Content uploaded by Daniela Jakubowicz
Author content
All content in this area was uploaded by Daniela Jakubowicz on Aug 28, 2017
Content may be subject to copyright.
Influences of Breakfast on Clock
Gene Expression and Postprandial
Glycemia in Healthy Individuals
and Individuals With Diabetes:
A Randomized Clinical Trial
https://doi.org/10.2337/dc16-2753
OBJECTIVE
The circadian clock regulates glucose metabolism by mediating the activity of met-
abolic enzymes, hormones, and transport systems. Breakfast skipping and night
eating have been associated with high HbA
1c
and postprandial hyperglycemia after
lunch and dinner. Our aim was to explore the acute effect of breakfast consumption
or omission on glucose homeostasis and clock gene expression in healthy individuals
and individuals with type 2 diabetes.
RESEARCH DESIGN AND METHODS
In a cross-over design, 18 healthy volunteers and 18 volunteers with 14.5 61.5 years
diabetes, BMI 30.7 61.1 kg/m
2
,andHbA
1c
7.6 60.1% (59.6 60.8 mmol/mol) were
randomly assigned to a test day with breakfast and lunch (YesB) and a test day with
only lunch (NoB). Postprandial clock and clock-controlled gene expression, plasma
glucose, insulin, intact glucagon-like peptide-1 (iGLP-1), and dipeptidyl peptidase IV
(DPP-IV) plasma activity were assessed after breakfast and lunch.
RESULTS
In healthy individuals, the expression level of Per1,Cry1,Rora,andSirt1 was lower
(P<0.05) but Clock was higher (P<0.05) after breakfast. In contrast, in individuals
with type 2 diabetes, Per1,Per2,andSirt1 only slightly, but significantly, decreased
and Roraincreased (P<0.05) after breakfast. In healthy individuals, the expression
level of Bmal1,Rora,andSirt1 was higher (P<0.05) after lunch on YesB day, whereas
the other clock genes remained unchanged. In individuals with type 2 diabetes,
Bmal1,Per1,Per2,Rev-erba,andAmpk increased (P<0.05) after lunch on the YesB
day. Omission of breakfast altered clock and metabolic gene expression in both
healthy and individuals with type 2 diabetes.
CONCLUSIONS
Breakfast consumption acutely affects clock and clock-controlled gene expression
leading to normal oscillation. Breakfast skipping adversely affects clock and clock-
controlled gene expression and is correlated with increased postprandial glycemic
response in both healthy individuals and individuals with diabetes.
1
Diabetes Unit, Wolfson Medical Center, Sackler
Faculty of Medicine, Tel Aviv University, Holon,
Israel
2
Diabetes Unit, Hadassah University Hospital,
Ein Kerem Hospital, Hadassah Medical School,
The Hebrew University of Jerusalem, Jerusalem,
Israel
3
Department of Clinical Sciences, Faculty of
Medicine, Lund University, Lund, Sweden
4
Institute of Biochemistry, Food Science and Nu-
trition, TheRobert H. Smith Faculty of Agriculture,
Food and Environment, The Hebrew University of
Jerusalem, Rehovot, Israel
5
Department of Molecular Genetics, Faculty of
Biochemistry, Weizmann Institute of Science,
Rehovot, Israel
Corresponding authors: Oren Froy, oren.froy@
mail.huji.ac.il, and Daniela Jakubowicz, daniela.
jak@gmail.com.
Received 25 December 2016 and accepted 30
July 2017.
Clinical trial reg. no. NCT01939782, clinicaltrials
.gov.
This article contains Supplementary Data online
at http://care.diabetesjournals.org/lookup/
suppl/doi:10.2337/dc16-2753/-/DC1.
D.J. and J.W. contributed equally to this work.
© 2017 by the American Diabetes Association.
Readers may use this article as long as the work
is properly cited, the use is educational and not
for profit, and the work is not altered. More infor-
mation is available at http://www.diabetesjournals
.org/content/license.
Daniela Jakubowicz,
1
Julio Wainstein,
1
Zohar Landau,
1
Itamar Raz,
2
Bo Ahren,
3
Nava Chapnik,
4
Tali Ganz,
1
Miriam Menaged,
1
Maayan Barnea,
5
Yosefa Bar-Dayan,
1
and Oren Froy
4
Diabetes Care 1
CARDIOVASCULAR AND METABOLIC RISK
Diabetes Care Publish Ahead of Print, published online August 22, 2017
The circadian clock controls the activity of
most enzymes, hormones, and transport
systems involved in glucose metabolism
(1,2). The central circadian clock, located
in the suprachiasmatic nuclei of the ante-
rior hypothalamus, generates endoge-
nous ;24-h rhythms. Similar clocks are
found in peripheral tissues, such as the
liver, b-cells, muscle, and adipose tissue
(3–5). The core clock mechanism in the
brain and peripheral tissues comprises
two loops. The positive loop consists of
the CLOCK and BMAL1 heterodimer that
mediates transcription of tissue-specific
genes and those of the negative feedback
loop. The negative feedback loop consists
of the period (PERs) and cryptochrome
(CRYs) proteins that inhibit CLOCK:BMAL1-
mediated transcription (6).
