Extracted Text
16647.full.pdf
PERSPECTIVE
Meal frequency and timing in health
and disease
Mark P. Mattson
a,b,1
, David B. Allison
c
, Luigi Fontana
d,e,f
, Michelle Harvie
g
, Valter D. Longo
h
, Willy J. Malaisse
i
,
Michael Mosley
j
, Lucia Notterpek
k
, Eric Ravussin
l
, Frank A. J. L. Scheer
m
, Thomas N. Seyfried
n
, Krista A. Varady
o
,
and Satchidananda Panda
p,1
a
Laboratory of Neurosciences, National Institute on Aging, Baltimore, MD 21224;
b
Department of Neuroscience, The Johns Hopkins University
School of Medicine, Baltimore, MD 21205;
c
Nutrition and Obesity Research Center, University of Alabama at Birmingham, Birmingham, AL
35294;
d
Department of Medicine, Washington University in St. Louis, St. Louis, MO 63130;
e
Department of Clinical and Experimental Sciences,
Brescia University, 25123 Brescia, Italy;
f
CEINGE Biotecnologie Avanzate, 80145 Naples, Italy;
g
Genesis Breast Cancer Prevention Centre,
University Hospital South Manchester, Wythenshaw, M23 9LT Manchester, United Kingdom;
h
Longevity Institute, Davis School of Gerontology
and Department of Biological Sciences, University of Southern California, Los Angeles, CA 90089;
i
Laboratory of Experimental Hormonology,
Brussels Free University, B-1070 Brussels, Belgium;
j
British Broadcasting Corporation, W1A 1AA London, United Kingdom;
k
Department of
Neuroscience, College of Medicine, McKnight Brain Institute, University of Florida, Gainesville, FL 32610;
l
Pennington Biomedical Research
Center, Baton Rouge, LA 70808;
m
Harvard Medical School and Brigham and Women’s Hospital, Boston, MA 02115;
n
Biology Department,
Boston College, Chestnut Hill, MA 02467;
o
Department of Kinesiology and Nutrition, University of Illinois at Chicago, Chicago, IL 60612;
and
p
Regulatory Biology Laboratory, Salk Institute for Biological Studies, La Jolla, CA 92037
Edited by Joseph S. Takahashi, Howard Hughes Medical Institute, University of Texas Southwestern Medical Center, Dallas, TX, and approved October 7, 2014 (received for review
July 23, 2014)
Although major research efforts have focused on how specific components of foodstuffs affect health, relatively little is known about a more
fundamental aspect of diet, the frequency and circadian timing of meals, and potential benefits of intermittent periods with no or very low
energy intakes. The most common eating pattern in modern societies, three meals plus snacks every day, is abnormal from an evolutionary
perspective. Emerging findings from studies of animal models and human subjects suggest that intermittent energy restriction periods of as
little as 16 h can improve health indicators and counteract disease processes. The mechanisms involve a metabolic shift to fat metabolism and
ketone production, and stimulation of adaptive cellular stress responses that prevent and repair molecular damage. As data on the optimal
frequency and timing of meals crystalizes, it will be critical to develop strategies to incorporate those eating patterns into health care policy
and practice, and the lifestyles of the population.
metabolism|circadian rhythm|time-restricted feeding|feeding behavior|obesity
Obesity and associated diseases of modern
societies (diabetes, cardiovascular/cerebrovas-
cular disease, cancers, and Alzheimer’sdis-
ease) are overwhelming health care systems.
Unfortunately, the common knowledge that
reducing overall calorie intake and regular
exercise can help optimize body weight and
reduce disease risk has, in many cases, not
been implemented successfully. Some of the
advice provided by physicians and dieticians
to their patients is consistent with the current
scientific evidence, including the benefits of
vegetables, fruits, fiber, nuts, and fish, and the
value of reducing or eliminating snacks.
However, there are many myths and pre-
sumptions concerning diet and health,
including that it is important to eat three
or more meals per day on a regular basis
(1, 2). Although many aspects of diet and
lifestyle influence metabolic status and dis-
ease trajectory during the life course, emerg-
ing findings suggest that the influences of
the frequency and timing of meals on health
may be large, but are difficult to characterize
with any generality. Here we describe the
physiological responses of laboratory ani-
mals and human subjects to controlled var-
iations in meal size, frequency, and circadian
timing, and their impact on health and dis-
ease in modern societies. Three experimental
dietary regimens are considered: (i)caloric
restriction (CR), in which daily calorie intake
is reduced by 20–40%, and meal frequency is
unchanged; (ii) intermittent energy restric-
tion (IER), which involves eliminating (fast-
ing) or greatly reducing (e.g., 500 calories
per day) daily intake food/caloric beverage
intake intermittently, for example 2 d/wk;
and (iii) time-restricted feeding (TRF), which
involves limiting daily intake of food and
caloric beverages to a 4- to 6-h time window.
We also consider the cultural and industrial
barriers to implementing evidence-based
healthy eating patterns, and strategies for
removing or circumventing those barriers.
Evolutionary and Cultural
Considerations
Unlike modern humans and domesticated
animals, the eating patterns of many mam-
mals are characterized by intermittent energy
intake. Carnivores may kill and eat prey only
a few times each week or even less frequently
(3, 4), and hunter-gatherer anthropoids, in-
cluding those living today, often eat inter-
mittently depending upon food availability
(5, 6). The ability to function at a high level,
both physically and mentally, during ex-
tended periods without food may have been
of fundamental importance in our evolu-
tionary history. Many adaptations for an in-
termittent food supply are conserved among
mammals, including organs for the uptake
and storage of rapidly mobilizable glucose
(liver glycogen stores) and longer-lasting
energy substrates, such as fatty acids in
adipose tissue. Behavioral adaptations that
Author contributions: M.P.M., D.B.A., L.F., M.H., V.D.L., W.J.M.,
M.M., L.N., E.R., F.A.J.L.S., T.N.S., K.A.V., and S.P. wrote the
paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
1
To whom correspondence may be addressed. Email: mark.mattson@
nih.gov or panda@salk.edu.
www.pnas.org/cgi/doi/10.1073/pnas.1413965111 PNAS |November 25, 2014 |vol. 111|no. 47|16647–16653
PERSPECTIVE
enable food acquisition and storage per-
meate the behavioral repertoire of all spe-
cies, including humans. Indeed, the higher
cognitive capabilities of humans compared
with other species likely evolved for the
purpose of acquiring food resources; evi-
dence suggests that the earliest tools (7)
and languages (8) were invented to aid in
food acquisition.
The agricultural revolution, which began
∼10,000 y ago, resulted in the constant year-
round availability of food typical of modern
societies. Our agrarian ancestors adopted
a three meals per day eating pattern, pre-
sumably because it provided both social and
practical benefits for the daily work and
school schedules. More recently, within the
past 50 y, high calorie density foodstuffs
(refined grains, sugar, cooking oils, corn
syrup, and so on) have permeated these three
daily meals (9). When superimposed on in-
creasingly sedentary lifestyles, the consump-
tion of high-energy meals multiple times each
day plausibly contributed to the emergence of
obesity and related diseases as major causes
of morbidity and mortality (Fig. 1). Obesity
has also become a major health problem
in dogs and cats, which are often fed ad
libitum (10), and even laboratory rodents
can often be considered overfed and sed-
entary (11, 12). Indeed, animals in the wild
and hunter-gatherer humans rarely, if ever,
suffer from obesity, diabetes, and cardio-
vascular disease (5).
Circadian Rhythms, Meal Timing, and
Health
Circadian rhythms are self-sustained∼24-h
oscillations in behavior, physiology, and
metabolism. These rhythms have evolved
and permit organisms to effectively respond
to the predictable daily change in the light:
dark cycle and the resultant rhythms in
food availability in nature. Gene-expression
studies have revealed more than 10% of
expressed genes in any given organ exhibit
circadian oscillation (13). These rhythmic
transcripts encode key rate-determining
steps in neuroendocrine, signaling, and met-
abolic pathways. Such regulation temporally
separatesincompatiblecellularprocessesand
optimizes cellular and organismal fitness.
Although the circadian clock is cell-autono-
mous and is present in the majority of tissue
types, the circadian system is organized in a
hierarchical manner in which the hypotha-
lamic suprachiasmatic nucleus (SCN) func-
tions as the master circadian clock that uses
both diffusible and synaptic mechanisms to
orchestrate circadian rhythms in the periph-
eral organs at appropriate phase. Photore-
ceptive retinal ganglion cells send ambient
light information to the SCN through
monosynaptic connection to ensure that the
circadian system is entrained to the daily
light:dark cycle (14).
Whereas light is the dominant timing cue
for the SCN oscillator, time of food intake
affects the phase of theclocks in peripheral
tissues (15), including liver, muscle, and ad-
ipose tissues. For millions of years in the
absence of artificial light, the circadian
clock—in conjunction with the retinal light
input—imposed diurnal rhythms in physi-
ology and behaviors, including the activity/
rest and feeding/fasting cycle. For many of
our ancestors, food was probably scarce and
primarily consumed during daylight hours,
leaving long hours of overnight fasting.
With the advent of affordable artificial light-
ing and industrialization, modern humans
began to experience prolonged hours of illu-
mination every day and resultant extended
consumption of food. The modern lifestyle
perturbed the human circadian system in
three primary ways: shift work, exposure to
prolonged hours of artificial light, and er-
ratic eating patterns. Although it is difficult
to separate the consequence of each of these
perturbations on metabolism and physiol-
ogy, animal models and recent experimental
human studies have begun to elucidate the
mechanisms and consequence of these cir-
cadian disruptions. In industrial societies
nearly 10% of the workforce performs
night-shift work: either permanent night
work, rotating shifts, or irregular sched-
ules, in which the individuals typically
switch their wakeful hours back to the
daytime during days off to maintain a typ-
ical social life on those days. During night-
shift work the individuals are subject to
both prolonged hours of artificial lighting
and an abnormal eating schedule. Fur-
thermore, during the weekend the ten-
dency to maintain a day-active social life
imposes a jet-lag–type paradigm in which
both central and peripheral clocks attempt
to adjust to a weekend lifestyle. Although
such internal desynchrony has never been
demonstrated directly in humans, based
on animal experimental work this is pre-
sumed to result in chronic disruption of
circadian rhythms, which may help ex-
plaintheknownassociationbetweennight
work and several diseases, including car-
diovascular disease, diabetes, obesity, cer-
tain types of cancer, and neurodegenerative
diseases (16, 17).
In addition to shift work, modern human
societies experience prolonged illumination
(18) and erratic eating patterns, both of
which are known to perturb the circadian
system. In nocturnal rodent models, extended
illumination has been shown to increase
predisposition to metabolic diseases. Con-
versely, in diurnal flies a shift to nighttime
feeding compromises fat metabolism and
fecundity (19). In humans, a 12-h shift of
the sleep/wake and fasting/feeding cycle
compared with the central circadian sys-
tem, while maintaining an isocaloric diet,
reduced glucose tolerance, increased blood
pressure, and decreased the satiety hormone
leptin (20). These studies highlight the
importance of temporal organization of
sleep and feeding relative to the circadian
system. Both nutrient quality and genetic
factors appear to affect meal timing in
rodents. Mutation in the circadian clock
genePer1affecting a conserved phosphory-
lation site causes mice to consume more
food during the daytime and predisposes
them to metabolic diseases (21). The widely
used diet-induced obesity model in mice also
perturbs feeding; mice fed a high-fat diet ad
libitum consume small meals throughout day
and night (22). Both diet-induced obesity and
obesity inPer1mutant mice can be prevented
by restricting access to high-fat diet only
during the nighttime (23). The surprising
effectiveness of TRF without altering calo-
ric intake or source of calories suggests a
potentially effective meal-timing intervention
for humans. Indeed, recent human studies
suggest that earlier meal timing associates
with improved effectiveness of weight-loss
therapy in overweight and obese patients
(24, 25).
