1. REVIEWS
Circadian rhythms and memory
formation
Jason R. Gerstner*§ and Jerry C. P. Yin*‡
Abstract | There has been considerable progress in elucidating the molecular mechanisms
that contribute to memory formation and the generation of circadian rhythms. However, it is
not well understood how these two processes interact to generate long-term memory.
Recent studies in both vertebrate and invertebrate models have shown time-of-day effects
on neurophysiology and memory formation, and have revealed a possible role for cycling
molecules in memory persistence. Together, these studies suggest that common mechanisms
underlie circadian rhythmicity and long-term memory formation.
Zeitgeber Circadian rhythms are basic biological phenomena that circadian alterations on neurophysiological processes
A German word that means exist throughout phylogeny. They are influenced by that involve synaptic plasticity (such as long-term poten-
‘time-giver’. It refers to an zeitgebers and regulate various physiological events that tiation) and on memory formation in nocturnal (night-
exogenous cue, such as the include the cell cycle, body temperature, metabolism, feed- active), diurnal (day-active), and crepuscular model
light–dark cycle, that entrains
a circadian rhythm.
ing and, perhaps most notably, the sleep–wake cycle. Far systems. On the basis of the cycling pattern of molecu-
less well understood is the relationship between circadian lar cascades that are involved in memory formation, we
Circadian rhythm rhythm biology and memory formation. The impact of address whether the cyclical reactivation of these cas-
The regular cycling of biological time-of-day effects and of circadian rhythms on cognitive cades over the 24-hour day is essentially independent
processes in an organism over
performance in humans1–3 and on memory in animals4–7 from inputs of the core time-keeping cells that are known
a ~24-hour period that occurs
regardless of the zeitgeber.
have been studied for decades, and there has been a to contribute to locomotor rhythm output. This Review
renewed interest in this topic in light of an increased expands on previously understood circadian effects on
understanding of the genetic, molecular and systems- memory at the behavioural and physiological level, by
level events that underlie these complex processes8. focusing on recent data that show a possible involvement
Recent discoveries have shown a high level of integration of circadian cycling of specific molecular pathways in
*Department of Genetics,
between cellular signalling cascades (such as the cyclic long-term memory formation. Further background
University of Wisconsin– AMP–mitogen-activated protein kinase (MAPK)– information has been published elsewhere on circadian
Madison, 3476 Genetics and cAMP-responsive element-binding protein (CREB) rhythms9,10 and memory formation11,12.
Biotechnology, 425 Henry pathway) that regulate circadian rhythms and memory
Mall, Madison, Wisconsin
processing. Disruption of circadian rhythms or specific Are clock genes memory genes?
53706, USA.
‡
Department of Neurology, signalling cascades that undergo time-of-day-depend- The initial characterization of the molecular players
University of Wisconsin– ent cycling, by behavioural, environmental, genetic or involved in the generation of circadian rhythms was
Madison, 3434 Genetics and pharmacological means, has negative consequences on carried out in the Drosophila melanogaster model. Over
Biotechnology, 425 Henry memory and cognitive performance in various tasks and three decades ago, work on fruitflies showed that the
Mall, Madison, Wisconsin
53706, USA.
in several species. Given that modern society is becoming periodic timing of the eclosion rhythm was dependent
§
Present address: Center for less dependent on the natural 24-hour light–dark cycle, on the strain of fly. This suggested a genetic basis for
Sleep and Respiratory an increased understanding of the functional relationship the circadian regulation of this process, prompting a for-
Neurobiology, University of between circadian rhythms and cognitive function has ward mutagenesis screen that identified the first clock
Pennsylvania School of
broad implications for public health9. gene, period (per)13. This gene was eventually cloned
Medicine, Translational
Research Laboratories, Here, we summarize studies that have shown a independently by separate laboratories14,15. Levels of per
125 South 31st Street, time-of-day effect on memory formation and compare mRNA and protein were shown to cycle in a circadian
Suite 2100, Philadelphia, the emerging common themes in various invertebrate manner in flies and mammals and to be a part of a phy-
Pennsylvania 19104‑3403, and vertebrate species. We first describe the molecular logenetically conserved transcriptional auto-regulatory
USA.
E‑mails: jrgerstn@gmail.com;
pathways and time-of-day-dependent neuronal activity feedback loop (FIG. 1) that is necessary for the synchro-
jcyin@wisc.edu patterns that are conserved in circadian pacemaker cells nized expression of the circadian rhythm of locomotor
doi:10.1038/nrn2881 in flies and rodents. Next, we present work that shows activity 16,17. In D. melanogaster, mutations in per result
NATuRE REvIEWs | NeuroscieNce vOLuME 11 | AuGusT 2010 | 577
© 2010 Macmillan Publishers Limited. All rights reserved

2. REVIEWS
Time-of-day effect in differences in the length of the eclosion rhythm, and activities under free-running conditions. This suggests
The effect of the specific point include long (perL), short (pers), and arrhythmic (per 0) that per regulates memory independently of its role in
in time during the day–night phenotypes. Interestingly, these mutations cause correla- eclosion or in the generation of circadian rhythms.