Unlike the suprachiasmatic nuclei clock,
which responds mainly to the light-dark
cycle, peripheral clocks respond to meal
content and timing, leading to coordi-
nated regulation of digestive and absorp-
tive functions, hormone secretion, thereby
preventing metabolic dysregulation (1–4).
BMAL1:CLOCK heterodimer plays a criti-
cal role in mediating the transcription of
coactivators that regulate the circadian
synthesis of most of the enzymesand hor-
mones involved in glucose homeostasis,
i.e., regulation of hepatic gluconeogene-
sis and pancreatic b-cell insulin secretion
(7). CRYs inhibit gluconeogenic gene ex-
pression by regulating CREBP activity and
their hepatic depletion increases circulat-
ing glucose, and CRY overexpression re-
duces fasting blood glucose and improves
whole-body insulin sensitivity in obese
mice (7). CLOCK:BMAL1 heterodimer reg-
ulates the expression of Rev-erbaand
Rora(8–10). REV-ERBa, the negative reg-
ulator of Bmal1 expression (8), is induced
during normal adipogenesis and mediates
a suppressive effect on hepatic gluconeo-
genesis and glucose output by regulating
the expression of PEPCK and glucose-
6-phosphatase (G6pase). Depletion of
REV-ERBaleads to hyperglycemia (2,7).
In contrast, RORa, the positive regulator
of Bmal1 expression (10), activates the
hepatic gluconeogenic enzyme G6pase
and regulates lipogenesis and lipid stor-
age in skeletal muscle (2,10).
AMPK, the cellular energy sensor, also
plays a key role in the clock mechanism by
enhancing degradation of PERs and CRYs
(11). Upregulation of AMPK signaling sig-
nificantly enhances GLUT4 translocation
and muscular glucose uptake, ensuring
metabolic efficiency and improving post-
prandial glucose and insulin responses
(12). AMPK exerts a positive effect on
SIRT1, associated with beneficial effects
on b-cell viability and insulin sensitivity,
which interacts directly with CLOCK and
deacetylates BMAL1 and PER2 (12–15).
Reduced glucose-stimulated insulin se-
cretion, insulin resistance, diminished
b-cell proliferation, and apoptosis have
been associated with asynchrony or defi-
ciencies in clock genes (16). Moreover,
lower transcripts of Bmal1 and Cry2 are
inversely correlated with HbA
1c
levels
(17,18). Furthermore, higher risk of obe-
sity, metabolic alterations, and type 2 di-
abetes have been found in shift workers
and individuals who underwent acute or
chronic forced circadian misalignment
(19,20), supporting the notion that the
clock plays an essential role in the preser-
vation of insulin sensitivity and b-cell
function.
It has recently been found that secre-
tion of glucagon-like peptide-1 (GLP-1), a
key incretin hormone that regulates glu-
cose-dependent insulin secretion from in-
testinal L cells, shows a rhythmic pattern
in rats and humans in vivo. In addition, its
secretion is altered by circadian disrup-
tors, such as constant light exposure, con-
sumption of a Western diet, and feeding
during the inactivity hours. The alter-
ations in the rhythm of GLP-1 secretory
response parallel changes in the pattern
of insulin responses (5). This would sug-
gest that it is important to preserve clock
functionality in subjects with type 2 dia-
betes. In fact, recent studies suggest that
breakfast skipping and night eating dis-
rupt circadian rhythms and, as a result,
impair glucose metabolism and b-cell
function (16,21,22).
We have recently shown that omission
of breakfast in patients with type 2 diabe-
tes led to increased postprandial glycemia
and attenuated insulin and GLP-1 re-
sponses after lunch and dinner (23). Al-
though the mechanism underlying this
effect is unclear, we hypothesize that
breakfast consumption triggers proper
cyclic clock gene expression, leading to
improved postprandial glycemia and in-
sulin and GLP-1 responses. Since the
acute effect of meals and specifically
breakfast on clock gene expression is
not known in humans, we compared
clock and clock-controlled gene expres-
sion in association with glucose, insu-
lin, and GLP-1 responses and plasma
dipeptidyl peptidase IV (DPP-IV) activity
in healthy individuals and individuals
with type 2 diabetes consuming or skip-
ping breakfast.
RESEARCH DESIGN AND METHODS
Participants
This was a randomized, open-label,
crossover-within-subject clinical trial. The
study population included 18 subjects
with type 2 diabetes and 18 healthy sub-
jects. All participants habitually ate break-
fast, did not work shifts within the last
5 years, and did not cross time zones
within the last month prior to the study.
Participants usually woke up at 0600–
0700 h and went to sleep at 2200–2400 h.