The mechanism underlying the beneficial
effect of TRF is likely complex and acts
on multiple pathways. The daily fasting and
feeding episodes trigger alternative activa-
tion of fasting-responsive cAMP response
element binding protein (CREB) and AMP
kinase, and feeding responsive insulin-
dependent mammalian target of rapamycin
(mTOR) pathways implicated in metabolic
homeostasis. In addition, these pathways
also impinge on the circadian clock and
Fig. 1.The rising tide of obesity is strongly associated
with daily calorie intake and sedentary lifestyle-promoting
transportation (refs. 84–86;www.earth-policy.org/
data_center/C23). *US, approximate value.
#
World-
wide auto production.
16648|www.pnas.org/cgi/doi/10.1073/pnas.1413965111 Mattson et al.
improve robustness of oscillation of clock
components and downstream targets (23).
Accordingly, gene-expression studies indicate
that TRF supports circadian rhythmicity of
thousands of hepatic transcripts (26).
The confluence of genomics and genetics
in mice is unraveling the pathways from the
core clock components to specific nutrient
metabolism. The nuclear hormone receptors
REV-ERBs are integral to the circadian clock
and directly regulate transcription of several
key rate-determining enzymes for fatty acid
and cholesterol metabolism (27). Although
cryptochrome proteins are strong transcrip-
tional suppressors, they also inhibit cAMP
signaling and thereby tune CREB-mediated
gluconeogenesis (28). Circadian clock down-
stream transcription factors DBP/TEF/HLF
regulate xenobiotic metabolism (29), and
KLF15 mediates nitrogen metabolism (30).
These and other modes of regulation (31)
provide a mechanistic framework by which
meal-timing affects the circadian clock and,
in turn, affects metabolic homeostasis in
mammals.
Not only does circadian phase influence
the metabolic response to food intake, food
intake itself has recently been demonstrated
to be under control by the endogenous
circadian system, independent of the sleep/
wake and fasting/feeding cycle (32), possi-
bly helping explain why breakfast is often
the smallest meal of the day or even skipped
all together.
Cellular and Molecular Mechanisms:
Insight from Intermittent Energy
Restriction and Fasting
Compared with those fed ad libitum, the
lifespans of organisms from yeast and worms,
to mice and monkeys can be extended by
dietary energy restriction (33–35). Data
collected from individuals practicing se-
vere dietary restriction indicate that humans
undergo many of the same molecular, met-
abolic, and physiologic adaptations typical
of long-lived CR rodents (36). IER/fasting
can forestall and even reverse disease pro-
cesses in animal models of various cancers,
cardiovascular disease, diabetes, and neu-
rodegenerative disorders (2). Here we briefly
highlight four general mechanisms by
which IER protects cells against injury
and disease.
Adaptive Stress Responses.Compared with
their usual ad libitum feeding conditions,
laboratory animals maintained on IER exhibit
numerous changes, suggesting heightened
adaptive stress responses at the molecular,
cellular, and organ system levels. Alternate-day
fasting prevents age-related decrements in the
antioxidant enzymes superoxide dismutase 1
and catalase in the liver cells of rats (37). IER
increases levels of the antioxidant enzymes
NADH-cytochromeb
5reductase and NAD
(P)H-quinone oxidoreductase 1 in muscle
cells of mice, and these effects are accentu-
ated by exercise (38). Numerous studies have
shown that IER can protect neurons against
oxidative, metabolic, and proteotoxic stress
in animal models of neurodegenerative
disorders, including Alzheimer’s and Par-
kinson’s diseases (39). IER can also pro-
tect the heart against ischemic damage in
an animal model of myocardial infarction
(40). Alternate-day fasting stimulates the
production of several different neuropro-
tective proteins, including the antioxidant
enzyme heme oxygenase 1, proteins involved
in mitochondrial function, and the protein
chaperones HSP-70 and GRP-78 (41, 42).
Moreover, IER increases the production of
trophic factors that promote neuronal sur-
vival, neurogenesis, and the formation and
strengthening of synapses in the brain (43).
Taken together, these data suggestthat ben-
eficial effects of IER involve the general
biological phenomenon of“hormesis”or
“preconditioning,”in which exposure of
cells and organisms to a mild stress results in
adaptive responses that protect against more
severe stress.
Bioenergetics.When humans switch from
eating three full meals per day to an IER diet,
such as one moderate size meal every other
dayoronly500–600 calories 2 d/wk, they
exhibit robust changes in energy metabolism
characterized by increased insulin sensitivity,
reduced levels of insulin and leptin, mobili-
zation of fatty acids, and elevation of ketone
levels (44–47). Ketones, such asβ-hydroxy-
butyrate, are known to have beneficial effects
on cells with a high energy demand, such as
neurons in the brain (Fig. 2) (48, 49). In mice,
alternate-day fasting can greatly increase
insulin sensitivity even without a major re-
duction in body weight (50), and in humans
IER can increase insulin sensitivity more
than daily calorie restriction that achieves
similar weight loss (45, 51). Dietary energy
restriction can prevent age-related decline
in mitochondrial oxidative capacity in skele-
tal muscle, and can induce mitochondrial
biogenesis (52). Brain bioenergetics may
also be bolstered by IER. For example,
brain-derived neurotrophic factor (BDNF),
which is up-regulated in hippocampal
neurons in response to IER and exercise,
activates the transcription factor CREB,
which then induces peroxisome proliferator-
activated receptor gamma coactivator 1-α
(PGC-1α) expression and mitochondrial
biogenesis (53). The latter study showed
Fig. 2.A metabolic shift to ketogenesis that occurs with fasting bolsters neuronal bioenergetics. Liver glycogen
stores are typically depleted within 10–12 h of fasting, which is followed by liberation of fatty acids from adipose
tissue cells into the blood. The fatty acids are then transported into liver cells where they are oxidized to generate
Acetyl-CoA. Acetyl-CoA is then converted to 3-hydroxy-3-methylgluaryl-CoA, which is in turn used to generate the
ketones acetoacetate andβ-hydroxybutyrate (β-OHB). The ketones are released into the blood and are transported
into various tissues, including the brain, where they are taken up by neurons and used to produce acetyl-CoA. Acetyl-
CoA enters the tricarboxylic acid (TCA) cycle to generate ATP.
Mattson et al. PNAS |November 25, 2014 |vol. 111|no. 47|16649
PERSPECTIVE
that PGC-1αand mitochondrial biogenesis
are critical for the formation of synapses in
developing hippocampal neurons and the
maintenance of synapses in the hippo-
campus of adult mice. Because impaired
mitochondrial biogenesis and function oc-
cur during aging and chronic disease states,
such as sarcopenia and neurodegenerative
disorders, it is important to consider the
impact of the frequency and circadian timing
of meals on the development and progression
of such disorders.
Whereas IER/fasting is beneficial and
overeating detrimental for many types of
normal cells, the converse is true for tumor
cells. Cells in tumors exhibit major mito-
chondrial abnormalities and generate their
ATP primarily from glycolysis rather than
oxidative phosphorylation (54). Moreover,
tumors are highly vascularized and so their
cells have access to large amounts of circu-
lating glucose. Animal models have consis-
tently shown that IER inhibits and even
reverses the growth of a range of tumors,
including neuroblastoma, breast, and ovarian
cancers (55). The shift to ketogenesis may
play an important role in suppression of
tumor growth by IER/fasting because many
tumor cells are largely unable to use ketones
as an energy source; accordingly, ketogenic
diets may potentiate the antitumor effects of
IER (54). Although preliminary, recent case
studies in human patients suggest potential
applications of IER in the treatment of a
range of cancers, including breast, ovar-
ian, prostate, and glioblastoma (56, 57).
Indeed, evolutionary theory predicts that
collected random mutations will prevent
tumor cells from making the necessary met-
abolic adaptations to IER (58).
Inflammation.All major diseases, including
cardiovascular disease, diabetes, neurodegen-
erative disorders, arthritis, and cancers in-
volve chronic inflammation in the affected
tissues and, in many cases, systemically (59).
Local tissue inflammation involves hyper-
activation of macrophages (microglia in the
brain) which produce proinflammatory
cytokines (TNF, IL-1β, IL-6) and reactive
oxygen species. Overweight and obesity
promote inflammation, and IER suppresses
inflammation in human subjects and ani-
mal models of diseases. Obese women who
changed their diet from multiple daily meals
to alternate-day energy restriction exhibited
significant reductions in levels of circulating
TNF and IL-6 (60). In asthma patients, 2 mo
of alternate-day energy restriction reduced
circulating TNF and markers of oxidative
stress, and improved asthma symptoms and
airway resistance (44). However, because
weight loss may reduce inflammation re-
gardless of the dietary change inducing the
weight loss, it will be important to deter-
mine if and how eating patterns modify
inflammation independently of weight loss.
Multiple studies have shown that fasting
can lessen symptoms in patients with rheu-
matoid arthritis (61), and data from animal
studies suggest that the pathogenesis of other
autoimmune disorders may also be counter-
acted by IER, including multiple sclerosis
(62), lupus erythematosus (63), and type I
diabetes (64). In a mouse model of stroke,
IER suppressed elevations of TNF and IL-1β
in the ischemic cerebral cortex and striatum,
which was associated with improved func-
tional outcome (41). Inflammation is in-
creasingly recognized as a contributing factor
for cancer cell growth (65) and, because ex-
cessive energy intake promotes inflammation,
it is likely that suppression of inflammation
plays a role in the inhibition of tumor growth
by IER. Whereas inhibiting immune re-
sponses to autoantigens and sterile tissue
injuries can be beneficial, suppression of
immune responses to infectious agents is
detrimental. It will therefore be impor-
tant to determine whether eating regimens
such as TRF and IER affect immune re-
sponses to pathogens, an as yet unexplored
area of investigation.
Improved Repair and Removal of Dam-
aged Molecules and Organelles.Cells
possess dedicated mechanisms for the re-
moval of damaged molecules and organelles.
One mechanism involves the molecular
“tagging”of damaged proteins with ubiq-
uitin, which targets them for degradation in
the proteasome (66). In a second and more
elaborate mechanism called autophagy,
damaged and dysfunctional proteins, mem-
branes, and organelles are directed to and
degraded in lysosomes (67). Energy and
nutrient (particularly amino acids) intake
have been shown to have consistent effects
on autophagy. When organisms ingest regu-
lar meals, their cells receive a relatively
steady supply of nutrients and so remain in
a“growth mode”in which protein synthesis
is robust and autophagy is suppressed (68).
The nutrient-responsive mTOR pathway
negatively regulates autophagy. Accordingly,
fasting inhibits the mTOR pathway and
stimulates autophagy in cells of many tissues,
including liver, kidney, and skeletal muscle
(69–71). In this way, fasting“cleanses”
cells of damaged molecules and organelles.
Rats maintained on energy-restricted
diets exhibit reduced accumulation of
polyubiquitinated proteins and evidence
of increased autophagy in peripheral
nerves compared with rats fed ad libi-
tum (72). In a mouse model of Charcot-
Marie-Tooth type 1A, an inherited disorder
characterized by demyelination of peripheral
nerves, IER improved motor performance
and reduced demyelination by a mecha-
nism involving enhanced autophagy and
reduced accumulation of myelin protein
PMP22 aggregates (73). A common feature of
many major chronic diseases is the abnormal/
excessive accumulation of protein aggregates
within and outside of cells; examples include
intracellularα-synuclein in Parkinson’sdis-
ease and extracellular amyloidβ-peptide and
intracellular Tau protein in Alzheimer’s
disease (74, 75). In addition to the frequency
of meals, the circadian timing of meals is
likely to affect the responses of the cellular
machineries for clearance of damaged
proteins and organelles (76). Autophagy is
regulated in a diurnal rhythm in many cell
types, and this rhythm can be altered by
changing the timing of food intake. It is
therefore reasonable to consider that meal
timing has an impact on diseases that involve
impaired or insufficient autophagy.
Society-Wide Implications
The high rates of childhood and adult obesity
and the diseases they foster is a major burden
to our society. As findings from basic re-
search studies and controlled interventional
trials accrue, consensus recommendations for
healthy patterns of meal frequency and
diurnal timing may eventually emerge. If
sufficient evidence does emerge to support
public health and clinical recommendations
to alter meal patterning, there will be nu-
merous forces at play in the acceptance or
resistance to such recommendations. First
and perhaps foremost is cultural tradition.