cycle on the biological tive changes in the periodicity of the circadian locomo- Previous work has shown that there is a role for
processes of an organism. The
effect can be dependent or
tor activity rhythm in adult flies. under constant dark another transcription factor, CREB, in the core circadian
independent of a zeitgeber. (DD) conditions, pers flies have a shortened circadian clock of flies20 and mammals21. In addition, a functional
rhythm, perL have a lengthened circadian rhythm, and cAMP-responsive element (CRE) site in the promoter of
Long-term potentiation per 0 are arrhythmic. This evidence suggests that the sin- mouse per genes that binds CREB has been described22,
A persistent increase in
gle clock gene per has pleiotropic effects on the timing of suggesting a link between CREB activity and PER activ-
synaptic strength following
high-frequency stimulation of a
two separate processes at different developmental stages. ity in circadian rhythm generation (FIG. 1). A functional
synapse. Do clock genes have a role in the time-of-day effects on relationship between CREB activity and per expression
memory formation? Curiously, in contrast to wild-type was also shown in D. melanogaster 20. Flies that carry
Crepuscular flies, in mutant per 0 flies and tim01 flies — which have a luciferase reporter downstream of the per gene pro-
Describes an organism that is
active during twilight or during
a mutation in the gene encoding Timeless (TIM), the moter (per-luc) have a disrupted and reduced amplitude
day-to-night or night-to-day binding partner of PER proteins — there is no time-of- of circadian transcriptional activity in a CREB-mutant
transitions. day effect on short-term olfactory avoidance memory background, indicating a functional link between CREB
under DD conditions18. In addition, as measured in a activity and circadian gene expression in D. mela-
Eclosion rhythm
courtship conditioning assay, per 0 flies are defective in nogaster. In addition, per expression affects the cycling
The timing of the emergence
of the adult fly from its pupal
long-term memory (LTM) formation — a phenotype of CRE-mediated activity. Flies that carry a luciferase
case, which usually occurs that can be rescued with a wild-type copy of the per gene reporter downstream of three CRE sites (CRE-luc) nor-
at dawn. in the per 0 background19. Overexpression of per in this mally show a circadian rhythm of luminescence under
paradigm has even been shown to enhance LTM19 despite conditions of 12-hour light followed by 12-hour dark
Clock gene
A gene that regulates aspects
these flies retaining rhythmic locomotor and mating (LD) as well as under DD conditions. This CRE-luc
of circadian rhythms.
D. melanogaster M. musculus
CREB
P P
? CREB
CKII CKIε
DBT CLK CLOCK
per CYC E-box CRE CRE E-box BMAL Per1, Per2
P CLK CLOCK P
PER tim CYC E-box E-box BMAL Cry1, Cry2 PERs
TIM CRYs
MAPK PER CLK CLOCK PERs MAPK
TIM CYC BMAL CRYs
PKA PKA
Ca2+ Ca2+
cAMP cAMP
Nucleus
? P P
MEL AC ATP CLK Cytoplasm BMAL ATP AC MEL
? CYC CLOCK
Figure 1 | Phylogenetic conservation of the core molecular clock. The molecular clock in flies and mammals is
Nature Reviews | Neuroscience
composed of transcriptional and translational feedback networks. In flies, CLOCK (CLK) and CYCLE (CYC) heterodimerize
and activate transcription of the period (per) and timeless (tim) genes by binding to E-box elements in their promoters. The
protein products PER and TIM heterodimerize and enter the nucleus following phosphorylation (P) by proteins such as
doubletime (DBT) or casein kinase II (CKII), and repress the transcriptional activity of CLK–CYC. In mammals, the circadian
clock comprises a similar feedback network, including CLOCK and the CYC homologue brain and muscle ARNT-like
(BMAL), which activate the transcription of per and cryptochrome (cry) genes via E-box elements. PERs and CRYs
heterodimerize in the cytoplasm following phosphorylation by proteins such as casein kinase Iε (CKIε), and enter the
nucleus where they inhibit CLOCK–BMAL transcriptional activation. Mitogen-activated protein kinase (MAPK)
phosphorylates BMAL146, repressing BMAL–CLOCK activity. Putative mechanisms linking melatonin (MEL) rhythms and
the circadian clock include repression of adenylate cyclase (AC) and protein kinase A (PKA), a pathway known to influence
cAMP-responsive element (CRE)-binding protein (CREB) activation. A second mechanism is thought to activate the
MAPK–CREB cascade147,148 through Ca2+ influx, leading to transcriptional activity through CRE elements in the per
promoters22. An analogous putative pathway is shown for Drosophila melanogaster, in which MAPK phosphorylation
represses CLK–CYC149 or activates CREB, leading to per transcription.
578 | AuGusT 2010 | vOLuME 11 www.nature.com/reviews/neuro
© 2010 Macmillan Publishers Limited. All rights reserved

3. REVIEWS
a D. melanogaster OC DN2
CREB regulates normal circadian behaviour in flies20.
PI These data support a reciprocal relationship between
DN1 LPN
LNd
CREB- and PER-mediated transcriptional regulation,
DN3 with functional relationships in the generation of cir-
Large LNv ?
? cadian rhythms. The precise relationship between
CREB- and PER-mediated transcriptional activity (for
example, through the cAMP–MAPK–CREB cascade)
MB
in the time-of-day-dependent regulation of memory is
still unclear.
POT
Anatomical relay of circadian centres
It is well established that certain molecules with cycling
H–B
activity patterns influence circadian rhythms, but what
is known about how neural networks generate the ulti-
OL Small LNv mate behavioural output? Considerable progress has
5th small LNv been made in elucidating the cellular components and
neuronal pathways that are responsible for the genera-
b M. Musculus Hippocampus Pineal gland tion of circadian rhythms, and some similarities have
been found across many species. Across phylogeny,
clock-containing circadian pacemaker cells in the cen-
OB tral nervous system receive photic input and can drive
changes in locomotor rhythms over the course of the
day. For example, in mammals, photic activation of
non-image-forming retinal ganglion cells, which con-
tain the photo-responsive pigment melanopsin, send
light information to the central pacemaker of circadian
rhythms — the suprachiasmatic nucleus (sCN)23 — via the
PVN SCG
LH retinohypothalamic tract (RHT)24. The core of the sCN
VLPO
receives photic input from the RHT and relays it to the
TMN subparaventricular zone (sPvz), which in turn relays
RHT sPVz DMH the information to other hypothalamic structures (FIG. 2).