None of the participants had impaired
thyroid, renal, or liver function; anemia;
pulmonary disease; psychiatric, immuno-
logical, or neoplastic diseases; or severe
diabetic complications or underwent bari-
atric surgery. All participants were insulin
na¨
ıve, and only patients taking metformin
were included. Patients taking oral hypo-
glycemic agents, such as GLP-1 analogs,
anorectic drugs, or steroids, were ex-
cluded. Those patients on metformin
were told to stop taking the medication
2 days before test days. All participants
gave their informed consent. The study
was registered at clinicaltrials.gov,
NCT01939782. The Helsinki Committee of
the Wolfson Medical Center, where the
trial was performed, approved the study.
Study Design
There were two groups of participants: a
group with type 2 diabetes and a group of
healthy control subjects. Both groups un-
derwent two separate test days in the
clinic, with two different experimental
conditions: 1) the participants ate break-
fast and lunch (YesB), or 2) the partici-
pants did not eat breakfast, i.e., continued
the overnight fast until lunchtime and then
hadlunch(NoB).Thetwotestdayswere
separated by 1–2 weeks of washout. A
person not involved in the study, using a
coin flip, randomized participants within
their group to start with the YesB or NoB
day. Participants ingested their last oral
therapy 18 h before the test days. The sub-
jects were instructed to eat a standardized
meal in the evening, between 2100 and
2200 h, prior to test day, consisting of
twowhite/wheatbreadslicesandanop-
tional drink, such as tea or coffee with an
artificial sweetener or diet soda. Eating a
standardized meal the night before the test
2 Breakfast Skipping Alters Circadian Expression Diabetes Care
prevents long overnight fasting in the YesB
group and more so in the NoB group. Ad-
ditionally, participants were instructed to
avoid alcohol and excessive physical activ-
ity 6 days preceding each test day. On the
YesB day, participants had two identical
standard meals that were provided in the
clinic as breakfast at 0830 h and lunch at
1200 h. On the NoB day, breakfast was
omitted and the individuals continued their
overnight fast until lunch at 1200 h. The
energy and content of each test meal
were identical (572 68 kcal; 19.4% fat,
49% carbohydrates, and 32% protein).
The primary outcome was to compare
clock and clock-controlled gene expres-
sion (Clock,Bmal1, Per1, Per2, Cry1,Rev-
erba,Rora,Ampk,andSirt1)inwhite
blood cells (WBCs) on YesB versus NoB
days in healthy individuals and individuals
with type 2 diabetes. Gene expression
profiling in the WBCs has been shown to
reflect food-related metabolic changes,
even in the postprandial state (24).
Gene expression was relative to actin, a
housekeeping gene, normalizing varia-
tions in WBC count. The secondary out-
come was to compare postprandial
plasma glucose, insulin, intact GLP-1
(iGLP-1), and DPP-IV plasma activity.
Meal Tests
On the day of the meal tests, each partic-
ipant reported to the laboratory at 0730 h
after an overnight fast. Anthropometric
data were collected in the morning. At
0800 h, a catheter was placed in the vein
of the nondominant arm and remained
there until 1530 h. Venous blood samples
for quantification of plasma glucose, insu-
lin, and iGLP-1 were collected before break-
fast (at 0830 h), at 60 and 120 min after
breakfast, and at 210 min (at 1200 h, just
before lunch). After lunch (at 1200 h), sam-
ples were collected at 60, 120, and 210 min
(420 min including breakfast). On the NoB
day, blood samples were taken at the same
time points as on the YesB day. The sam-
ples for the assessment of clock and clock-
controlled gene expression and DPP-IV
activity were taken after the overnight fast
(at 0830 h), before lunch (at 1200 h), and
210 min after lunch (at 1530 h).
Biochemical and Hormonal Blood
Analyses
Plasma glucose was immediately analyzed
with hexokinase using a Cobas analyzer
(Roche Diagnostics, Madison, WI). Blood
samples for determining iGLP-1 were
collected into chilled tubes containing
EDTA and diprotin A (0.1 mmol/L). Samples
were centrifuged at 3,000 rpm at 4°C
for 10 min and stored at 280°C. Insulin
was determined by electrochemilumines-
cence immunoassay using a Cobas ana-
lyzer (Roche Diagnostics) according to
the manufacturer’s instructions. iGLP-1
was quantified using ELISA (IBL Amer-
ica, Minneapolis, MN). DPP-IV plasma
activity was analyzed as previously
described (25).
Analysis of Gene Expression in
Leukocytes
Blood for gene expression was collected
in Tempus tubes (Applied Biosystems,
Foster City, CA) and total RNA extracted
according to the manufacturer’s instruc-
tions. Total RNA was DNase I treated us-
ing RQ1 DNase (Promega, Madison, WI)
for 2 h at 37°C, as previously described
(26). Two micrograms of DNase I–treated
RNA was reverse transcribed using
MMuLV reverse transcriptase and ran-
dom hexamers (Promega). One-twentieth
of the reaction was then subjected to
quantitative real-time PCR using primers
spanning exon-exon boundaries and the
ABI Prism 7300 Sequence Detection Sys-
tem (Applied Biosystems). The fold
change in target gene expression was
calculated by the 2
2DDCt
relative quanti-
fication method (Applied Biosystems).