Three meals plus snacks daily has become
the norm during the past half-century, such
that a majority of American children are
accustomed to this eating pattern. Second,
the agriculture, food processing, food retail,
and restaurant industries and all of the affil-
iated industries that serve or promote food—
from airlines to concert stadiums to television
cooking shows to advertising and others—
still all have established practices and finan-
cial interests and these interests may affect
receptivity to proposed shifts in eating pat-
terns and potential decreases in total food
purchased. Third, the willingness and ability
of the American health care system, includ-
ing medical training and practice, to em-
phasize prevention and lifestyles will be a
key factor in success or lack thereof.
We believe that it is important to consider
how“prescriptions”for meal frequency and
timing can be developed, validated, and
16650|www.pnas.org/cgi/doi/10.1073/pnas.1413965111 Mattson et al.
implemented in light of the current in-
dustrial, cultural, and institutional pressures
to maintain the status quo of daily over-
consumption of food. In doing so, it will be
importanttoensurethatweprovidethe
public with accurate information on eating
patterns and health. For example, despite
equivocal and even contradictory scientific
evidence, breakfast is often touted as
a weight-control aid (77), but recent evi-
dence has suggested that it may not be (78).
Primary education and media outlets
should dispense up-to-date information on
healthy eating, including the frequency and
circadian timing of meals. Although regu-
latory agencies must play an important role
in developing recommendations and facili-
tating their implementation, it may also be
helpful for parents to lead by example and
establish healthy eating patterns in their
children. Additionally, the inclusion of
science-based information on eating pat-
terns and health in primary and secondary
education may help stem the rising tide of
overeating and related poor health in our
children. The medical community could
play a central role in developing and
implementing prescriptions for long-term
daily energy restriction or IER that can be
incorporated into most daily home and
workplace environments. Examples of such
prescriptions include fasting or caloric re-
striction (e.g., 500 calories) on alternate days
or 2 d each week, or forgoing breakfast and
lunch several days each week (Fig. 3). The
available evidence suggests that patients may
be able to comply with such diets when
there is rigorous follow-up (44, 45, 47), and
it will be important to determine if com-
pliance would increase further if patients
were able to choose an eating pattern-based
prescription that best fits their weekly rou-
tines. Recent findings suggest that it may be
possible for many people to adopt a long-
term change in their lifestyle from eating
three meals plus snacks every day to an IER
diet if they are able to keep on the new
eating pattern during a transition period of
approximately 1 mo (45). Moreover, for
many people who are overweight IER may
facilitate their maintenance of an overall
reduction in energy intake compared with
prescriptions for daily caloric restriction.
Future Directions
Further animal studies are required to better
elucidate the cellular and molecular mecha-
nisms by which meal frequency, IER, and
TRF affect health and disease susceptibility,
as well as the impact of eating patterns on
extant disease processes in various experi-
mental models. For example, it will be of
great interest to know the effects of IER and
TRF on gene expression, epigenetic mark-
ers (methylation and acetylation), and dis-
ease-relevant pathways in multiple tissues
throughout the body and nervous system.
The overlapping and complementary effects
of exercise and healthy eating patterns
on functionality and disease resistance
should be elucidated. Intervention studies
of IER and TRF, particularly randomized
controlled trials (RCTs), should be per-
formed in various groups of human sub-
jects, including those who are healthy and
those with diseases, such as obesity, di-
abetes, cancer, cardiovascular disease, and
neurodegenerative disorders. RCTs should
include functional outcomes as well as bio-
markers relevant to disease risk and patho-
genesis. Thus, far, relatively few RCTs of
IER and TRF have been performed in hu-
man subjects, with the results of several
studies of alternate-day and twice weekly
energy restriction demonstrating weight
loss and abdominal fat reduction and
suggesting improvements in indicators
of energy and lipid metabolism and in-
flammation (44–46, 51, 61). On the other
hand, a study of TRF in which healthy
normal weight subjects consumed a bal-
anced daily food intake within a 4-h or
12-h time period each day revealed no
improvement (79, 80), which is similar to
the lack of any short-term benefit of TRF
in mice when the animals are fed a bal-
anced diet (23). This finding suggests that
the short-term benefits of TRF might depend
on the diet and body composition. It will
also be critical to evaluate long-term ad-
herence of different subject populations
to IER and TRF protocols to evaluate
their feasibility for broad applications for
sustained weight reduction and disease
risk reduction.
Genetic factors can determine whether the
lifespan of a particular strain of mouse or
rat is increased, unaffected, or even de-
creased, by lifelong CR or TRF, with inbred
animals generally responding less well to CR
(81). Understanding the mechanism of TRF
will help to predict whether a certain eating
pattern is beneficial or whether individuals
with specific genotype are predisposed to
erratic eating patterns. Missense mutation in
circadian clock component Per1 has been
shown to affect eating patterns in mice
(21). However, the presence of intact food
anticipatory activity in SCN ablated rodents
or those lacking functional circadian oscil-
lator genes points to yet-unidentified genes
and circuits in eating-pattern determination
(82, 83). Humans are highly heterogeneous
with regard to their genetic composition,
epigenetic landscape, and the environmental
factors to which they are exposed through-
out life. It is therefore likely that there will be
considerable variability among human sub-
jects in the responses of their cells and organ
systems (and overall health) to different
eating patterns. Although there is sufficient
evidence to suggest that CR and IER can
improve health indicators in most or all
obese human subjects (2), data are lacking
with regard to normal weight subjects.
Fig. 3.Patterns of daily and weekly food consumption. The upper illustration shows five different patterns of
food consumption during a 24-h period. A: Eating three large meals plus snacks spread throughout a 16-h period
of wakefulness; this is the common eating pattern of food consumption upon which the epidemic of obesity,
diabetes, and associated chronic diseases has emerged. B–D: Examples if time-restricted eating patterns in which
food is consumed as three (B) or two (C) regular size meals, or three small meals (D). E: Complete fast. Examples of
weekly eating schedules are shown in thelower right. ER, energy restriction; IER, intermittent energy restriction;
TRF, time-restricted feeding.
Mattson et al. PNAS |November 25, 2014 |vol. 111|no. 47|16651
PERSPECTIVE
Insight into genetic and epigenetic factors
that affect responses to specific eating
patterns could be obtained from RCTs of
TRF and IER regimens in normal weight
subjects in which biomarkers of health and
disease risk are measured (blood pressure,
heart rate variability, insulin resistance,
lipid profiles, adipokines, ketones, and
so forth).
It would be particularly valuable to de-
sign RCT in human subjects with head-to-
head comparisons of multiple eating pat-
terns, such as those shown in Fig. 3. Once
the eating patterns that promote optimal
health are established, what can be done to
encourage, enable and empower individ-
uals to modify their food choices and
eating patterns? Implementing any such
changes will be challenging, as a half-cen-
tury of research on behavioral approaches to
weight control suggests. That said, the field
of behavioral science is continually evolv-
ing, as is the growth and quality of mobile
information technology, which may serve
to buttress efforts. We are hopeful that in
thefuture,wemaybebetterabletohelp
individuals achieve the healthy behavior
changes they desire.
ACKNOWLEDGMENTS. This article incorporates in-
formation from a workshop on“Eating Patterns and
Disease,”which can be viewed atvideocast.nih.gov/
summary.asp?Live=13746&bhcp=1, and was supported
by the National Institute on Aging Intramural Re-
search Program and the Glenn Foundation for Medi-
cal Research. Relevant research in the authors’
laboratories are supported by NIH intramural support
(to M.P.M.); NIH Grants P30DK056336 (to D.B.A.),
P01AG034906 (to V.D.L.), R01NS041012 (to L.N.),
P30DK072476 (to E.R.), R01DK099512 (to F.A.J.L.S.),
R01NS055195 (to T.N.S.), R01HL106228 (to K.A.V.),
and R01DK091618 (to S.P.); the European Union’s
Seventh Framework Programme MOPACT [mobilising
the potential of active ageing in Europe; FP7-SSH-
2012-1 Grant 320333 (to L.F.)]; a grant from Genesis
Breast Cancer Prevention, UK (to M.H.); and Belgian
Foundation for Scientific Medical Research Grant
3.4520.07 (to W.J.M.).
1Casazza K, et al. (2013) Myths, presumptions, and facts about
obesity.N Engl J Med368(5):446–454.
2Longo VD, Mattson MP (2014) Fasting: Molecular mechanisms
and clinical applications.Cell Metab19(2):181–192.
3Gervasi V, et al. (2012) Predicting the potential demographic
impact of predators on their prey: A comparative analysis of two
carnivore-ungulate systems in Scandinavia.J Anim Ecol81(2):
443–454.
4Metz MC, Smith DW, Vucetich JA, Stahler DR, Peterson RO (2012)
Seasonal patterns of predation for gray wolves in the multi-prey
system of Yellowstone National Park.J Anim Ecol81(3):553–563.
5Cordain L, Eaton SB, Miller JB, Mann N, Hill K (2002) The
paradoxical nature of hunter-gatherer diets: Meat-based, yet non-
atherogenic.Eur J Clin Nutr56(Suppl 1):S42–S52.
6Ströhle A, Hahn A, Sebastian A (2010) Estimation of the diet-
dependent net acid load in 229 worldwide historically studied hunter-
gatherer societies.Am J Clin Nutr91(2):406–412.
7van Schaik CP, Deaner RO, Merrill MY (1999) The conditions for
tool use in primates: Implications for the evolution of material
culture.J Hum Evol36(6):719–741.
8Allen JS (2012)“Theory of food”as a neurocognitive adaptation.
Am J Hum Biol24(2):123–129.
9Tillotson JE (2004) America’s obesity: Conflicting public policies,
industrial economic development, and unintended human
consequences.Annu Rev Nutr24:617–643.
10Zoran DL (2010) Obesity in dogs and cats: A metabolic and
endocrine disorder.Vet Clin North Am Small Anim Pract40(2):
221–239.
11Martin B, Ji S, Maudsley S, Mattson MP (2010)“Control”
laboratory rodents are metabolically morbid: Why it matters.Proc
Natl Acad Sci USA107(14):6127–6133.
12Klimentidis YC, et al. (2011) Canaries in the coal mine: A cross-
species analysis of the plurality of obesity epidemics.Proc Biol Sci
278(1712):1626–1632.
13Panda S, Hogenesch JB, Kay SA (2002) Circadian rhythms from
flies to human.Nature417(6886):329–335.
14Hatori M, Panda S (2010) The emerging roles of melanopsin in
behavioral adaptation to light.Trends Mol Med16(10):435–446.
15Stokkan KA, Yamazaki S, Tei H, Sakaki Y, Menaker M (2001)
Entrainment of the circadian clock in the liver by feeding.Science
291(5503):490–493.
16Stevens RG, Brainard GC, Blask DE, Lockley SW, Motta ME (2014)
Breast cancer and circadian disruption from electric lighting in the
modern world.CA Cancer J Clin64(3):207–218.
17Morris CJ, Yang JN, Scheer FA (2012) The impact of the circadian
timing system on cardiovascular and metabolic function.Prog Brain
Res199:337–358.
18Wright KP Jr, et al. (2013) Entrainment of the human circadian
clock to the natural light-dark cycle.Curr Biol23(16):1554–1558.
19Gill S, Panda S (2011) Feeding mistiming decreases reproductive
fitness in flies.Cell Metab13(6):613–614.
20Scheer FA, Hilton MF, Mantzoros CS, Shea SA (2009) Adverse
metabolic and cardiovascular consequences of circadian
misalignment.Proc Natl Acad Sci USA106(11):4453–4458.
21Liu Z, et al. (2014) PER1 phosphorylation specifies feeding
rhythm in mice.Cell Rep7(5):1509–1520.
22Kohsaka A, et al. (2007) High-fat diet disrupts behavioral and
molecular circadian rhythms in mice.Cell Metab6(5):414–421.
23Hatori M, et al. (2012) Time-restricted feeding without reducing
caloric intake prevents metabolic diseases in mice fed a high-fat diet.
Cell Metab15(6):848–860.
24Garaulet M, et al. (2013) Timing of food intake predicts weight
loss effectiveness.Int J Obes (Lond)37(4):604–611.