SCN These hypothalamic structures are known to regulate
Figure 2 | Anatomical circadian pathways in flies and mice. a | Reviews | Neuroscience
Nature In fruitflies many physiological processes, including thermoregula-
(Drosophila melanogaster), various light-receiving cells are involved in functional tion, hormone secretion, feeding behaviour and arousal–
neuroanatomical connections, such as those in the Hofbauer–Buchner (H–B) eyelets and sleep states. The sCN is therefore thought to regulate
ocelli (OC), or from the optic lobes (OL). These project to circadian pacemaker cells, the the circadian timing of these processes. Pathways con-
lateral neurons (LN), via the posterior optic tract (POT). LN subtypes include the large, necting the sCN to limbic structures that are involved
small, and 5th small ventral LN (LNv), as well as the dorsal LN (LNd). Little is known about
in memory processing, such as the hippocampus and
the functional connectivity between these pacemaker cells and other clock cells, such as
the dorsal neurons (DN1, DN2 and DN3 subtypes) the lateral posterior neurons (LPN) or
amygdala, have been shown25,26. Other indirect connec-
cells that are involved in sleep and memory formation, such as the pars intercerebralis (PI) tions — such as through hypocretin-expressing cells in
and mushroom bodies (MB). DNs and LNs comprise the ~150 cells of the clock network in the lateral hypothalamus27 or through superior cervical
the fly brain. b | In the mouse (Mus musculus), the suprachiasmatic nucleus (SCN) receives ganglion-stimulated melatonin release from the pineal
photic input through the retinohypothalamic tract (RHT). The SCN projects to the dorsal gland28 — could relay sCN-derived circadian input to
medial hypothalamus (DMH) through the subparaventricular zone (sPVz), which projects the hippocampus (FIG. 2). Whether these connections are
to various regions in the hypothalamus, including the ventrolateral preoptic area (VLPO), responsible for the time-of-day-dependent expression of
the lateral hypothalamus (LH) and the paraventricular nucleus (PVN). There are reciprocal memory and/or synaptic plasticity is not known.
connections between the VLPO and the tuberomammilary nucleus (TMN), which are In D. melanogaster, photic input entrains a circadian
thought to be partly responsible for the proper timing of sleep–wake rhythms. Functional
rhythm in circadian pacemaker cells (of which there are
connections between a circadian centre and a memory forming-centre, such as the
hippocampus, are not well known. They may be partially gated through hypocretin- or
~150) through at least three pathways29: the eyes30, the
orexin-expressing cells of the LH, or by melatonin secretion from the pineal gland Hofbauer–Buchner eyelets31–33 and/or the blue-light photo
following signalling from the PVN to the superior cervical ganglion (SCG)150. Note that pigment cryptochrome (CRY)30,34. Photoreceptive cells
C57BL/6J mice lack melatonin. OB, olfactory bulb. in the optic lobe are thought to project to the lateral neu-
ronal cells via the posterior optic tract 35 (POT) (FIG. 2).
This has been supported by recent findings that describe
cycling is coordinately altered in per mutants: in perL functional connectivity between the contralateral optic
flies, the CRE-luc cycling pattern is lengthened, whereas lobe and the large ventral lateral neurons (LNv) via the
it is shortened in pers flies, compared with wild-type POT36. Additionally, the Hofbauer–Buchner eyelets send
Suprachiasmatic nucleus controls, and in per0 flies the CRE-luc activity pattern projections to the accessory medulla37, where they syn-
A hypothalamic bilateral
structure that is the central
is arrhythmic across the day 20. A mutation in the Creb2 apse with the small LNv pacemaker clock neurons38,39.
pacemaker of circadian gene of D. melanogaster also produces a shortened Anatomical connections among pacemaker cells that
rhythms in mammals. circadian cycle of locomotor activity, suggesting that receive and relay this photic input to other brain regions
NATuRE REvIEWs | NeuroscieNce vOLuME 11 | AuGusT 2010 | 579
© 2010 Macmillan Publishers Limited. All rights reserved

4. REVIEWS
— such as the mushroom bodies, a pair of neuropil events downstream of central pacemaker cells, such as
structures in the insect brain known to regulate sleep40,41 stimulation of brain regions that are involved in learning
and memory formation42,43 — in D. melanogaster are not and memory.
well characterized. Although the small LNv cells have Melatonin is a signalling molecule that is widely
been shown to project to terminals near the mushroom expressed throughout phylogeny and is secreted in a
bodies44, a direct functional connection between pace- time-of-day-dependent manner 53. Recently, melatonin
maker cells and cells that are responsible for memory synthesis has been suggested to interact with core circa-
formation is still lacking. The genetically tractable model dian mechanisms54. The temporal release of melatonin
of D. melanogaster will be valuable in determining the is regulated over the 24-hour day in humans55 and many
neural circuitry that is responsible for the integration of other species, including zebrafish56 (Danio rerio), sea
these two complex processes. slugs57 (Apylsia californica), mice58 (Mus musculus) and
flies59 (D. melanogaster). Interestingly, melatonin affects
Time-of-day effects on neurophysiology the firing rate of the mammalian sCN60–62, suggesting
It is unclear whether a time-of-day-dependent relay that the time-of-day-dependent regulation of hormo-
exists between clock pacemaker neurons and memory- nal secretion may alter the firing patterns of circadian
forming cells, such as between the sCN and the hippoc- pacemaker cells. Melatonin also affects the firing rate of
ampus of mammals or between the lateral neurons and hippocampal CA1 neurons63. Therefore, circadian hor-
the mushroom bodies of flies. However, cyclical changes monal modulation of neuronal firing could be a general
over 24-hour periods have been observed in baseline mechanism throughout the brain. It would be interest-
physiological properties of central pacemaker cells in ing to know whether there are circadian fluctuations in
both mammals and flies. In nocturnal rodents, sCN the baseline sFRs and RMPs in hippocampal or mush-
neurons show circadian changes in spontaneous fir- room body neurons, and whether these oscillations
ing rate (sFR) and resting membrane potential (RMP), occur in phase with those of the sCN and large LNvs,
with an elevated sFR and more depolarized RMP in the respectively. If these circadian firing patterns occur, an
light phase than the dark phase45–50 (FIG. 3). A similar important next step will be to address whether they
effect was observed in the large LNv pacemaker cells result from functional connectivity between pacemaker
in crepuscular D. melanogaster 36,51,52. Taken together, cells and memory-encoding regions, or whether they
these data further support a phylogenetically conserved are caused by autonomous cycling molecules or cir-
mechanism of circadian neurophysiology in pacemaker culating hormones such as melatonin. Further stud-
cells (FIG. 3). This conserved mechanism may influence ies are needed to examine the functional connectivity
time-of-day-dependent expression of physiological between these brain regions that are involved in circa-
dian rhythms and memory, and to compare baseline
neurophysiological properties.