Sample Size and Power Analysis
Asamplesizeofn= 16 healthyparticipants
and n= 16 participants with type 2 di-
abetes in crossover was determined
to have .90% power to detect a true,
between-group difference of 40 625%
relative mRNA expression levels of clock
genes with a Pvalue = 0.05. The current
study provided 80% power to detect 5%
difference between-group in overall post-
prandial plasma area under the curve
Figure 1—Gene expression of healthy participants and participants with type 2 diabetes with or without
breakfast. Blood samples were collected after overnight fasting (time point 0), after which breakfast was
given (YesB) or fasting continued (NoB). Hashes denote statistical differences (P,0.05) between time
point 0 and time point 210 min, 3.5 h after breakfast or no breakfast in the healthy group. Asterisks
denote statistical differences (P,0.05) between time point 0 and time point 210 min, 3.5 h after
breakfast or no breakfast in the group with type 2 diabetes. Data are means 6SE.
care.diabetesjournals.org Jakubowicz and Associates 3
(AUC) for iGLP-1, insulin, and glucose. To
allow discontinuation, 36 participants
were recruited.
Statistical Analysis
Thirty-six subjects were enrolled in the
study and four subjects dropped out.
They were excluded from the analyses;
therefore, the results are based on n=32
subjects. AUC was calculated by the trap-
ezoidal rule. For time series, a two-way
ANOVA (time 3diet) was performed
and a least significant difference paired
Student ttest post hoc analysis was
used for comparison between the YesB
and NoB test days at each time point. For
clock gene expression, fasting (time point
0)wasdeterminedasbaselinefortheYesB
and NoB groups 3.5 h after (time point
210 min). Time point 210 min was deter-
mined as the baseline for the postlunch
(time point 420 min) gene expression. Stu-
dent ttest analyses were performed to
evaluate differences between time points
within each group and treatment. The re-
sults are expressed as mean 6SEM. A
Pvalue #0.05 was considered statistically
significant. Statistical analysis was per-
formed with JMP software (version 12;
SAS Institute Inc., Cary, NC).
RESULTS
Participants
Thirty-six individuals (18 with type 2 di-
abetes and 18 healthy) were enrolled in
the study, with 32 participants complet-
ing (16 with type 2 diabetes and 16
healthy). During the first meal test day,
four dropped out (two healthy and two
with diabetes) because of problems in the
installation of the intravenous catheter.
Dropouts did not differ significantly from
those who continued the study. The pa-
tients with type 2 diabetes were 66.8 6
1.9 years of age and had the disease
for 14.5 61.5 years with 59.6 60.8
mmol/mol (7.6 60.1%) HbA
1c
and BMI
30.7 61.1 kg/m
2
(Supplementary Table
1). Ten patients were treated with
diet alone, whereas six were treated
with diet and metformin. Seven pa-
tients had a history of hypertension and
were treated with ACE inhibitors and/or
calcium channel antagonists. The partic-
ipants in the healthy group were 44.3
62.9 years of age with 4.9 60.1%
(30.1 60.6 mmol/mol) HbA
1c
and BMI
23.1 60.4 kg/m
2
(Supplementary Table
1). They were not taking any medications.
Clock Gene Expression on YesB and
NoB Days in Healthy Individuals
Versus Individuals With Diabetes
Before Lunch
In healthy individuals, the expression
level of Per1,Cry1,Rora,andSirt1 was
lower (Student ttest, P,0.05), but Clock
was higher (Student ttest, P,0.05) after
breakfast (Fig. 1). In contrast, in individu-
als with type 2 diabetes, Per1,Per2,and
Sirt1 only slightly, but significantly, de-
creased and Roraincreased (Student
ttest, P,0.05) after breakfast. Omission
of breakfast altered clock and metabolic
gene expression in both healthy people
and individuals with type 2 diabetes. In
healthy individuals, there was an in-
crease in the expression level of Bmal1,
Cry1,Rev-erba,Roraand Sirt1 (Student
ttest, P,0.05), whereas in individuals
with type 2 diabetes, Clock expression
increased (Student ttest, P,0.05) (Fig.
1). In both healthy individuals and
individuals with diabetes, Per1 and Per2
expression was not affected whether
breakfast was consumed or not (Fig. 1).
Clock Gene Expression on YesB and
NoB Days in Healthy Individuals
Versus Individuals With Diabetes After
Lunch
In healthyindividuals, the expression level
of Bmal1,Rora,andSirt1 was higher (Stu-
dent ttest, P,0.05) after lunch on the
YesB day, whereas the other clock genes
remained unchanged (Fig. 2). In individu-
als with type 2 diabetes, Bmal1,Per1,
Per2,Rev-erba,andAmpk increased (Stu-
dent ttest, P,0.05) after lunch on the
YesB day. Omission of breakfast altered
clock and metabolic gene expression after
lunch in both healthy individuals and in-
dividuals with type 2 diabetes. In healthy
individuals, Per2 and Ampk expression in-
creased and Per1,Cry1,andRev-erbade-
creased (Student ttest, P,0.05) (Fig. 2),
Figure 2—Gene expression of healthy participants and participants with type 2 diabetes after lunch.