25Jakubowicz D, Barnea M, Wainstein J, Froy O (2013) High caloric
intake at breakfast vs. dinner differentially influences weight loss of
overweight and obese women.Obesity (Silver Spring)21(12):
2504–2512.
26Vollmers C, et al. (2012) Circadian oscillations of protein-coding
and regulatory RNAs in a highly dynamic mammalian liver
epigenome.Cell Metab16(6):833–845.
27Cho H, et al. (2010) Regulation of circadian behaviour and
metabolism by REV-ERB-αand REV-ERB-β.Nature485(7396):
123–127.
28Zhang EE, et al. (2010) Cryptochrome mediates circadian
regulation of cAMP signaling and hepatic gluconeogenesis.Nat Med
16(10):1152–1156.
29Gachon F, Olela FF, Schaad O, Descombes P, Schibler U (2006)
The circadian PAR-domain basic leucine zipper transcription factors
DBP, TEF, and HLF modulate basal and inducible xenobiotic
detoxification.Cell Metab4(1):25–36.
30Jeyaraj D, et al. (2006) Klf15 orchestrates circadian nitrogen
homeostasis.Cell Metab15(3):311–323.
31Bass J, Takahashi JS (2010) Circadian integration of metabolism
and energetics.Science330(6009):1349–1354.
32Scheer FA, Morris CJ, Shea SA (2013) The internal circadian clock
increases hunger and appetite in the evening independent of food
intake and other behaviors.Obesity (Silver Spring)21(3):421–423.
33Goodrick CL, Ingram DK, Reynolds MA, Freeman JR, Cider NL
(1983) Effects of intermittent feeding upon growth, activity, and
lifespan in rats allowed voluntary exercise.Exp Aging Res9(3):
203–209.
34Honjoh S, Yamamoto T, Uno M, Nishida E (2009) Signalling
through RHEB-1 mediates intermittent fasting-induced longevity in
C. elegans.Nature457(7230):726–730.
35Colman RJ, et al. (2014) Caloric restriction reduces age-related
and all-cause mortality in rhesus monkeys.Nat Commun5:3557.
36Rizza W, Veronese N, Fontana L (2014) What are the roles of
calorie restriction and diet quality in promoting healthy longevity?
Ageing Res Rev13:38–45.
37Pieri C, et al. (1992) Food restriction in female Wistar rats: V. Lipid
peroxidation and antioxidant enzymes in the liver.Arch Gerontol
Geriatr14(1):93–99.
38Rodríguez-Bies E, Navas P, López-Lluch G (2014) Age-dependent
effect of every-other-day feeding and aerobic exercise in ubiquinone
levels and related antioxidant activities in mice muscle.J Gerontol A
Biol Sci Med Sci, 10.1093/gerona/glu002.
39Mattson MP (2012) Energy intake and exercise as determinants
of brain health and vulnerability to injury and disease.Cell Metab
16(6):706–722.
40Ahmet I, Wan R, Mattson MP, Lakatta EG, Talan M (2005)
Cardioprotection by intermittent fasting in rats.Circulation112(20):
3115–3121.
41Arumugam TV, et al. (2010) Age and energy intake interact to
modify cell stress pathways and stroke outcome.Ann Neurol67(1):
41–52.
42Poon HF, et al. (2006) Proteomics analysis provides insight into
caloric restriction mediated oxidation and expression of brain proteins
associated with age-related impaired cellular processes:
Mitochondrial dysfunction, glutamate dysregulation and impaired
protein synthes-is.Neurobiol Aging27(7):1020–1034.
43Marosi K, Mattson MP (2014) BDNF mediates adaptive brain and
body responses to energetic challenges.Trends Endocrinol Metab
25(2):89–98.
44Johnson JB, et al. (2007) Alternate day calorie restriction
improves clinical findings and reduces markers of oxidative stress and
inflammation in overweight adults with moderate asthma.Free Radic
Biol Med42(5):665–674.
45Harvie MN, et al. (2011) The effects of intermittent or continuous
energy restriction on weight loss and metabolic disease risk markers:
A randomized trial in young overweight women.Int J Obes (Lond)
35(5):714–727.
46Klempel MC, Kroeger CM, Varady KA (2013) Alternate day
fasting (ADF) with a high-fat diet produces similar weight loss and
cardio-protection as ADF with a low-fat diet.Metabolism62(1):
137–143.
47Varady KA, et al. (2013) Alternate day fasting for weight loss in
normal weight and overweight subjects: A randomized controlled
trial.Nutr J12(1):146.
48Kashiwaya Y, et al. (2013) A ketone ester diet exhibits anxiolytic
and cognition-sparing properties, and lessens amyloid and tau
pathologies in a mouse model of Alzheimer’s disease.Neurobiol
Aging34(6):1530–1539.
49Newman JC, Verdin E (2014) Ketone bodies as signaling
metabolites.Trends Endocrinol Metab25(1):42–52.
50Anson RM, et al. (2003) Intermittent fasting dissociates beneficial
effects of dietary restriction on glucose metabolism and neuronal
resistance to injury from calorie intake.Proc Natl Acad Sci USA
100(10):6216–6220.
51Harvie M, et al. (2013) The effect of intermittent energy and
carbohydrate restriction v. daily energy restriction on weight loss and
metabolic disease risk markers in overweight women.Br J Nutr
110(8):1534–1547.
52Martin-Montalvo A, de Cabo R (2013) Mitochondrial metabolic
reprogramming induced by calorie restriction.Antioxid Redox Signal
19(3):310–320.
53Cheng A, et al. (2012) Involvement of PGC-1αin the formation
and maintenance of neuronal dendritic spines.Nat Commun3:1250.
54Seyfried TN, Flores RE, Poff AM, D’Agostino DP (2014) Cancer as
a metabolic disease: Implications for novel therapeutics.
Carcinogenesis35(3):515–527.
55Lee C, et al. (2012) Fasting cycles retard growth of tumors and
sensitize a range of cancer cell types to chemotherapy.Sci Transl Med
4(124):124ra27.
56Safdie FM, et al. (2009) Fasting and cancer treatment in humans:
A case series report.Aging (Albany, NY Online)1(12):988–1007.
57Zuccoli G, et al. (2010) Metabolic management of glioblastoma
multiforme using standard therapy together with a restricted
ketogenic diet: Case report.Nutr Metab (Lond)7:33.
58Seyfried TN (2012)Cancer as a Metabolic Disease: On the
Origin, Management and Prevention of Cancer(John Wiley & Sons,
Hoboken, NJ), pp 261–275.
59McDade TW (2012) Early environments and the ecology of
inflammation.Proc Natl Acad Sci USA109(Suppl 2):17281–17288.
60Kroeger CM, et al. (2012) Improvement in coronary heart disease
risk factors during an intermittent fasting/calorie restriction regimen:
Relationship to adipokine modulations.Nutr Metab (Lond)9(1):98.
61Müller H, de Toledo FW, Resch KL (2001) Fasting followed by
vegetarian diet in patients with rheumatoid arthritis: A systematic
review.Scand J Rheumatol30(1):1–10.
62Esquifino AI, Cano P, Jimenez-Ortega V, Fernández-Mateos MP,
Cardinali DP (2007) Immune response after experimental allergic
encephalomyelitis in rats subjected to calorie restriction.
J Neuroinflammation4:6.
63Muthukumar A, Zaman K, Lawrence R, Barnes JL, Fernandes G
(2003) Food restriction and fish oil suppress atherogenic risk factors in
lupus-prone (NZB x NZW) F1 mice.J Clin Immunol23(1):23–33.
16652|www.pnas.org/cgi/doi/10.1073/pnas.1413965111 Mattson et al.
64Belkacemi L, et al. (2012) Intermittent fasting modulation of the
diabetic syndrome in streptozotocin-injected rats.Int J Endocrinol
2012:962012.
65Elinav E, et al. (2013) Inflammation-induced cancer: Crosstalk
between tumours, immune cells and microorganisms.Nat Rev
Cancer13(11):759–771.
66Maupin-Furlow J (2012) Proteasomes and protein conjugation
across domains of life.Nat Rev Microbiol10(2):100–111.
67Rubinsztein DC, Mariño G, Kroemer G (2011) Autophagy and
aging.Cell146(5):682–695.
68Speakman JR, Mitchell SE (2011) Caloric restriction.Mol Aspects
Med32(3):159–221.
69Donati A, Recchia G, Cavallini G, Bergamini E (2008) Effect of
aging and anti-aging caloric restriction on the endocrine regulation of
rat liver autophagy.J Gerontol A Biol Sci Med Sci63(6):550–555.
70Kume S, et al. (2010) Calorie restriction enhances cell adaptation
to hypoxia through Sirt1-dependent mitochondrial autophagy in
mouse aged kidney.J Clin Invest120(4):1043–1055.
71Wohlgemuth SE, Seo AY, Marzetti E, Lees HA, Leeuwenburgh C
(2010) Skeletal muscle autophagy and apoptosis during aging:
Effects of calorie restriction and life-long exercise.Exp Gerontol45(2):
138–148.
72Lee S, Notterpek L (2013) Dietary restriction supports peripheral
nerve health by enhancing endogenous protein quality control
mechanisms.Exp Gerontol48(10):1085–1090.
73Madorsky I, et al. (2009) Intermittent fasting alleviates the
neuropathic phenotype in a mouse model of Charcot-Marie-Tooth
disease.Neurobiol Dis34(1):146–154.
74Moreno-Gonzalez I, Soto C (2011) Misfolded protein aggregates:
Mechanisms, structures and potential for disease transmission.Semin
Cell Dev Biol22(5):482–487.
75Jucker M, Walker LC (2011) Pathogenic protein seeding in
Alzheimer disease and other neurodegenerative disorders.Ann
Neurol70(4):532–540.
76Ma D, Li S, Molusky MM, Lin JD (2012) Circadian autophagy
rhythm: A link between clock and metabolism?Trends Endocrinol
Metab23(7):319–325.
77Brown AW, Bohan Brown MM, Allison DB (2013) Belief beyond
the evidence: Using the proposed effect of breakfast on obesity to
show 2 practices that distort scientific evidence.Am J Clin Nutr98(5):
1298–1308.
78Dhurandhar EJ, et al. (2014) The effectiveness of breakfast
recommendations on weight loss: A randomized controlled trial.Am
J Clin Nutr100(2):507–513.
79Stote KS, et al. (2007) A controlled trial of reduced meal
frequency without caloric restriction in healthy, normal-weight,
middle-aged adults.Am J Clin Nutr85(4):981–988.
80Carlson O, et al. (2007) Impact of reduced meal frequency without
caloric restriction on glucose regulation in healthy, normal-weight
middle-aged men and women.Metabolism56(12):1729–1734.
81Swindell WR (2012) Dietary restriction in rats and mice: A meta-
analysis and review of the evidence for genotype-dependent effects
on lifespan.Ageing Res Rev11(2):254–270.
82Pendergast JS, Oda GA, Niswender KD, Yamazaki S (2012) Period
determination in the food-entrainable and methamphetamine-
sensitive circadian oscillator(s).Proc Natl Acad Sci USA109(35):
14218–14223.
83Mulder CK, Papantoniou C, Gerkema MP, Van Der Zee EA
(2014) Neither the SCN nor the adrenals are required for
circadian time-place learning in mice.Chronobiol Int31(9):
1075–1092.
84Fryar CD, Carroll MD, Ogden CL (2012)Prevalence of
Overweight, Obesity, and Extreme Obesity Among Adults: United
States, Trends 1960–1962 Through 2009–2010(National Center for
Health Statistics, Hyattsville, MD). Available at www.cdc.gov/nchs/
data/hestat/obesity_adult_09_10/obesity_adult_09_10.htm.
Accessed November 3, 2014.
85United States Department of Agriculture (2003) Profiling food
consumption in America.Agriculture Fact Book(US Department of
Agriculture, Washington, DC), pp 13–21.
86Day JC, Janus A, Davis J (2005)Computer and Internet Use in the
United States: 2003(US Census Bureau, Washington, DC). Available
at www.census.gov/prod/2005pubs/p23-208.pdf. Accessed
November 3, 2014.