Fly large LNv Rodent SCN
Time-of-day effects on synaptic plasticity
Given that circadian changes in molecular and neuro-
physiological properties of pacemaker cells are observed
2 10 throughout the animal kingdom, an obvious question
is whether neural correlates of plasticity-related neuro-
SFR (Hz)
physiology are also regulated by circadian rhythms. In the
hippocampus, long-term potentiation (LTP) — a form
0 0 of synaptic plasticity (BOX 1) that is thought to underlie
–40 –50 learning and memory 64 — has been shown to change
depending on the time of day in various nocturnal
RMP (mV)
rodents65–70 (FIG. 4). These circadian effects on LTP can be
considered a naturally occurring form of metaplasticity71,
–60 –60 in that the synaptic efficacy for a given amount of stimu-
lation varies based on the time of day. Circadian changes
Figure 3 | Time-of-day-dependent neurophysiology in in plasticity may serve as a useful model with which to
Nature Reviews | Neuroscience study the neurophysiological and molecular mechanisms
pacemaker cells. Fruitflies, which are crepuscular, have a
higher spontaneous firing rate (SFR) and resting membrane of metaplasticity. It would also be interesting to compare
Melatonin potential (RMP) in their clock cells, such as the large ventral plasticity-related mechanisms in the sCN with those in
A catecholamine hormone lateral neurons (LNv), near the dark-to-light transition and the hippocampus. Molecular pathways (FIG. 1) that are
derived from serotonin. during the daytime, than in the light-to-dark transition involved in establishing plastic changes in circadian
and during the night-time36,52. Similar to fly clock cells, neurophysiology may be more fundamental than previ-
Hofbauer–Buchner eyelets neurons in the nocturnal rodent suprachiasmatic nucleus
Photoreceptor cells that are ously appreciated and could be shared with other known
(SCN) have an elevation in SFR and RMP during the light
located between the retina and
period compared with the dark period, despite a difference plasticity-related brain regions, such as those involved in
the lamina of the fly eye. memory formation. Lastly, knowledge of how baseline
in the locomotor activity rhythm of these two species49,50.
Metaplasticity Top traces show a schematic representation of spike trace conditions can be modulated by the time of day is cru-
Alterations in the ability of the frequency over the day–night cycle. White bars represent cial for understanding the effect on synaptic plasticity of
synapse to change in strength. ‘lights on’; dark green bars represent ‘lights off’. other experimental manipulations.
580 | AuGusT 2010 | vOLuME 11 www.nature.com/reviews/neuro
© 2010 Macmillan Publishers Limited. All rights reserved

5. REVIEWS
Box 1 | Physiological analysis of long-term potentiation
Two examples of the methods used to study the physiological manifestations of long-term potentiation (LTP) are shown in
the figure. Rodent brain slices are obtained at the level of the suprachiasmatic nucleus (SCN; a), or the hippocampus (b).
Following stimulation (STIM) of the optic tract or the Schaffer collateral pathway, recordings are made from electrodes
(REC) that are implanted in cells of the SCN or area CA1 of the hippocampus, respectively. Example graphs depicting
representative traces of recordings following stimulation show LTP in the form of a population spike or a field potential.
Example traces (inset) are shown for the pre-stimulation baseline (shown by the orange line) and post-stimulation LTP
(shown by the green line). LTP is expressed as a percentage increase of the slope or amplitude of electrical potential from
the baseline over time.
ab REC Population spike
300
STIM
1 mV
20 ms
% Baseline
b
a 100
Hippocampus STIM
0 1h
Field potential
REC 250
STIM STIM
% Baseline
100
1 mV
10 ms
SCN 0
0 1h
Nature Reviews | Neuroscience
Synaptic plasticity in the suprachiasmatic nucleus. Time- the light phase in area CA1 (FIG. 4). A clear influence
of-day-dependent changes in synaptic plasticity have of the time of day on LTP in the hippocampal CA1
been observed in the sCN. stimulation of the optic nerve region was also observed in the syrian hamster, with
can elicit potentiated responses in the sCN that last for increased LTP in animals that were killed during the
hours, analogous to those observed following schaffer light phase than during the dark phase73. However,
collateral stimulation of the CA1 region of the hippo- unlike the rat studies analysed previously 67, this study
campus (BOX 1). In the sCN of rats, following stimulation used tissue that was harvested during the opposite
of the optic nerve and with changes recorded in field time of day from when the electrophysiology was
potentials at six time points over the course of the day completed. ‘Daytime’ hippocampal slices were pre-
(starting 1 hour after normal lights-on), a potentiation of pared between zeitgeber time (ZT) phase 4.5–5.5, but
synaptic strength was observed during the day phase LTP was not evaluated until the night-time (between
of the circadian cycle69 (FIG. 4). However, this study was ZT13.5–19.5). Conversely, ‘night-time’ hippocampal
unable to detect the previously reported time-of-day- slices were prepared between ZT18.5–19.5, but LTP
dependent changes in LTP in the hippocampus67, which was not evaluated until ZT4. LTP in the CA1 was
raises questions about how time-of-day effects on LTP greater in hamster tissue that was harvested during
may be regulated. the light period and then tested later during the dark
period than in hamster tissue that was harvested and
Inhibitory avoidance Hippocampal long-term potentiation. Hippocampus- tested in the reverse conditions73 (FIG. 4). This prompted
conditioning
A form of learning in which an
dependent tasks, such as inhibitory avoidance conditioning, examination of whether the time-of-day effects on
animal learns to avoid a are thought to elicit LTP-like responses following train- LTP are dependent on the time of tissue harvest or
stimulus (for example, a ing 72, suggesting a functional relationship between the time of testing 70. In hippocampal CA1 tissue har-
darkened compartment) that synaptic plasticity mechanisms and memory forma- vested from nocturnal C3H and C57Bl/6J strains of
delivers a shock.