Blood samples were collected 3.5 h after breakfast or no breakfast (time point 210 min) and 3.5 h
after lunch (time point 420 min). Hashes denote statistical differences (P,0.05) between time
point 210 min and time point 420 min in the healthy group. Asterisks denote statistical differences
(P,0.05) between time point 210 min and time point 420 min in the group with type 2 diabetes.
Data are means 6SE.
4 Breakfast Skipping Alters Circadian Expression Diabetes Care
whereas in individuals with type 2 diabe-
tes, Bmal1,Per1, and Ampk expression
decreased and Roraexpression increased
(Student ttest, P,0.05) after lunch on
the NoB day (Fig. 2). Taken together,
breakfast and lunch led to increased ex-
pression of the positive loop and down-
regulation of the negative feedback loop,
whereas breakfast skipping altered this
expression pattern. Lunch, as the first
meal of the day, could not rectify the al-
tered clock expression. In addition, indi-
viduals with type 2 diabetes who skip
breakfast have a greater disruption of
their circadian rhythms compared with
those who consume breakfast.
Glucose, Insulin, and iGLP-1 Levels and
DPP-IV Activity on YesB and NoB Days
In the healthy group and group with di-
abetes, the AUC
glucose
after lunch was 15–
18% higher during the NoB day versus
the YesB day (P,0.001) (Fig. 3 and
Supplementary Table 2). In healthy par-
ticipants, AUC
insulin
after lunch was not
significantly different between the tests
(P= 0.5). In contrast, in the group with
diabetes, the AUC
insulin
after lunch was
25% lower on the NoB day compared
with the YesB day (P,0.004) (Fig. 3
and Supplementary Table 2). In both the
healthy group and group with diabetes,
the AUC
iGLP-1
after lunch was ;35% lower
on the NoB day compared with the
YesB day (P,0.0001) (Fig. 3 and
Supplementary Table 2). No significant
change in plasma DPP-IV activity was
found after lunch between the groups
on YesB or NoB day.
CONCLUSIONS
In this study, we report for the first time
an acute postprandial effect on clock and
clock-controlled gene expression in both
healthy individuals and individuals with
type 2 diabetes. We showed that break-
fast skipping acutely disrupts circadian
rhythms in both healthy individuals and
people with type 2 diabetes. Further-
more, clock gene expression after lunch
was different and depended on whether
breakfast was consumed or not. Thus, the
study results suggest that breakfast con-
sumption is of high relevance for preserv-
ing clock gene activity, and this is true for
both healthy subjects and subjects with
diabetes.
Several reports have suggested that
meal timing exerts a pivotal influence
on peripheral clocks and clock output
systems involved in the regulation of met-
abolic pathways (1,26–29). Breakfast
skipping (22,30) and/or eating at hours
designed to sleep (21), for the long run,
lead to disruption of clock gene expres-
sion and are associated with weight gain
and diabetes (31,32). In contrast, acclima-
tion to timed feeding, high-energy break-
fast, and fasting during the hours
designed for sleep reversed impaired
clock gene expression and enhanced
AMPK expression, resulting in a more
efficient weight loss, insulin sensitivity,
and reduction of lipid accumulation
(21,22,30–32).
In healthy people, breakfast and lunch
acutely led to an overall increased expres-
sion of the positive loop of the core clock
mechanism, whereas the expression of
the negative feedback loop was downre-
gulated (Fig. 4). This pattern maintains
the normal cyclic expression of output
Figure 3—Line charts of healthy participants (A) and participants with type 2 diabetes (B) on YesB
and NoB daysfor glucose, insulin, iGLP-1, and DPP-IV.Breakfast was given to the YesB group at time
point 0. Lunchwas given to both groups at timepoint 210 min. Asterisks denote statisticaldifference
between YesB and NoB at a specific time point. Data are means 6SE.
care.diabetesjournals.org Jakubowicz and Associates 5
systems (Fig. 4). In contrast, breakfast
skipping in healthy people led to an al-
tered clock gene expression pattern. After
lunchontheNoBday,Sirt1,Clock,Bmal1,
and Rorawere upregulated and Per1 was
downregulated, similarly to the effect on
the YesB day (Fig. 4). However, unlike on
the YesB day, after lunch, Per2 and Ampk
were upregulated and Cry1 did not
change, disrupting the normal cyclic ex-
pression (Fig. 4).
In patients with type 2 diabetes, break-
fast and lunch eventually led to an overall
altered clock gene expression (Fig. 4).
However, Ampk expression was upregu-
lated after lunch. As AMPK activation
leads to glucose uptake, this upregulation
may indicate improved glycemic control,
as indeed was found herein. In contrast,
breakfast skipping in people with diabetes
further imparted a greater disru ption, as it
also led to reduced Ampk levels, accentu-
ating blood glucose dysregulation (Fig. 4).