Mattson et al. PNAS |November 25, 2014 |vol. 111|no. 47|16653
PERSPECTIVE
Meal frequency and timing in health
and disease
Mark P. Mattson
a,b,1
, David B. Allison
c
, Luigi Fontana
d,e,f
, Michelle Harvie
g
, Valter D. Longo
h
, Willy J. Malaisse
i
,
Michael Mosley
j
, Lucia Notterpek
k
, Eric Ravussin
l
, Frank A. J. L. Scheer
m
, Thomas N. Seyfried
n
, Krista A. Varady
o
,
and Satchidananda Panda
p,1
a
Laboratory of Neurosciences, National Institute on Aging, Baltimore, MD 21224;
b
Department of Neuroscience, The Johns Hopkins University
School of Medicine, Baltimore, MD 21205;
c
Nutrition and Obesity Research Center, University of Alabama at Birmingham, Birmingham, AL
35294;
d
Department of Medicine, Washington University in St. Louis, St. Louis, MO 63130;
e
Department of Clinical and Experimental Sciences,
Brescia University, 25123 Brescia, Italy;
f
CEINGE Biotecnologie Avanzate, 80145 Naples, Italy;
g
Genesis Breast Cancer Prevention Centre,
University Hospital South Manchester, Wythenshaw, M23 9LT Manchester, United Kingdom;
h
Longevity Institute, Davis School of Gerontology
and Department of Biological Sciences, University of Southern California, Los Angeles, CA 90089;
i
Laboratory of Experimental Hormonology,
Brussels Free University, B-1070 Brussels, Belgium;
j
British Broadcasting Corporation, W1A 1AA London, United Kingdom;
k
Department of
Neuroscience, College of Medicine, McKnight Brain Institute, University of Florida, Gainesville, FL 32610;
l
Pennington Biomedical Research
Center, Baton Rouge, LA 70808;
m
Harvard Medical School and Brigham and Women’s Hospital, Boston, MA 02115;
n
Biology Department,
Boston College, Chestnut Hill, MA 02467;
o
Department of Kinesiology and Nutrition, University of Illinois at Chicago, Chicago, IL 60612;
and
p
Regulatory Biology Laboratory, Salk Institute for Biological Studies, La Jolla, CA 92037
Edited by Joseph S. Takahashi, Howard Hughes Medical Institute, University of Texas Southwestern Medical Center, Dallas, TX, and approved October 7, 2014 (received for review
July 23, 2014)
Although major research efforts have focused on how specific components of foodstuffs affect health, relatively little is known about a more
fundamental aspect of diet, the frequency and circadian timing of meals, and potential benefits of intermittent periods with no or very low
energy intakes. The most common eating pattern in modern societies, three meals plus snacks every day, is abnormal from an evolutionary
perspective. Emerging findings from studies of animal models and human subjects suggest that intermittent energy restriction periods of as
little as 16 h can improve health indicators and counteract disease processes. The mechanisms involve a metabolic shift to fat metabolism and
ketone production, and stimulation of adaptive cellular stress responses that prevent and repair molecular damage. As data on the optimal
frequency and timing of meals crystalizes, it will be critical to develop strategies to incorporate those eating patterns into health care policy
and practice, and the lifestyles of the population.
metabolism|circadian rhythm|time-restricted feeding|feeding behavior|obesity
Obesity and associated diseases of modern
societies (diabetes, cardiovascular/cerebrovas-
cular disease, cancers, and Alzheimer’sdis-
ease) are overwhelming health care systems.
Unfortunately, the common knowledge that
reducing overall calorie intake and regular
exercise can help optimize body weight and
reduce disease risk has, in many cases, not
been implemented successfully. Some of the
advice provided by physicians and dieticians
to their patients is consistent with the current
scientific evidence, including the benefits of
vegetables, fruits, fiber, nuts, and fish, and the
value of reducing or eliminating snacks.
However, there are many myths and pre-
sumptions concerning diet and health,
including that it is important to eat three
or more meals per day on a regular basis
(1, 2). Although many aspects of diet and
lifestyle influence metabolic status and dis-
ease trajectory during the life course, emerg-
ing findings suggest that the influences of
the frequency and timing of meals on health
may be large, but are difficult to characterize
with any generality. Here we describe the
physiological responses of laboratory ani-
mals and human subjects to controlled var-
iations in meal size, frequency, and circadian
timing, and their impact on health and dis-
ease in modern societies. Three experimental
dietary regimens are considered: (i)caloric
restriction (CR), in which daily calorie intake
is reduced by 20–40%, and meal frequency is
unchanged; (ii) intermittent energy restric-
tion (IER), which involves eliminating (fast-
ing) or greatly reducing (e.g., 500 calories
per day) daily intake food/caloric beverage
intake intermittently, for example 2 d/wk;
and (iii) time-restricted feeding (TRF), which
involves limiting daily intake of food and
caloric beverages to a 4- to 6-h time window.
We also consider the cultural and industrial
barriers to implementing evidence-based
healthy eating patterns, and strategies for
removing or circumventing those barriers.
Evolutionary and Cultural
Considerations
Unlike modern humans and domesticated
animals, the eating patterns of many mam-
mals are characterized by intermittent energy
intake. Carnivores may kill and eat prey only
a few times each week or even less frequently
(3, 4), and hunter-gatherer anthropoids, in-
cluding those living today, often eat inter-
mittently depending upon food availability
(5, 6). The ability to function at a high level,
both physically and mentally, during ex-
tended periods without food may have been
of fundamental importance in our evolu-
tionary history. Many adaptations for an in-
termittent food supply are conserved among
mammals, including organs for the uptake
and storage of rapidly mobilizable glucose
(liver glycogen stores) and longer-lasting
energy substrates, such as fatty acids in
adipose tissue. Behavioral adaptations that
Author contributions: M.P.M., D.B.A., L.F., M.H., V.D.L., W.J.M.,
M.M., L.N., E.R., F.A.J.L.S., T.N.S., K.A.V., and S.P. wrote the
paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
1
To whom correspondence may be addressed. Email: mark.mattson@
nih.gov or panda@salk.edu.
www.pnas.org/cgi/doi/10.1073/pnas.1413965111 PNAS |November 25, 2014 |vol. 111|no. 47|16647–16653
PERSPECTIVE
enable food acquisition and storage per-
meate the behavioral repertoire of all spe-
cies, including humans. Indeed, the higher
cognitive capabilities of humans compared
with other species likely evolved for the
purpose of acquiring food resources; evi-
dence suggests that the earliest tools (7)
and languages (8) were invented to aid in
food acquisition.
The agricultural revolution, which began
∼10,000 y ago, resulted in the constant year-
round availability of food typical of modern
societies. Our agrarian ancestors adopted
a three meals per day eating pattern, pre-
sumably because it provided both social and
practical benefits for the daily work and
school schedules. More recently, within the
past 50 y, high calorie density foodstuffs
(refined grains, sugar, cooking oils, corn
syrup, and so on) have permeated these three
daily meals (9). When superimposed on in-
creasingly sedentary lifestyles, the consump-
tion of high-energy meals multiple times each
day plausibly contributed to the emergence of
obesity and related diseases as major causes
of morbidity and mortality (Fig. 1). Obesity
has also become a major health problem
in dogs and cats, which are often fed ad
libitum (10), and even laboratory rodents
can often be considered overfed and sed-
entary (11, 12). Indeed, animals in the wild
and hunter-gatherer humans rarely, if ever,
suffer from obesity, diabetes, and cardio-
vascular disease (5).
Circadian Rhythms, Meal Timing, and
Health
Circadian rhythms are self-sustained∼24-h
oscillations in behavior, physiology, and
metabolism. These rhythms have evolved
and permit organisms to effectively respond
to the predictable daily change in the light:
dark cycle and the resultant rhythms in
food availability in nature. Gene-expression
studies have revealed more than 10% of
expressed genes in any given organ exhibit
circadian oscillation (13). These rhythmic
transcripts encode key rate-determining
steps in neuroendocrine, signaling, and met-
abolic pathways. Such regulation temporally
separatesincompatiblecellularprocessesand
optimizes cellular and organismal fitness.
Although the circadian clock is cell-autono-
mous and is present in the majority of tissue
types, the circadian system is organized in a
hierarchical manner in which the hypotha-
lamic suprachiasmatic nucleus (SCN) func-
tions as the master circadian clock that uses
both diffusible and synaptic mechanisms to
orchestrate circadian rhythms in the periph-
eral organs at appropriate phase. Photore-
ceptive retinal ganglion cells send ambient
light information to the SCN through
monosynaptic connection to ensure that the
circadian system is entrained to the daily
light:dark cycle (14).
Whereas light is the dominant timing cue
for the SCN oscillator, time of food intake
affects the phase of theclocks in peripheral
tissues (15), including liver, muscle, and ad-
ipose tissues. For millions of years in the
absence of artificial light, the circadian
clock—in conjunction with the retinal light
input—imposed diurnal rhythms in physi-
ology and behaviors, including the activity/
rest and feeding/fasting cycle. For many of
our ancestors, food was probably scarce and
primarily consumed during daylight hours,
leaving long hours of overnight fasting.
With the advent of affordable artificial light-
ing and industrialization, modern humans
began to experience prolonged hours of illu-
mination every day and resultant extended
consumption of food. The modern lifestyle
perturbed the human circadian system in
three primary ways: shift work, exposure to
prolonged hours of artificial light, and er-
ratic eating patterns. Although it is difficult
to separate the consequence of each of these
perturbations on metabolism and physiol-
ogy, animal models and recent experimental
human studies have begun to elucidate the
mechanisms and consequence of these cir-
cadian disruptions. In industrial societies
nearly 10% of the workforce performs
night-shift work: either permanent night
work, rotating shifts, or irregular sched-
ules, in which the individuals typically
switch their wakeful hours back to the
daytime during days off to maintain a typ-
ical social life on those days. During night-
shift work the individuals are subject to
both prolonged hours of artificial lighting
and an abnormal eating schedule. Fur-
thermore, during the weekend the ten-
dency to maintain a day-active social life
imposes a jet-lag–type paradigm in which
both central and peripheral clocks attempt
to adjust to a weekend lifestyle. Although
such internal desynchrony has never been
demonstrated directly in humans, based
on animal experimental work this is pre-
sumed to result in chronic disruption of
circadian rhythms, which may help ex-
plaintheknownassociationbetweennight
work and several diseases, including car-
diovascular disease, diabetes, obesity, cer-
tain types of cancer, and neurodegenerative
diseases (16, 17).
In addition to shift work, modern human
societies experience prolonged illumination
(18) and erratic eating patterns, both of
which are known to perturb the circadian
system. In nocturnal rodent models, extended
illumination has been shown to increase
predisposition to metabolic diseases. Con-
versely, in diurnal flies a shift to nighttime
feeding compromises fat metabolism and
fecundity (19). In humans, a 12-h shift of
the sleep/wake and fasting/feeding cycle
compared with the central circadian sys-
tem, while maintaining an isocaloric diet,
reduced glucose tolerance, increased blood
pressure, and decreased the satiety hormone
leptin (20). These studies highlight the
importance of temporal organization of
sleep and feeding relative to the circadian
system. Both nutrient quality and genetic
factors appear to affect meal timing in
rodents. Mutation in the circadian clock
genePer1affecting a conserved phosphory-
lation site causes mice to consume more
food during the daytime and predisposes
them to metabolic diseases (21). The widely
used diet-induced obesity model in mice also
perturbs feeding; mice fed a high-fat diet ad
libitum consume small meals throughout day
and night (22). Both diet-induced obesity and
obesity inPer1mutant mice can be prevented
by restricting access to high-fat diet only
during the nighttime (23). The surprising
effectiveness of TRF without altering calo-
ric intake or source of calories suggests a
potentially effective meal-timing intervention
for humans. Indeed, recent human studies
suggest that earlier meal timing associates
with improved effectiveness of weight-loss
therapy in overweight and obese patients
(24, 25).