tion. In 1977, it was shown that synaptic excitability mice during the light period, LTP was greater in tissue
Zeitgeber time of the rat hippocampal dentate gyrus exhibited a cir- that was examined during the dark period than during
(ZT). Standardized notation for cadian rhythm65. Previously, a meta-analysis examined the light period70 (FIG. 4). These data are in contrast
the time during an entrained over 170 studies for a circadian component of hippoc- to the meta-analysis of nocturnal rat LTP67, but are con-
circadian cycle. ZT0 is the start ampal LTP in rats, and found that the incidence and sistent with data on nocturnal hamsters73, and together
of the light phase and ZT12 is
the beginning of the dark
the magnitude of LTP were dependent on the time of suggest that the time-of-day effects on LTP are depend-
phase, during a 24-hour day 67. Interestingly, LTP was elevated during the dark ent on the time of testing, and not the time of harvest.
light–dark cycle. phase in the dentate gyrus but was elevated during Furthermore, these results support the hypothesis that
NATuRE REvIEWs | NeuroscieNce vOLuME 11 | AuGusT 2010 | 581
© 2010 Macmillan Publishers Limited. All rights reserved

6. REVIEWS
a Rat hippocampal LTP Rat SCN LTP formation in rats78, suggesting a broad action of mela-
300 300 tonin on synaptic plasticity in other brain regions. These
Light phase data are interesting given the influence of melatonin on
Dark phase memory formation 79 and the timing of melatonin
% Baseline
peaks in the circadian rhythm of memory formation
in various species (FIG. 5). In addition, studies in adrena-
lectomized (ADX) rats have implicated hormonal mod-
ulation in the time-of-day effect on hippocampal LTP68.
100
DG CA1
100 In ADX rats, the peak of LTP in the dentate gyrus
shifted from the night-time to the daytime, suggesting
b Hamster hippocampal LTP that hippocampal LTP is regulated by other circulating
300
Time of death
500 hormones, such as those found in the adrenal gland.
As the circadian regulation of adrenal corticosterone
% Baseline
Light > dark is under the influence of the sCN80, the altered cycling
Dark > light of hippocampal LTP that is observed in ADX rats may
result from a disruption in signalling from the sCN
100 100
CA1 0 30 to the hippocampus via the adrenal gland. Given that
the time of testing, rather than the time of harvest,
b Mouse (C3H strain) hippocampal LTP seems to affect hippocampal LTP, it is plausible that
500 500
cycling hormones affect the ‘clock-time’ of hippocam-
pal cells before tissue harvest. These data suggest that
% Baseline
Dark Light > dark circulating hormones could operate as a zeitgeber, set-
Light ting the clock in hippocampal cells such that they retain
100 100 the rhythm during testing that was previously entrained
0 60 0 60 by hormone signalling. Whether there is a relation-
Figure 4 | Time-of-day effects on synaptic plasticity. ship between rhythms of LTP and rhythms of memory
Nature Reviews | Neuroscience
Time-of-day-dependent changes in long-term formation remains to be determined.
potentiation (LTP) occur in various rodent species,
including rats, hamsters and mice. a | Interestingly, in the Circadian effects on memory formation
nocturnal rat hippocampus, LTP is greater during the dark Time-of-day and circadian effects on cognitive per-
phase than in the light phase in the dentate gyrus (DG) formance and memory formation have been observed
but is greater during the light phase than in the dark in various behavioural paradigms81–84. The influence of
phase in area CA1 (ReF. 67). LTP is also time-of-day- sleep and behavioural state on memory consolidation
dependent in the rat suprachiasmatic nucleus (SCN), with
are reviewed elsewhere85–87.
an elevation during the day phase compared to the night
phase69. b | In hippocampal area CA1 of the hamster, LTP is
Interestingly, circadian rhythms of locomotor activ-
higher when tissue is isolated during the light phase and ity are not an accurate predictor of the timing of optimal
tested during the dark phase (light > dark) and lower performance in memory tasks over the course of the
when tissue is isolated during the dark phase and tested day. The circadian rhythms in memory formation in
during the light phase (dark > light)73. c | In area CA1 of nocturnal, diurnal and crepuscular species do not show
the nocturnal mouse, LTP is lower during the light period, a clear correlation with their specific rhythm of locomo-
but enhancement of LTP (metaplasticity) occurs when tor activity. For example, peak performance in memory
tissue is harvested during the light period and tested tasks occurs during the ‘active phase’ of the diurnal
during the dark period70. Taken together, these data zebrafish79 (D. rerio) and sea slug 18,88 (A. californica),
suggest that there are endogenous cellular oscillators
but during the ‘inactive phase’ of the nocturnal house
that continue to function when dissociated from the
intact brain, driving time-of-day-dependent changes in
mouse89 (M. musculus) and the crepuscular fruitfly 90
synaptic plasticity. (D. melanogaster) (FIG. 5). This suggests that activity lev-
els are not responsible for the changes in memory forma-
tion that occur over the course of 24 hours, as there is a
an independent circadian pacemaker controls the time- clear dissociation of the rest–activity rhythm and cogni-
of-day-dependent changes in hippocampal plasticity tive performance in the various chronobiological mod-
and that arousal state or sleep per se are not necessary els. This indicates that factors other than the behavioural
for circadian changes in LTP. state are involved, supporting the theory that specific
cellular or molecular events that cycle over the 24-hour
Hormonal regulation of long-term potentia- day are responsible for the enhancement of memory at
tion. Hormones have been shown to modulate LTP in specific times. Processes that could contribute to this
the sCN and hippocampus. For example, melatonin has memory enhancement include gene transcription and
Epigenetic mechanism been shown to block LTP in both regions74–77. C57Bl/6J translation, epigenetic mechanisms, neurotransmitter
A process that alters the state mice, which lack melatonin, undergo less dramatic release, synaptic excitability, neuronal activity and hor-
of gene expression through
changes in chromatin structure
time-of-day-dependent changes in hippo campal LTP mone secretion. In addition, each of these processes may
(that is, DNA or histone than C3H mice, which express melatonin70. Melatonin have different effects on various phases (acquisition,
modifications). also inhibits neocortex-dependent LTP and memory consolidation and retrieval) of memory.