Our results show that the effect of
breakfast is acute, as only 1 day of skipping
breakfast led to such a deleterious effect
on clock and clock-controlled gene expres-
sion. The rapid change in clock gene ex-
pression on YesB versus NoB days is
consistent with reports in animal models,
in which a 30-min feeding stimulus al-
tered clock gene expression within 2–4h
(33,34). As feeding poses such an acute
effect, it is clear why long-term abnormal
patterns of clock gene expression are
manifested in obese individuals and indi-
viduals with diabetes.
A change in clock and clock-controlled
gene expression, due to breakfast omis-
sion, is expected to yield altered metabo-
lism. Impaired circadian rhythms have
been closely associated with the patho-
physiology of type 2 diabetes(4). Reduced
glucose-stimulated insulin secretion, insu-
lin resistance, diminished b-cell prolifera-
tion, and apoptosis have been associated
with asynchrony or deficiencies in clock
genes (16). Indeed, in people with diabe-
tes, omission of breakfast led to a signif-
icant increase in postprandial glucose
excursions with reduced insulin and
iGLP-1 responses after lunch compared
with YesB day. Furthermore, this delete-
rious effect of breakfast skipping was also
observed in healthy participants, although
with a lower magnitude.
The reduction of hyperglycemia and
enhanced insulin response after lunch
with prior consumption of breakfast has
been described as the second meal
phenomenon and has been previously re-
ported in several studies in healthy in-
dividuals and individuals with type 2
diabetes (23,35). This has been explained
by enhanced b-cell responsiveness at the
second meal induced by the first meal
(36,37). This explanation is based on the
findings that both the first and the second
phase of insulin release are influenced by
b-cell memory, and the magnitude of in-
sulin release is enhanced significantly by
previous glucose exposure (37). The ab-
sence of glucose elevation due to fasting
until noon may diminish b-cell respon-
siveness and memory, leading to a re-
duced and delayed insulin response
after lunch on the NoB day. Indeed, it
was recently reported that lower insulin
release by b-cells in response to nutrient
depletion or starvation induces lysosomal
degradation of nascent secretory insulin
granules, and this is controlled by protein
kinase D (PKD), a key player in secretory
granule biogenesis (38). The impaired in-
sulin secretion at lunch on the NoB day
may also be due to perturbed incretin
regulation, since both b-cell memory
and sensitivity to glucose are enhanced
by GLP-1. Therefore, the higher levels of
GLP-1 on the YesB day may explain both
the enhanced insulin secretion and the
reduced glycemic response after lunch.
At the cellular level, the reduction of post-
prandial glycemia after lunch on the YesB
day can be explained by the fact that
breakfast triggers correct cyclic expres-
sion of clockgenes, which, in turn, assures
metabolic efficiency, leading to normal
glucose, insulin, and GLP-1 responses af-
ter lunch. Ampk upregulation in partici-
pants with type 2 diabetes may enhance
GLUT4 translocation and muscular glu-
cose uptake, leading to improved post-
prandial glycemia.
At baseline, clock gene levels were dif-
ferent between the healthy group and
the group with type 2 diabetes, presum-
ably due to differences related to age, body
weight, and/or HbA
1c
levels. Although the
age and body weight of the group with
type 2 diabetes was significantly different
from the healthy group, and may be a
limitation in our study, changes in clock
gene expression as a result of breakfast
skipping cannot be attributed to these
parameters, as each group was normal-
ized to its own baseline. In addition, the
crossover study assured that the same
person had both YesB and NoB treat-
ments, reiterating the effect of breakfast
skipping. Another limitation of our study
is that it was performed until after lunch.
Figure 4—Effect of lunch with or without prior breakfast on the interrelations between the clock
mechanism and metabolic proteins. CLOCK and BMAL1 mediate the expression of clock and clock-
controlled genes regulating circadian hormone secretion (insulin and glucagon), metabolism (glu-
coneogenesis and glycolysis), and glucosehomeostasis. PER1, PER2, and CRY1 serve as the negative
feedback loop that inhibits CLOCK:BMAL1-mediated expression. SIRT1 and AMPK, when activated
under low cellular energy levels, relieve the inhibition mediated by the negative feedback loop.
BMAL1 expression is positively regulated by RORaand negatively regulated by REV-ERBa. Breakfast
consumption followed by lunch assures this overall cyclic regulation is maintained. Omission
of breakfast disruptsthis cyclic regulation in both healthy people and patientswith type 2 diabetes.
(A high-quality color representation of this figure is available in the online issue.)