The mechanism underlying the beneficial
effect of TRF is likely complex and acts
on multiple pathways. The daily fasting and
feeding episodes trigger alternative activa-
tion of fasting-responsive cAMP response
element binding protein (CREB) and AMP
kinase, and feeding responsive insulin-
dependent mammalian target of rapamycin
(mTOR) pathways implicated in metabolic
homeostasis. In addition, these pathways
also impinge on the circadian clock and
Fig. 1.The rising tide of obesity is strongly associated
with daily calorie intake and sedentary lifestyle-promoting
transportation (refs. 84–86;www.earth-policy.org/
data_center/C23). *US, approximate value.
#
World-
wide auto production.
16648|www.pnas.org/cgi/doi/10.1073/pnas.1413965111 Mattson et al.
improve robustness of oscillation of clock
components and downstream targets (23).
Accordingly, gene-expression studies indicate
that TRF supports circadian rhythmicity of
thousands of hepatic transcripts (26).
The confluence of genomics and genetics
in mice is unraveling the pathways from the
core clock components to specific nutrient
metabolism. The nuclear hormone receptors
REV-ERBs are integral to the circadian clock
and directly regulate transcription of several
key rate-determining enzymes for fatty acid
and cholesterol metabolism (27). Although
cryptochrome proteins are strong transcrip-
tional suppressors, they also inhibit cAMP
signaling and thereby tune CREB-mediated
gluconeogenesis (28). Circadian clock down-
stream transcription factors DBP/TEF/HLF
regulate xenobiotic metabolism (29), and
KLF15 mediates nitrogen metabolism (30).
These and other modes of regulation (31)
provide a mechanistic framework by which
meal-timing affects the circadian clock and,
in turn, affects metabolic homeostasis in
mammals.
Not only does circadian phase influence
the metabolic response to food intake, food
intake itself has recently been demonstrated
to be under control by the endogenous
circadian system, independent of the sleep/
wake and fasting/feeding cycle (32), possi-
bly helping explain why breakfast is often
the smallest meal of the day or even skipped
all together.
Cellular and Molecular Mechanisms:
Insight from Intermittent Energy
Restriction and Fasting
Compared with those fed ad libitum, the
lifespans of organisms from yeast and worms,
to mice and monkeys can be extended by
dietary energy restriction (33–35). Data
collected from individuals practicing se-
vere dietary restriction indicate that humans
undergo many of the same molecular, met-
abolic, and physiologic adaptations typical
of long-lived CR rodents (36). IER/fasting
can forestall and even reverse disease pro-
cesses in animal models of various cancers,
cardiovascular disease, diabetes, and neu-
rodegenerative disorders (2). Here we briefly
highlight four general mechanisms by
which IER protects cells against injury
and disease.
Adaptive Stress Responses.Compared with
their usual ad libitum feeding conditions,
laboratory animals maintained on IER exhibit
numerous changes, suggesting heightened
adaptive stress responses at the molecular,
cellular, and organ system levels. Alternate-day
fasting prevents age-related decrements in the
antioxidant enzymes superoxide dismutase 1
and catalase in the liver cells of rats (37). IER
increases levels of the antioxidant enzymes
NADH-cytochromeb
5reductase and NAD
(P)H-quinone oxidoreductase 1 in muscle
cells of mice, and these effects are accentu-
ated by exercise (38). Numerous studies have
shown that IER can protect neurons against
oxidative, metabolic, and proteotoxic stress
in animal models of neurodegenerative
disorders, including Alzheimer’s and Par-
kinson’s diseases (39). IER can also pro-
tect the heart against ischemic damage in
an animal model of myocardial infarction
(40). Alternate-day fasting stimulates the
production of several different neuropro-
tective proteins, including the antioxidant
enzyme heme oxygenase 1, proteins involved
in mitochondrial function, and the protein
chaperones HSP-70 and GRP-78 (41, 42).
Moreover, IER increases the production of
trophic factors that promote neuronal sur-
vival, neurogenesis, and the formation and
strengthening of synapses in the brain (43).
Taken together, these data suggestthat ben-
eficial effects of IER involve the general
biological phenomenon of“hormesis”or
“preconditioning,”in which exposure of
cells and organisms to a mild stress results in
adaptive responses that protect against more
severe stress.
Bioenergetics.When humans switch from
eating three full meals per day to an IER diet,
such as one moderate size meal every other
dayoronly500–600 calories 2 d/wk, they
exhibit robust changes in energy metabolism
characterized by increased insulin sensitivity,
reduced levels of insulin and leptin, mobili-
zation of fatty acids, and elevation of ketone
levels (44–47). Ketones, such asβ-hydroxy-
butyrate, are known to have beneficial effects
on cells with a high energy demand, such as
neurons in the brain (Fig. 2) (48, 49). In mice,
alternate-day fasting can greatly increase
insulin sensitivity even without a major re-
duction in body weight (50), and in humans
IER can increase insulin sensitivity more
than daily calorie restriction that achieves
similar weight loss (45, 51). Dietary energy
restriction can prevent age-related decline
in mitochondrial oxidative capacity in skele-
tal muscle, and can induce mitochondrial
biogenesis (52). Brain bioenergetics may
also be bolstered by IER. For example,
brain-derived neurotrophic factor (BDNF),
which is up-regulated in hippocampal
neurons in response to IER and exercise,
activates the transcription factor CREB,
which then induces peroxisome proliferator-
activated receptor gamma coactivator 1-α
(PGC-1α) expression and mitochondrial
biogenesis (53). The latter study showed
Fig. 2.A metabolic shift to ketogenesis that occurs with fasting bolsters neuronal bioenergetics. Liver glycogen
stores are typically depleted within 10–12 h of fasting, which is followed by liberation of fatty acids from adipose
tissue cells into the blood. The fatty acids are then transported into liver cells where they are oxidized to generate
Acetyl-CoA. Acetyl-CoA is then converted to 3-hydroxy-3-methylgluaryl-CoA, which is in turn used to generate the
ketones acetoacetate andβ-hydroxybutyrate (β-OHB). The ketones are released into the blood and are transported
into various tissues, including the brain, where they are taken up by neurons and used to produce acetyl-CoA. Acetyl-
CoA enters the tricarboxylic acid (TCA) cycle to generate ATP.
Mattson et al. PNAS |November 25, 2014 |vol. 111|no. 47|16649
PERSPECTIVE
that PGC-1αand mitochondrial biogenesis
are critical for the formation of synapses in
developing hippocampal neurons and the
maintenance of synapses in the hippo-
campus of adult mice. Because impaired
mitochondrial biogenesis and function oc-
cur during aging and chronic disease states,
such as sarcopenia and neurodegenerative
disorders, it is important to consider the
impact of the frequency and circadian timing
of meals on the development and progression
of such disorders.
Whereas IER/fasting is beneficial and
overeating detrimental for many types of
normal cells, the converse is true for tumor
cells. Cells in tumors exhibit major mito-
chondrial abnormalities and generate their
ATP primarily from glycolysis rather than
oxidative phosphorylation (54). Moreover,
tumors are highly vascularized and so their
cells have access to large amounts of circu-
lating glucose. Animal models have consis-
tently shown that IER inhibits and even
reverses the growth of a range of tumors,
including neuroblastoma, breast, and ovarian
cancers (55). The shift to ketogenesis may
play an important role in suppression of
tumor growth by IER/fasting because many
tumor cells are largely unable to use ketones
as an energy source; accordingly, ketogenic
diets may potentiate the antitumor effects of
IER (54). Although preliminary, recent case
studies in human patients suggest potential
applications of IER in the treatment of a
range of cancers, including breast, ovar-
ian, prostate, and glioblastoma (56, 57).
Indeed, evolutionary theory predicts that
collected random mutations will prevent
tumor cells from making the necessary met-
abolic adaptations to IER (58).
Inflammation.All major diseases, including
cardiovascular disease, diabetes, neurodegen-
erative disorders, arthritis, and cancers in-
volve chronic inflammation in the affected
tissues and, in many cases, systemically (59).
Local tissue inflammation involves hyper-
activation of macrophages (microglia in the
brain) which produce proinflammatory
cytokines (TNF, IL-1β, IL-6) and reactive
oxygen species. Overweight and obesity
promote inflammation, and IER suppresses
inflammation in human subjects and ani-
mal models of diseases. Obese women who
changed their diet from multiple daily meals
to alternate-day energy restriction exhibited
significant reductions in levels of circulating
TNF and IL-6 (60). In asthma patients, 2 mo
of alternate-day energy restriction reduced
circulating TNF and markers of oxidative
stress, and improved asthma symptoms and
airway resistance (44). However, because
weight loss may reduce inflammation re-
gardless of the dietary change inducing the
weight loss, it will be important to deter-
mine if and how eating patterns modify
inflammation independently of weight loss.
Multiple studies have shown that fasting
can lessen symptoms in patients with rheu-
matoid arthritis (61), and data from animal
studies suggest that the pathogenesis of other
autoimmune disorders may also be counter-
acted by IER, including multiple sclerosis
(62), lupus erythematosus (63), and type I
diabetes (64). In a mouse model of stroke,
IER suppressed elevations of TNF and IL-1β
in the ischemic cerebral cortex and striatum,
which was associated with improved func-
tional outcome (41). Inflammation is in-
creasingly recognized as a contributing factor
for cancer cell growth (65) and, because ex-
cessive energy intake promotes inflammation,
it is likely that suppression of inflammation
plays a role in the inhibition of tumor growth
by IER. Whereas inhibiting immune re-
sponses to autoantigens and sterile tissue
injuries can be beneficial, suppression of
immune responses to infectious agents is
detrimental. It will therefore be impor-
tant to determine whether eating regimens
such as TRF and IER affect immune re-
sponses to pathogens, an as yet unexplored
area of investigation.
Improved Repair and Removal of Dam-
aged Molecules and Organelles.Cells
possess dedicated mechanisms for the re-
moval of damaged molecules and organelles.
One mechanism involves the molecular
“tagging”of damaged proteins with ubiq-
uitin, which targets them for degradation in
the proteasome (66). In a second and more
elaborate mechanism called autophagy,
damaged and dysfunctional proteins, mem-
branes, and organelles are directed to and
degraded in lysosomes (67). Energy and
nutrient (particularly amino acids) intake
have been shown to have consistent effects
on autophagy. When organisms ingest regu-
lar meals, their cells receive a relatively
steady supply of nutrients and so remain in
a“growth mode”in which protein synthesis
is robust and autophagy is suppressed (68).
The nutrient-responsive mTOR pathway
negatively regulates autophagy. Accordingly,
fasting inhibits the mTOR pathway and
stimulates autophagy in cells of many tissues,
including liver, kidney, and skeletal muscle
(69–71). In this way, fasting“cleanses”
cells of damaged molecules and organelles.
Rats maintained on energy-restricted
diets exhibit reduced accumulation of
polyubiquitinated proteins and evidence
of increased autophagy in peripheral
nerves compared with rats fed ad libi-
tum (72). In a mouse model of Charcot-
Marie-Tooth type 1A, an inherited disorder
characterized by demyelination of peripheral
nerves, IER improved motor performance
and reduced demyelination by a mecha-
nism involving enhanced autophagy and
reduced accumulation of myelin protein
PMP22 aggregates (73). A common feature of
many major chronic diseases is the abnormal/
excessive accumulation of protein aggregates
within and outside of cells; examples include
intracellularα-synuclein in Parkinson’sdis-
ease and extracellular amyloidβ-peptide and
intracellular Tau protein in Alzheimer’s
disease (74, 75). In addition to the frequency
of meals, the circadian timing of meals is
likely to affect the responses of the cellular
machineries for clearance of damaged
proteins and organelles (76). Autophagy is
regulated in a diurnal rhythm in many cell
types, and this rhythm can be altered by
changing the timing of food intake. It is
therefore reasonable to consider that meal
timing has an impact on diseases that involve
impaired or insufficient autophagy.
Society-Wide Implications
The high rates of childhood and adult obesity
and the diseases they foster is a major burden
to our society. As findings from basic re-
search studies and controlled interventional
trials accrue, consensus recommendations for
healthy patterns of meal frequency and
diurnal timing may eventually emerge. If
sufficient evidence does emerge to support
public health and clinical recommendations
to alter meal patterning, there will be nu-
merous forces at play in the acceptance or
resistance to such recommendations. First
and perhaps foremost is cultural tradition.