582 | AuGusT 2010 | vOLuME 11 www.nature.com/reviews/neuro
© 2010 Macmillan Publishers Limited. All rights reserved

7. REVIEWS
a ‘Active-phase’ enhancement b ‘Inactive-phase’ enhancement
D. rerio A. californica M. musculus D. melanogaster
Activity
Activity
Activity
Activity
Memory
Memory
Memory
Memory
Melatonin
Melatonin
Melatonin
Melatonin
Figure 5 | Melatonin and circadian rhythms of memory. Low levels of melatonin correlate with high levels of memory
Nature Reviews | Neuroscience
performance and are seemingly independent of the activity state in vertebrate and invertebrate species. In the diurnal
zebrafish (Danio rerio) and sea slug (Aplysia californica), there is a time-of-day-dependent enhancement of memory
formation during the ‘active phase’ (a) of their circadian rhythm. By constrast, in the nocturnal mouse (Mus musculus) and
the crepuscular fruitfly (Drosophila melanogaster), there is memory enhancement during the ‘inactive phase’ (b) of their
circadian rhythm. Each species, regardless of activity state, has a corresponding anti-phase relationship between memory
performance and melatonin levels. White bars represent ‘lights on’; dark green bars represent ‘lights off’. Figure is based
on data from ReFS 56,57,59,79,88–90,93.
Does melatonin influence memory? A better predictor of cascades influence the time-of-day effects on memory
the circadian variation of peak performance in memory formation. However, the melatonin theory is contro-
tasks may be cycling molecules with a periodicity that fol- versial as C57Bl/6J mice exhibit time-of-day-dependent
lows the rhythm of performance — rather than the rhythm changes in memory formation89 despite the absence of
of locomotor activity — across species. One molecular melatonin58,92,93, and so further studies are warranted.
correlate that seems to be a good candidate is melatonin.
Interestingly, levels of melatonin are inversely related to Time of acquisition versus time of recall. Differences
cognitive function: peak melatonin release occurs during in time-of-day effects on memory acquisition versus
the lower periods of performance over the course of the recall have been observed on various tasks in various
day in humans8,91 and other species (FIG. 5). This suggests models79,88,89,94–96. In A. californica, the circadian rhythm of
that levels of melatonin are good predictors of the nadir long-term sensitization (LTs) was examined88. LTs train-
(lowest) period in the time-of-day cycling of memory. ing consists of a series of shocks delivered to the side of
Recently, the role of melatonin in the regulation of the sea slug, which elicits a siphon withdrawal response,
memory was investigated using an active-avoidance followed by a post-training electrical shock to the tail,
conditioning paradigm in zebrafish79. Fish were trained which elicits a ‘sensitized’ siphon withdrawal. Memory
in a tank to make an ‘unsafe’ association in a dark com- is expressed as the ratio between pre-training and post-
partment, in which they received electric shocks, and training durations of the siphon withdrawal behaviour.
a ‘safe’ association within a lit compartment, in which A time-of-day effect on LTs memory was observed, with
there were no electric shocks. A clear time-of-day effect peak responses during the ‘active period’ in both LD and
was observed in acquisition (learning) and memory DD conditions (FIG. 5). Time-of-day-dependent regula-
formation, and both were improved during the daytime tion of the baseline siphon withdrawal response or the
(active) period. These effects were maintained under DD withdrawal duration was not observed in either LD or
— a condition that is necessary to evaluate whether an DD conditions, suggesting that the memory results from
endogenous circadian system controls the time-of-day an endogenous circadian mechanism. This time-of-day
effect. Treatment of fish with melatonin before training effect on LTs depended on the time of acquisition, rather
had no effect on acquisition but significantly reduced than on the time of recall. The LTs response was greater
LTM formation. Fish that were treated with melatonin, in animals that were trained at circadian time 9 (CT9;
either following training or just before testing, did not when LTs is enhanced) and tested at CT21 (when LTs is
show differences in LTM or retrieval, respectively. This suppressed) than in animals that were trained and tested
Circadian time finding supports a role for melatonin in the earlier at CT21. Animals that were trained at CT21 and tested at
(CT). Standardized notation for stages of LTM formation. Furthermore, this phenotype CT3 (when LTs is normally enhanced) did not show an
an organism’s relative was rescued by treating the fish with melatonin recep- elevated LTs response, suggesting that the time of training
(subjective) time. CT0 is the tor antagonists, either simultaneously with melatonin (learning) and not the time of testing (recall) is responsi-
start of subjective daytime and
CT12 is the start of subjective
during the daytime or alone during the night-time ble for the circadian rhythm of LTs in A. californica. This
night-time, under constant phase (when endogenous melatonin is high and LTM is in contrast to the result in mice89, which suggested that
dark conditions, over 24 hours. is low). These results suggest that melatonin signalling optimal recall times are under circadian control.