6 Breakfast Skipping Alters Circadian Expression Diabetes Care
Analyses of subsequent dinner and other
time points during the nighttime are re-
quired in order to determine overall daily
rhythms and how each meal affects this
oscillatory pattern. In summary, our study
demonstrates that breakfast consump-
tion exerts a rapid influence on clock
and clock-controlled genes involved in
glucose regulation, such as Ampk,and
this effect is associated with a significant
reduction in postprandial glycemia and
enhanced insulin and GLP-1 responses
after subsequent lunch. In contrast, break-
fast skipping adversely affects the expres-
sion of clock and clock-controlled genes
involved in glucose metabolism in both
healthy individuals and individuals with di-
abetes. This study emphasizes the con-
sumption of breakfast as an important
strategy when targeting glycemic control
in type 2 diabetes. As the circadian clock
also regulates blood pressure, heart rate,
and cardiovascular activity, as well as adi-
pose tissue and other metabolic organs
(39), meal timing may affect overall me-
tabolism and influence chronic complica-
tions of obesity and type 2 diabetes.
Funding. This study was funded by D-Cure and
the Ministry of Health, Israel (grant 3-0000-
11406).
Duality of Interest. No potential conflicts of in-
terest relevant to this article were reported.
Author Contributions. D.J. and O.F. contrib-
uted to the conception and design of the study,
acquired, analyzed, and interpreted data, and
drafted and revised the manuscript. J.W. con-
tributed to the conception and design of the
study, acquired and interpreted data, and
drafted the manuscript. Z.L., N.C., T.G., M.M.,
and Y.B.-D. contributed to the conception and
designof the study,acquired andinterpreteddata,
organized the randomization, and drafted the
manuscript. I.R. and B.A. researched data, con-
tributed to the interpretation of the data, and
draftedand revisedthe manuscript.M.B. analyzed
and interpreted data. D.J. is the guarantor of this
work and, as such, had full access to all the data in
the study and takes responsibility forthe integrity
of the data and the accuracy of the data analysis.
Prior Presentation. This study was presented
orally at the 77th Scientific Sessions of the Amer-
ican Diabetes Association, San Diego, CA, 9–13
June 2017.
References
1. Froy O. Metabolism and circadian rhythms–
implications for obesity. Endocr Rev 2010;31:1–24
2. Jordan SD, Lamia KA. AMPK at the crossroads
of circadian clocks and metabolism. Mol Cell En-
docrinol 2013;366:163–169
3. Zvonic S, Ptitsyn AA, Conrad SA, et al. Charac-
terization of peripheral circadian clocks in adipose
tissues. Diabetes 2006;55:962–970
4. Ando H, Ushijima K, Yanagihara H, et al. Clock
gene expression in the liver and adipose tissues of
non-obese type 2 diabetic Goto-Kakizaki rats. Clin
Exp Hypertens 2009;31:201–207
5. Brubaker PL, Gil-Lozano M. Glucagon-like
peptide-1: The missing link inthe metabolic clock?
J Diabetes Investig 2016;7(Suppl. 1):70–75
6. Reppert SM, Weaver DR. Coordination of cir-
cadian timing in mammals. Nature2002;418:935–
941
7. Feng D,Lazar MA. Clocks, metabolism, and the
epigenome. Mol Cell 2012;47:158–167
8. Preitner N, Damiola F, Lopez-Molina L, et al.
The orphan nuclear receptor REV-ERBalpha con-
trols circadian transcription within the positive
limb of the mammalian circadian oscillator. Cell
2002;110:251–260
9. UedaHR,ChenW,AdachiA,etal.Atranscrip-
tion factor response element for gene expression
during circadian night. Nature 2002;418:534–539
10. Sato TK, Panda S, Miraglia LJ, et al. A func-
tional genomics strategy reveals Rora as a com-
ponent of the mammalian circ adian clock. Neuron
2004;43:527–537
11. Lamia KA, Sachdeva UM, DiTacchio L, et al.
AMPK regulates the circadian clock by crypto-
chrome phosphorylation and degradation. Sci-
ence 2009;326:437–440
12. Hardie DG, Ross FA, Hawley SA. AMPK: a nu-
trient and energy sensor that maintains energy
homeostasis. Nat Rev Mol Cell Biol 2012;13:
251–262
13. Pinho AV, Bensellam M, Wauters E, et al.
Pancreas-specific Sirt1-deficiencyin mice compro-
mises beta-cell function without development of
hyperglycemia. PLoS One 2015;10:e0128012
14. Asher G, Gatfield D, Stratmann M, et al. SIRT1
regulates circadianclock gene expressionthrough
PER2 deacetylation. Cell 2008;134:317–328
15. Nakahata Y, Sahar S, Astarita G, Kaluzova M,
Sassone-Corsi P. Circadian control of the NAD+
salvage pathway by CLOCK-SIRT1. Science 2009;