Three meals plus snacks daily has become
the norm during the past half-century, such
that a majority of American children are
accustomed to this eating pattern. Second,
the agriculture, food processing, food retail,
and restaurant industries and all of the affil-
iated industries that serve or promote food—
from airlines to concert stadiums to television
cooking shows to advertising and others—
still all have established practices and finan-
cial interests and these interests may affect
receptivity to proposed shifts in eating pat-
terns and potential decreases in total food
purchased. Third, the willingness and ability
of the American health care system, includ-
ing medical training and practice, to em-
phasize prevention and lifestyles will be a
key factor in success or lack thereof.
We believe that it is important to consider
how“prescriptions”for meal frequency and
timing can be developed, validated, and
16650|www.pnas.org/cgi/doi/10.1073/pnas.1413965111 Mattson et al.
implemented in light of the current in-
dustrial, cultural, and institutional pressures
to maintain the status quo of daily over-
consumption of food. In doing so, it will be
importanttoensurethatweprovidethe
public with accurate information on eating
patterns and health. For example, despite
equivocal and even contradictory scientific
evidence, breakfast is often touted as
a weight-control aid (77), but recent evi-
dence has suggested that it may not be (78).
Primary education and media outlets
should dispense up-to-date information on
healthy eating, including the frequency and
circadian timing of meals. Although regu-
latory agencies must play an important role
in developing recommendations and facili-
tating their implementation, it may also be
helpful for parents to lead by example and
establish healthy eating patterns in their
children. Additionally, the inclusion of
science-based information on eating pat-
terns and health in primary and secondary
education may help stem the rising tide of
overeating and related poor health in our
children. The medical community could
play a central role in developing and
implementing prescriptions for long-term
daily energy restriction or IER that can be
incorporated into most daily home and
workplace environments. Examples of such
prescriptions include fasting or caloric re-
striction (e.g., 500 calories) on alternate days
or 2 d each week, or forgoing breakfast and
lunch several days each week (Fig. 3). The
available evidence suggests that patients may
be able to comply with such diets when
there is rigorous follow-up (44, 45, 47), and
it will be important to determine if com-
pliance would increase further if patients
were able to choose an eating pattern-based
prescription that best fits their weekly rou-
tines. Recent findings suggest that it may be
possible for many people to adopt a long-
term change in their lifestyle from eating
three meals plus snacks every day to an IER
diet if they are able to keep on the new
eating pattern during a transition period of
approximately 1 mo (45). Moreover, for
many people who are overweight IER may
facilitate their maintenance of an overall
reduction in energy intake compared with
prescriptions for daily caloric restriction.
Future Directions
Further animal studies are required to better
elucidate the cellular and molecular mecha-
nisms by which meal frequency, IER, and
TRF affect health and disease susceptibility,
as well as the impact of eating patterns on
extant disease processes in various experi-
mental models. For example, it will be of
great interest to know the effects of IER and
TRF on gene expression, epigenetic mark-
ers (methylation and acetylation), and dis-
ease-relevant pathways in multiple tissues
throughout the body and nervous system.
The overlapping and complementary effects
of exercise and healthy eating patterns
on functionality and disease resistance
should be elucidated. Intervention studies
of IER and TRF, particularly randomized
controlled trials (RCTs), should be per-
formed in various groups of human sub-
jects, including those who are healthy and
those with diseases, such as obesity, di-
abetes, cancer, cardiovascular disease, and
neurodegenerative disorders. RCTs should
include functional outcomes as well as bio-
markers relevant to disease risk and patho-
genesis. Thus, far, relatively few RCTs of
IER and TRF have been performed in hu-
man subjects, with the results of several
studies of alternate-day and twice weekly
energy restriction demonstrating weight
loss and abdominal fat reduction and
suggesting improvements in indicators
of energy and lipid metabolism and in-
flammation (44–46, 51, 61). On the other
hand, a study of TRF in which healthy
normal weight subjects consumed a bal-
anced daily food intake within a 4-h or
12-h time period each day revealed no
improvement (79, 80), which is similar to
the lack of any short-term benefit of TRF
in mice when the animals are fed a bal-
anced diet (23). This finding suggests that
the short-term benefits of TRF might depend
on the diet and body composition. It will
also be critical to evaluate long-term ad-
herence of different subject populations
to IER and TRF protocols to evaluate
their feasibility for broad applications for
sustained weight reduction and disease
risk reduction.
Genetic factors can determine whether the
lifespan of a particular strain of mouse or
rat is increased, unaffected, or even de-
creased, by lifelong CR or TRF, with inbred
animals generally responding less well to CR
(81). Understanding the mechanism of TRF
will help to predict whether a certain eating
pattern is beneficial or whether individuals
with specific genotype are predisposed to
erratic eating patterns. Missense mutation in
circadian clock component Per1 has been
shown to affect eating patterns in mice
(21). However, the presence of intact food
anticipatory activity in SCN ablated rodents
or those lacking functional circadian oscil-
lator genes points to yet-unidentified genes
and circuits in eating-pattern determination
(82, 83). Humans are highly heterogeneous
with regard to their genetic composition,
epigenetic landscape, and the environmental
factors to which they are exposed through-
out life. It is therefore likely that there will be
considerable variability among human sub-
jects in the responses of their cells and organ
systems (and overall health) to different
eating patterns. Although there is sufficient
evidence to suggest that CR and IER can
improve health indicators in most or all
obese human subjects (2), data are lacking
with regard to normal weight subjects.
Fig. 3.Patterns of daily and weekly food consumption. The upper illustration shows five different patterns of
food consumption during a 24-h period. A: Eating three large meals plus snacks spread throughout a 16-h period
of wakefulness; this is the common eating pattern of food consumption upon which the epidemic of obesity,
diabetes, and associated chronic diseases has emerged. B–D: Examples if time-restricted eating patterns in which
food is consumed as three (B) or two (C) regular size meals, or three small meals (D). E: Complete fast. Examples of
weekly eating schedules are shown in thelower right. ER, energy restriction; IER, intermittent energy restriction;
TRF, time-restricted feeding.
Mattson et al. PNAS |November 25, 2014 |vol. 111|no. 47|16651
PERSPECTIVE
Insight into genetic and epigenetic factors
that affect responses to specific eating
patterns could be obtained from RCTs of
TRF and IER regimens in normal weight
subjects in which biomarkers of health and
disease risk are measured (blood pressure,
heart rate variability, insulin resistance,
lipid profiles, adipokines, ketones, and
so forth).
It would be particularly valuable to de-
sign RCT in human subjects with head-to-
head comparisons of multiple eating pat-
terns, such as those shown in Fig. 3. Once
the eating patterns that promote optimal
health are established, what can be done to
encourage, enable and empower individ-
uals to modify their food choices and
eating patterns? Implementing any such
changes will be challenging, as a half-cen-
tury of research on behavioral approaches to
weight control suggests. That said, the field
of behavioral science is continually evolv-
ing, as is the growth and quality of mobile
information technology, which may serve
to buttress efforts. We are hopeful that in
thefuture,wemaybebetterabletohelp
individuals achieve the healthy behavior
changes they desire.
ACKNOWLEDGMENTS. This article incorporates in-
formation from a workshop on“Eating Patterns and
Disease,”which can be viewed atvideocast.nih.gov/
summary.asp?Live=13746&bhcp=1, and was supported
by the National Institute on Aging Intramural Re-
search Program and the Glenn Foundation for Medi-
cal Research. Relevant research in the authors’
laboratories are supported by NIH intramural support
(to M.P.M.); NIH Grants P30DK056336 (to D.B.A.),
P01AG034906 (to V.D.L.), R01NS041012 (to L.N.),
P30DK072476 (to E.R.), R01DK099512 (to F.A.J.L.S.),
R01NS055195 (to T.N.S.), R01HL106228 (to K.A.V.),
and R01DK091618 (to S.P.); the European Union’s
Seventh Framework Programme MOPACT [mobilising
the potential of active ageing in Europe; FP7-SSH-
2012-1 Grant 320333 (to L.F.)]; a grant from Genesis
Breast Cancer Prevention, UK (to M.H.); and Belgian
Foundation for Scientific Medical Research Grant
3.4520.07 (to W.J.M.).
1Casazza K, et al. (2013) Myths, presumptions, and facts about
obesity.N Engl J Med368(5):446–454.
2Longo VD, Mattson MP (2014) Fasting: Molecular mechanisms
and clinical applications.Cell Metab19(2):181–192.
3Gervasi V, et al. (2012) Predicting the potential demographic
impact of predators on their prey: A comparative analysis of two
carnivore-ungulate systems in Scandinavia.J Anim Ecol81(2):
443–454.
4Metz MC, Smith DW, Vucetich JA, Stahler DR, Peterson RO (2012)
Seasonal patterns of predation for gray wolves in the multi-prey
system of Yellowstone National Park.J Anim Ecol81(3):553–563.
5Cordain L, Eaton SB, Miller JB, Mann N, Hill K (2002) The
paradoxical nature of hunter-gatherer diets: Meat-based, yet non-
atherogenic.Eur J Clin Nutr56(Suppl 1):S42–S52.
6Ströhle A, Hahn A, Sebastian A (2010) Estimation of the diet-
dependent net acid load in 229 worldwide historically studied hunter-
gatherer societies.Am J Clin Nutr91(2):406–412.
7van Schaik CP, Deaner RO, Merrill MY (1999) The conditions for
tool use in primates: Implications for the evolution of material
culture.J Hum Evol36(6):719–741.
8Allen JS (2012)“Theory of food”as a neurocognitive adaptation.
Am J Hum Biol24(2):123–129.
9Tillotson JE (2004) America’s obesity: Conflicting public policies,
industrial economic development, and unintended human
consequences.Annu Rev Nutr24:617–643.
10Zoran DL (2010) Obesity in dogs and cats: A metabolic and
endocrine disorder.Vet Clin North Am Small Anim Pract40(2):
221–239.
11Martin B, Ji S, Maudsley S, Mattson MP (2010)“Control”
laboratory rodents are metabolically morbid: Why it matters.Proc
Natl Acad Sci USA107(14):6127–6133.
12Klimentidis YC, et al. (2011) Canaries in the coal mine: A cross-
species analysis of the plurality of obesity epidemics.Proc Biol Sci
278(1712):1626–1632.
13Panda S, Hogenesch JB, Kay SA (2002) Circadian rhythms from
flies to human.Nature417(6886):329–335.
14Hatori M, Panda S (2010) The emerging roles of melanopsin in
behavioral adaptation to light.Trends Mol Med16(10):435–446.
15Stokkan KA, Yamazaki S, Tei H, Sakaki Y, Menaker M (2001)
Entrainment of the circadian clock in the liver by feeding.Science
291(5503):490–493.
16Stevens RG, Brainard GC, Blask DE, Lockley SW, Motta ME (2014)
Breast cancer and circadian disruption from electric lighting in the
modern world.CA Cancer J Clin64(3):207–218.
17Morris CJ, Yang JN, Scheer FA (2012) The impact of the circadian
timing system on cardiovascular and metabolic function.Prog Brain
Res199:337–358.
18Wright KP Jr, et al. (2013) Entrainment of the human circadian
clock to the natural light-dark cycle.Curr Biol23(16):1554–1558.
19Gill S, Panda S (2011) Feeding mistiming decreases reproductive
fitness in flies.Cell Metab13(6):613–614.
20Scheer FA, Hilton MF, Mantzoros CS, Shea SA (2009) Adverse
metabolic and cardiovascular consequences of circadian
misalignment.Proc Natl Acad Sci USA106(11):4453–4458.
21Liu Z, et al. (2014) PER1 phosphorylation specifies feeding
rhythm in mice.Cell Rep7(5):1509–1520.
22Kohsaka A, et al. (2007) High-fat diet disrupts behavioral and
molecular circadian rhythms in mice.Cell Metab6(5):414–421.
23Hatori M, et al. (2012) Time-restricted feeding without reducing
caloric intake prevents metabolic diseases in mice fed a high-fat diet.
Cell Metab15(6):848–860.
24Garaulet M, et al. (2013) Timing of food intake predicts weight
loss effectiveness.Int J Obes (Lond)37(4):604–611.