NATuRE REvIEWs | NeuroscieNce vOLuME 11 | AuGusT 2010 | 583
© 2010 Macmillan Publishers Limited. All rights reserved

8. REVIEWS
The mouse inbred strains C3H and C57Bl/6J have the formation of 4-day-long memory. This effect does
significant time-of-day effects on memory forma- not occur when CREB2A is stimulated in the middle
tion in a fear conditioning paradigm89. They display of the daytime (before training), suggesting a time-sen-
optimal memory recall during the ‘inactive phase’ sitive window when the encoding of CREB-mediated
(the light period) for contextual and cued fear con- enhancement can occur. Additionally, it was shown that
ditioning. These mice also show a time-of-day effect time-of-day-dependent cycling of various components
on memory acquisition, with an elevation during the of the cAMP cascade is necessary for LTM formation
light period, at ZT3, compared with the dark period, in mice96. Circadian cycling of cAMP and MAPK phos-
at ZT15. A possible relationship between the effect of phorylation paralleled time-of-day-dependent oscilla-
time of acquisition and the effect of the time of recall tions in RAs activity and the phosphorylation of MAPK
was also examined. Mice were trained during the day kinase and CREB in the hippocampus. Disruption of the
period, at ZT3, under LD conditions. They showed circadian rhythm of this cAMP–MAPK–CREB cascade
circadian cycling of memory that was maintained for in the hippocampus, by pharmacological approaches
at least 3 days, with a period of ~24 hours and a peak or by exposure of animals to constant light conditions,
enhancement reoccurring at the same time as train- impaired memory formation.
ing. Interestingly, in DD conditions with training at In A. californica, although baseline activity of MAPK
CT3, mice displayed the same ~24-hour periodicity of does not follow a circadian rhythm in the pleural gan-
peak memory, reoccurring at CT3 for 3 consecutive glia, LTs training during different times of day produces
days89. This suggests that the peak time of acquisition different levels of activated MAPK101. For example, phos-
may coincide with the peak time of recall. To test this, phorylation of MAPK increases following LTs training
mice were trained during the time of poorer acquisition during the daytime as compared with the night-time,
(ZT15), and examined for effects on recall. The circa- and this phosphorylation correlates with patterns of
dian cycling of the peak time of recall (ZT3) was pre- LTM enhancement. This poor performance during the
served to the third day of testing. To determine whether night-time following LTs training can be rescued using
this effect was truly circadian, animals were trained in compounds that activate MAPK activity or MAPK-
DD conditions at CT15 and, surprisingly, still showed dependent transcription. It therefore seems that, as in
memory enhancement at CT3 on subsequent days of mice96, the circadian clock is also able to modulate LTM
testing. This suggests that the peak time of memory formation in A. californica through the MAPK cascade.
recall is under time-of-day-dependent control, that it Taken together, these data strongly suggest that cAMP–
is independent of the time of training and is regulated MAPK–CREB is a phylogenetically conserved pathway
by an endogenous circadian system. for the time-of-day dependence of memory.
However, a time-of-day effect on recall not was
observed in a subsequent study using fear conditioning Epigenetic factors. Epigenetic factors provide an
in C57Bl/6J mice96. specifically, there was no elevation in interesting link between the molecular mechanisms
recall at ZT4 when animals were trained at ZT16 (ReF. 96). that underlie circadian rhythms and the formation of
However, this study had lower temporal resolution memory. In the mature nervous system these factors
than the study described above89, which might account influence changes in synaptic plasticity and complex
for the apparent conflict of results. It is also possible that behaviour, such as drug addiction, memory and cir-
variation in the strength of the training protocol could cadian rhythms102. Changes in the epigenetic state of
skew results, obscuring the circadian effects on memory specific cells may be a principal mechanism by which
processes. In addition, when animals are tested repeat- altered gene expression exerts a circadian influence on
edly 89, testing would include combined effects of recall, memory formation. Chromatin remodelling follow-
reconsolidation and extinction, making it difficult to ing stimulation occurs in hippocampal cells103 and is
discern the circadian influence on solely the recall stage induced by light in sCN cells104. It is therefore possible
of memory. Further studies examining the role of sev- that epigenetics is a common mechanism relaying the
eral memory stages, such as acquisition versus recall, cyclical change in various memory-related processes.
are required to determine the relative contributions Rhythmic changes in chromatin states are altered in
of these factors to the time-of-day effects on memory a circadian manner 105–107. Furthermore, the protein
formation. product of the circadian gene CLOCK itself has histone
acetyltransferase activity 108 and has recently been shown
Molecular oscillators and memory — with its binding partner brain and muscle ARNT-
The cAMP–MAPK–CREB pathway. The cAMP like (BMAL) (FIG. 1) — to regulate LTM formation109.
Fear conditioning signalling cascade has a central role in memory Memory formation elicits histone modifications110,111.
A form of learning in which fear
formation97. For example, in D. melanogaster, overex- It is possible that specific molecular pathways, such
is associated with a neutral
stimulus, by pairing the pression of a repressor isoform of CREB (CREB2B) as the cAMP–MAPK–CREB cascade, may stimulate
neutral stimulus with an selectively abolishes LTM formation without affect- these epigenetic changes that ultimately drive LTM
aversive stimulus. In contrast ing short-term memory 98. By contrast, overexpression through circadian gene expression. Future study of the
to inhibitory avoidance of an activator isoform of CREB (CREB2A) enhances relationships between these molecular pathways and
conditioning, the animal
cannot choose to avoid the
LTM99, and this is dependent on the time of day 100. their influence on epigenetic mechanisms may provide
conditioned stimulus upon When stimulated near the light–dark transition (before new insight into the circadian regulation of memory
testing. training), conditional expression of CREB2A enhances persistence.