324:654–657
16. Vieira E, Burris TP, Quesada I. Clock genes,
pancreatic function, and diabetes. Trends Mol
Med 2014;20:685–693
17. Pappa KI, Gazouli M, Anastasiou E, Iliodromiti
Z, Antsaklis A, Anagnou NP. Circadian clock gene
expression is impaired in gestational diabetes
mellitus. Gynecol Endocrinol 2013;29:331–335
18. Ando H, Takamura T, Matsuzawa-Nagata N,
et al. Clock gene expression in peripheral leuco-
cytes of patients with type 2 diabetes. Diabetolo-
gia 2009;52:329–335
19. Vetter C, Devore EE, Ramin CA, Speizer FE,
WillettWC, Schernhammer ES.Mismatch of sleep
and work timing and risk of type 2 diabetes. Di-
abetes Care 2015;38:1707–1713
20. Scheer FA, Hilton MF, Ma ntzoros CS, Shea SA.
Adverse metabolic and cardiovascular conse-
quences of circadian misalignment. Proc Natl
Acad Sci U S A 2009;106:4453–4458
21. Arble DM, Bass J, Laposky AD, Vitaterna MH,
Turek FW. Circadian timing of food intakecontrib-
utes to weight gain. Obesity (Silver Spring) 2009;
17:2100–2102
22. Yoshida C, Shikata N, Seki S, Koyama N,
Noguchi Y. Early nocturnal meal skipping alters
the peripheral clock and increases lipogenesis in
mice. Nutr Metab (Lond) 2012;9:78
23. Jakubowicz D,Wainstein J, AhrenB, Landau Z,
Bar-Dayan Y, Froy O. Fasting until noon triggers
increased postprandial hyperglycemia and im-
paired insulin response after lunch and dinner in
individuals with type 2 diabetes: a randomized
clinical trial. Diabetes Care 2015;38:1820–1826
24. Kawakami Y, Yamanaka-Okumura H, Sakuma
M, et al. Gene expression profiling in peripheral
white blood cells in response tothe intake of food
with different glycemic index using a DNA micro-
array. J Nutrigenet Nutrigenomics 2013;6:154–
168
25. Gunnarsson PT, Winzell MS, Deacon CF, etal.
Glucose-inducedincretin hormonerelease and in-
activation are differently modulated by oral fat
and protein in mice. Endocrinology 2006;147:
3173–3180
26. Sherman H, Frumin I, Gutman R, et al. Long-
term restricted feeding alterscircadian expression
and reducesthe level of inflammatory and disease
markers. J Cell Mol Med 2011;15:2745–2759
27. Hatori M, VollmersC, Zarrinpar A, et al. Time-
restricted feeding without reducing caloric intake
preventsmetabolic diseasesin mice fed a high-fat
diet. Cell Metab 2012;15:848–860
28. Chaix A, Zarrinpar A, Miu P, Panda S. Time-
restricted feeding is a preventative and therapeutic
intervention against diverse nutritional challenges.
Cell Metab 2014;20:991–1005
29. Yasumoto Y, Hashimoto C, Nakao R, et al.
Short-term feeding at the wrong time is sufficient
to desynchronize peripheral clocks and induce obesity
with hyperphagia, physical inactivity and metabolic
disorders in mice. Metabolism 2016;65:714–727
30. Wu T, Sun L, ZhuGe F, et al. Differential roles
of breakfast and supper in rats of a daily three-
meal schedule upon circadian regulation and
physiology. Chronobiol Int 2011;28:890–903
31. Kudo T, Akiyama M, Kuriyama K, Sudo M,
Moriya T, Shibata S. Night-time restricted feeding
normalises clock genes and Pai-1 gene expression
in the db/db mouse liver. Diabetologia 2004;47:
1425–1436
32. FuseY, Hirao A, Kuroda H, Otsuka M, Tahara Y,
Shibata S. Differentialroles of breakfast only (one
meal per day) and a bigger breakfast with a small
dinner (two meals per day) in mice fed a high-fat
diet with regard to induced obesity and lipid me-
tabolism. J Circadian Rhythms 2012;10:4
33. Wu T, Ni Y, Kato H, Fu Z. Feeding-induced
rapid resetting of the hepatic circadian clock is as-
sociated with acute induction of Per2 and Dec1
transcriptioninrats.ChronobiolInt2010;27:1–18
34. Wu T, Fu O, Yao L, Sun L, Zhuge F, Fu Z. Dif-
ferential responses of peripheral circadian clocks
to a short-term feeding stimulus. Mol Biol Rep
2012;39:9783–9789
35. Jovanovic A, Gerrard J, Taylor R. The second-
meal phenomenon in type 2 diabetes. Diabetes
Care 2009;32:1199–1201
36. Nesher R, Cerasi E. Modeling phasic insulin
release: immediate and time-dependent effects
of glucose. Diabetes 2002;51(Suppl. 1):S53–S59
37. Vagn Korsgaard T, Colding-Jørgensen M.
Time-dependent mechanisms in beta-cell glucose
sensing. J Biol Phys 2006;32:289–306
38. Goginashvili A, Zhang Z, Erbs E, et al. Insulin
granules. Insulin secretory granules control auto-
phagy in pancreatic bcells. Science 2015;347:
878–882
39. Richards J, Diaz AN, Gumz ML. Clock genes in
hypertension: novel insights from rodent models.
Blood Press Monit 2014;19:249–254
care.diabetesjournals.org Jakubowicz and Associates 7