25Jakubowicz D, Barnea M, Wainstein J, Froy O (2013) High caloric
intake at breakfast vs. dinner differentially influences weight loss of
overweight and obese women.Obesity (Silver Spring)21(12):
2504–2512.
26Vollmers C, et al. (2012) Circadian oscillations of protein-coding
and regulatory RNAs in a highly dynamic mammalian liver
epigenome.Cell Metab16(6):833–845.
27Cho H, et al. (2010) Regulation of circadian behaviour and
metabolism by REV-ERB-αand REV-ERB-β.Nature485(7396):
123–127.
28Zhang EE, et al. (2010) Cryptochrome mediates circadian
regulation of cAMP signaling and hepatic gluconeogenesis.Nat Med
16(10):1152–1156.
29Gachon F, Olela FF, Schaad O, Descombes P, Schibler U (2006)
The circadian PAR-domain basic leucine zipper transcription factors
DBP, TEF, and HLF modulate basal and inducible xenobiotic
detoxification.Cell Metab4(1):25–36.
30Jeyaraj D, et al. (2006) Klf15 orchestrates circadian nitrogen
homeostasis.Cell Metab15(3):311–323.
31Bass J, Takahashi JS (2010) Circadian integration of metabolism
and energetics.Science330(6009):1349–1354.
32Scheer FA, Morris CJ, Shea SA (2013) The internal circadian clock
increases hunger and appetite in the evening independent of food
intake and other behaviors.Obesity (Silver Spring)21(3):421–423.
33Goodrick CL, Ingram DK, Reynolds MA, Freeman JR, Cider NL
(1983) Effects of intermittent feeding upon growth, activity, and
lifespan in rats allowed voluntary exercise.Exp Aging Res9(3):
203–209.
34Honjoh S, Yamamoto T, Uno M, Nishida E (2009) Signalling
through RHEB-1 mediates intermittent fasting-induced longevity in
C. elegans.Nature457(7230):726–730.
35Colman RJ, et al. (2014) Caloric restriction reduces age-related
and all-cause mortality in rhesus monkeys.Nat Commun5:3557.
36Rizza W, Veronese N, Fontana L (2014) What are the roles of
calorie restriction and diet quality in promoting healthy longevity?
Ageing Res Rev13:38–45.
37Pieri C, et al. (1992) Food restriction in female Wistar rats: V. Lipid
peroxidation and antioxidant enzymes in the liver.Arch Gerontol
Geriatr14(1):93–99.
38Rodríguez-Bies E, Navas P, López-Lluch G (2014) Age-dependent
effect of every-other-day feeding and aerobic exercise in ubiquinone
levels and related antioxidant activities in mice muscle.J Gerontol A
Biol Sci Med Sci, 10.1093/gerona/glu002.
39Mattson MP (2012) Energy intake and exercise as determinants
of brain health and vulnerability to injury and disease.Cell Metab
16(6):706–722.
40Ahmet I, Wan R, Mattson MP, Lakatta EG, Talan M (2005)
Cardioprotection by intermittent fasting in rats.Circulation112(20):
3115–3121.
41Arumugam TV, et al. (2010) Age and energy intake interact to
modify cell stress pathways and stroke outcome.Ann Neurol67(1):
41–52.
42Poon HF, et al. (2006) Proteomics analysis provides insight into
caloric restriction mediated oxidation and expression of brain proteins
associated with age-related impaired cellular processes:
Mitochondrial dysfunction, glutamate dysregulation and impaired
protein synthes-is.Neurobiol Aging27(7):1020–1034.
43Marosi K, Mattson MP (2014) BDNF mediates adaptive brain and
body responses to energetic challenges.Trends Endocrinol Metab
25(2):89–98.
44Johnson JB, et al. (2007) Alternate day calorie restriction
improves clinical findings and reduces markers of oxidative stress and
inflammation in overweight adults with moderate asthma.Free Radic
Biol Med42(5):665–674.
45Harvie MN, et al. (2011) The effects of intermittent or continuous
energy restriction on weight loss and metabolic disease risk markers:
A randomized trial in young overweight women.Int J Obes (Lond)
35(5):714–727.
46Klempel MC, Kroeger CM, Varady KA (2013) Alternate day
fasting (ADF) with a high-fat diet produces similar weight loss and
cardio-protection as ADF with a low-fat diet.Metabolism62(1):
137–143.
47Varady KA, et al. (2013) Alternate day fasting for weight loss in
normal weight and overweight subjects: A randomized controlled
trial.Nutr J12(1):146.
48Kashiwaya Y, et al. (2013) A ketone ester diet exhibits anxiolytic
and cognition-sparing properties, and lessens amyloid and tau
pathologies in a mouse model of Alzheimer’s disease.Neurobiol
Aging34(6):1530–1539.
49Newman JC, Verdin E (2014) Ketone bodies as signaling
metabolites.Trends Endocrinol Metab25(1):42–52.
50Anson RM, et al. (2003) Intermittent fasting dissociates beneficial
effects of dietary restriction on glucose metabolism and neuronal
resistance to injury from calorie intake.Proc Natl Acad Sci USA
100(10):6216–6220.
51Harvie M, et al. (2013) The effect of intermittent energy and
carbohydrate restriction v. daily energy restriction on weight loss and
metabolic disease risk markers in overweight women.Br J Nutr
110(8):1534–1547.
52Martin-Montalvo A, de Cabo R (2013) Mitochondrial metabolic
reprogramming induced by calorie restriction.Antioxid Redox Signal
19(3):310–320.
53Cheng A, et al. (2012) Involvement of PGC-1αin the formation
and maintenance of neuronal dendritic spines.Nat Commun3:1250.
54Seyfried TN, Flores RE, Poff AM, D’Agostino DP (2014) Cancer as
a metabolic disease: Implications for novel therapeutics.
Carcinogenesis35(3):515–527.
55Lee C, et al. (2012) Fasting cycles retard growth of tumors and
sensitize a range of cancer cell types to chemotherapy.Sci Transl Med
4(124):124ra27.
56Safdie FM, et al. (2009) Fasting and cancer treatment in humans:
A case series report.Aging (Albany, NY Online)1(12):988–1007.
57Zuccoli G, et al. (2010) Metabolic management of glioblastoma
multiforme using standard therapy together with a restricted
ketogenic diet: Case report.Nutr Metab (Lond)7:33.
58Seyfried TN (2012)Cancer as a Metabolic Disease: On the
Origin, Management and Prevention of Cancer(John Wiley & Sons,
Hoboken, NJ), pp 261–275.
59McDade TW (2012) Early environments and the ecology of
inflammation.Proc Natl Acad Sci USA109(Suppl 2):17281–17288.
60Kroeger CM, et al. (2012) Improvement in coronary heart disease
risk factors during an intermittent fasting/calorie restriction regimen:
Relationship to adipokine modulations.Nutr Metab (Lond)9(1):98.
61Müller H, de Toledo FW, Resch KL (2001) Fasting followed by
vegetarian diet in patients with rheumatoid arthritis: A systematic
review.Scand J Rheumatol30(1):1–10.
62Esquifino AI, Cano P, Jimenez-Ortega V, Fernández-Mateos MP,
Cardinali DP (2007) Immune response after experimental allergic
encephalomyelitis in rats subjected to calorie restriction.
J Neuroinflammation4:6.
63Muthukumar A, Zaman K, Lawrence R, Barnes JL, Fernandes G
(2003) Food restriction and fish oil suppress atherogenic risk factors in
lupus-prone (NZB x NZW) F1 mice.J Clin Immunol23(1):23–33.
16652|www.pnas.org/cgi/doi/10.1073/pnas.1413965111 Mattson et al.
64Belkacemi L, et al. (2012) Intermittent fasting modulation of the
diabetic syndrome in streptozotocin-injected rats.Int J Endocrinol
2012:962012.
65Elinav E, et al. (2013) Inflammation-induced cancer: Crosstalk
between tumours, immune cells and microorganisms.Nat Rev
Cancer13(11):759–771.
66Maupin-Furlow J (2012) Proteasomes and protein conjugation
across domains of life.Nat Rev Microbiol10(2):100–111.
67Rubinsztein DC, Mariño G, Kroemer G (2011) Autophagy and
aging.Cell146(5):682–695.
68Speakman JR, Mitchell SE (2011) Caloric restriction.Mol Aspects
Med32(3):159–221.
69Donati A, Recchia G, Cavallini G, Bergamini E (2008) Effect of
aging and anti-aging caloric restriction on the endocrine regulation of
rat liver autophagy.J Gerontol A Biol Sci Med Sci63(6):550–555.
70Kume S, et al. (2010) Calorie restriction enhances cell adaptation
to hypoxia through Sirt1-dependent mitochondrial autophagy in
mouse aged kidney.J Clin Invest120(4):1043–1055.
71Wohlgemuth SE, Seo AY, Marzetti E, Lees HA, Leeuwenburgh C
(2010) Skeletal muscle autophagy and apoptosis during aging:
Effects of calorie restriction and life-long exercise.Exp Gerontol45(2):
138–148.
72Lee S, Notterpek L (2013) Dietary restriction supports peripheral
nerve health by enhancing endogenous protein quality control
mechanisms.Exp Gerontol48(10):1085–1090.
73Madorsky I, et al. (2009) Intermittent fasting alleviates the
neuropathic phenotype in a mouse model of Charcot-Marie-Tooth
disease.Neurobiol Dis34(1):146–154.
74Moreno-Gonzalez I, Soto C (2011) Misfolded protein aggregates:
Mechanisms, structures and potential for disease transmission.Semin
Cell Dev Biol22(5):482–487.
75Jucker M, Walker LC (2011) Pathogenic protein seeding in
Alzheimer disease and other neurodegenerative disorders.Ann
Neurol70(4):532–540.
76Ma D, Li S, Molusky MM, Lin JD (2012) Circadian autophagy
rhythm: A link between clock and metabolism?Trends Endocrinol
Metab23(7):319–325.
77Brown AW, Bohan Brown MM, Allison DB (2013) Belief beyond
the evidence: Using the proposed effect of breakfast on obesity to
show 2 practices that distort scientific evidence.Am J Clin Nutr98(5):
1298–1308.
78Dhurandhar EJ, et al. (2014) The effectiveness of breakfast
recommendations on weight loss: A randomized controlled trial.Am
J Clin Nutr100(2):507–513.
79Stote KS, et al. (2007) A controlled trial of reduced meal
frequency without caloric restriction in healthy, normal-weight,
middle-aged adults.Am J Clin Nutr85(4):981–988.
80Carlson O, et al. (2007) Impact of reduced meal frequency without
caloric restriction on glucose regulation in healthy, normal-weight
middle-aged men and women.Metabolism56(12):1729–1734.
81Swindell WR (2012) Dietary restriction in rats and mice: A meta-
analysis and review of the evidence for genotype-dependent effects
on lifespan.Ageing Res Rev11(2):254–270.
82Pendergast JS, Oda GA, Niswender KD, Yamazaki S (2012) Period
determination in the food-entrainable and methamphetamine-
sensitive circadian oscillator(s).Proc Natl Acad Sci USA109(35):
14218–14223.
83Mulder CK, Papantoniou C, Gerkema MP, Van Der Zee EA
(2014) Neither the SCN nor the adrenals are required for
circadian time-place learning in mice.Chronobiol Int31(9):
1075–1092.
84Fryar CD, Carroll MD, Ogden CL (2012)Prevalence of
Overweight, Obesity, and Extreme Obesity Among Adults: United
States, Trends 1960–1962 Through 2009–2010(National Center for
Health Statistics, Hyattsville, MD). Available at www.cdc.gov/nchs/
data/hestat/obesity_adult_09_10/obesity_adult_09_10.htm.
Accessed November 3, 2014.
85United States Department of Agriculture (2003) Profiling food
consumption in America.Agriculture Fact Book(US Department of
Agriculture, Washington, DC), pp 13–21.
86Day JC, Janus A, Davis J (2005)Computer and Internet Use in the
United States: 2003(US Census Bureau, Washington, DC). Available
at www.census.gov/prod/2005pubs/p23-208.pdf. Accessed
November 3, 2014.
Mattson et al. PNAS |November 25, 2014 |vol. 111|no. 47|16653
PERSPECTIVE