584 | AuGusT 2010 | vOLuME 11 www.nature.com/reviews/neuro
© 2010 Macmillan Publishers Limited. All rights reserved

9. REVIEWS
Directed Autonomous Integrated Conclusions and perspectives
The conservation of a time-of-day effect on memory
in many species points towards a common molecular
mechanism. This is likely to involve the cAMP–MAPK–
CREB cascade, which is implicated in the generation of
memory and circadian rhythm processes and is phy-
Memory circuit Memory circuit Memory circuit logenetically conserved. Given the circadian influence
on the regulation of memory-related molecules, neuro-
physiology, synaptic plasticity and behaviour, it is crucial
that neuroscientists consider the effect of time-of-day-
dependent variation on the experimental design, analysis
and interpretation of future studies. Further study is
needed to examine the signalling pathways involving
molecules that have been studied for roles in memory
and circadian rhythms, such as fragile X protein. This
protein has been shown to regulate circadian rhythms
Central pacemaker Central pacemaker Central pacemaker and memory in flies 118–121 and mice 122–124. In addi-
tion, levels of vasoactive intestinal peptide (vIP) cycle
Figure 6 | Models for the circadian regulation of memory. Cellular and molecular in the rodent sCN and contribute to the signalling of
correlates of circadian cycling exist in various regions of the brain. Such regions include
Nature Reviews | Neuroscience the cAMP–MAPK–CREB cascade23,125. Interestingly,
the mammalian suprachiasmatic nucleus (the central clock), as well as other cell
vIP knockout mice, which lack a circadian locomotor
populations, including those in the mammalian hippocampus (the memory centre). The
degree to which the central clock is responsible for directing oscillations of the memory rhythm, retain time-of-day effects on memory, albeit to
circuit, versus autonomous control by the memory cells themselves, remains to be a lesser extent than wild-type mice126. Further research
determined. An integrated model seems most likely, whereby peripheral oscillators have is needed to determine whether these pleiotropic mole-
some control that is coordinated with inputs (either directly or indirectly) from a central cules exert circadian effects on memory through known
pacemaker. Orange circles represent central pacemaker-driven oscillators; blue circles core clock mechanisms or through secondary effects
represent autonomous oscillators. Solid arrows indicate direct control; dashed arrows outside of the known circadian pathways.
indicate an influence on cellular oscillations. What is the influence of core clock-controlled pace-
maker cells on the neuronal networks that are respon-
sible for memory formation? As circadian clocks
Central versus cell-autonomous oscillators are autonomous in many tissues127, it is possible that
Which cells are responsible for driving the circadian reg- cell-autonomous clocks in different regions of the
ulation of memory? Is a central pacemaker required, or brain independently control the circadian timing of
do cell-autonomous oscillators in the memory-encoding neurophysiology and memory-related processes. For
neurons themselves regulate the time-of-day effects on example, in A. californica, time-of-day effects on LTs
memory (FIG. 6)? Although sCN ablation inhibits cycling persist in the absence of an ocular circadian oscillator 128,
of Per2 in the amygdala and hippocampus of hamsters112, and in D. melanogaster, circadian rhythms in the olfactory
it does not prevent expression of the time-stamp memory system are autonomous from lateral neuron pacemaker
in a conditioned place-avoidance task113. It is possible cells129. similarly, in rodents, ablation of sCN pacemaker cells
that the time stamp could be partially encoded by mem- does not affect circadian oscillations in gene expres-
ory cells using the same molecular machinery that is sion in the olfactory bulb130,131 and daily oscillations of
involved in maintaining circadian rhythms within pace- gene expression in brain regions outside of the sCN are
maker cells. Furthermore, mouse knockouts of neuronal anti-phase to those in the sCN130,132. Interestingly, Per2
PAs domain protein 2 (NPAs2), which is not expressed expression oscillates in isolated hippocampal tissue133,
in the sCN but is a binding partner of the essential clock suggesting that circadian gene expression in these extra-
protein BMAL, have LTM deficits in cued and contex- sCN regions may be autonomous. Further studies that
tual fear conditioning 114. However time–place memory analyse the contribution of endogenous oscillators in
— a form of learning in which an association is formed memory-forming brain regions (versus central pacemaker
between a specific location and the time of day — still cells) to the time-of-day effects on memory are vital for
requires the Cry genes in mice115 (FIG. 1). Together, these our understanding of normal brain physiology. For exam-
data suggest that the cycling of specific components of ple, does the learning event itself serve as a zeitgeber in
the core molecular oscillatory pathway is required, per- memory-encoding regions, setting a memory ‘clock’ of its
haps in the sCN or extra-sCN regions, for the expres- own such that the memory is better retrieved at certain
Time stamp
sion of time-stamp memory. In addition, rhythms of times of day, as time-stamp conditioning suggests? Are
The time of day that produces
optimal performance in a passive avoidance conditioning in rats5,116 require an the epigenetic mechanisms that are involved downstream
memory task and is associated intact sCN117, supporting a role for central circadian of the light zeitgeber in the clock pacemaker cells (for
with the memory. pacemaker cells in the regulation of time-of-day effects example, the sCN) analogous to the epigenetic mecha-
on memory. Further work is therefore needed to deter- nisms that produce an engram in memory centres such
Engram
A hypothetical representation
mine how circadian transcriptional mechanisms in the as the hippocampus? studies of functional connectivity
of the physiological storage of core pacemaker (the sCN), versus the peripheral cells between clock cells and learning and memory centres
memory. (extra-sCN), contribute to the persistence of LTM. over circadian time following training, consolidation and
NATuRE REvIEWs | NeuroscieNce vOLuME 11 | AuGusT 2010 | 585
© 2010 Macmillan Publishers Limited. All rights reserved