Download Insect olfactory memory in time and space

Insect olfactory memory in time and space
Xu Liu1 and Ronald L Davis1,2
Recent studies using functional optical imaging have revealed
that cellular memory traces form in different areas of the insect
brain after olfactory classical conditioning. These traces are
revealed as increased calcium signals or synaptic release from
defined neurons, and include a short-lived trace that forms
immediately after conditioning in antennal lobe projection
neurons, an early trace in dopaminergic neurons, and a
medium-term trace in dorsal paired medial neurons. New
molecular genetic tools have revealed that for normal
behavioral memory performance, synaptic transmission from
the mushroom body neurons is required only during retrieval,
whereas synaptic transmission from dopaminergic neurons is
required at the time of acquisition and synaptic transmission
from dorsal paired medial neurons is required during the
consolidation period. Such experimental results are helping to
identify the types of neurons that participate in olfactory
learning and when their participation is required. Olfactory
learning often occurs alongside crossmodal interactions of
sensory information from other modalities. Recent studies have
revealed complex interactions between the olfactory and the
visual senses that can occur during olfactory learning, including
the facilitation of learning about subthreshold olfactory stimuli
due to training with concurrent visual stimuli.
Addresses
1
Department of Molecular and Cellular Biology
2
Menninger Department of Psychiatry and Behavioral Sciences,
Baylor College of Medicine, Houston, TX 77030 USA
Corresponding author: Davis, Ronald L ([email protected])
Current Opinion in Neurobiology 2006, 16:679–685
This review comes from a themed issue on
Neurobiology of behaviour
Edited by John H Byrne and Wendy Suzuki
genes involved in olfactory learning were identified using
the fruit fly, Drosophila melanogaster. Several of the
identified genes were found to encode components in
the cAMP signaling pathway, such as the adenylyl cyclase
encoded by the rutabaga (rut) gene; the cAMP phosphodiesterase encoded by the dunce (dnc) gene; the cAMPdependent protein kinase (PKA); the predicted product
of the amnesiac (amn) gene, which has similarities to
pituitary adenylyl cyclase activating peptide (PACAP);
and the transcription factor cAMP-response element
binding protein (CREB) [1,2]. At the level of neural
circuits, a large body of evidence has accumulated that
suggests that mushroom bodies (MBs) are primary sites
for olfactory learning [2], and this evidence has biased
investigators to focus their studies almost exclusively on
the contribution of the MBs.
These discoveries have begun to reveal some of the
mechanisms underlying olfactory memory formation,
and they have also posed new and challenging questions.
For instance, is olfactory memory formed and stored
exclusively in the MBs until retrieval? Where do cellular
changes (cellular memory traces) occur within the olfactory nervous system in response to learning? What is the
relationship of cellular memory traces that form in the
brain to the various temporal phases (short-, medium-,
and long-term) of behavioral memory? How do different
sensory modalities interact to support learning? To
answer these questions, powerful new tools were needed.
Recently, researchers have developed new techniques to
visualize the formation of olfactory memory traces in the
brain and to dissect the contributions of various neurons
to temporal phases of olfactory memory using molecular
genetic strategies. We focus, here, on the progress made
in these areas in the past few years.
Available online 3rd November 2006
0959-4388/$ – see front matter
# 2006 Elsevier Ltd. All rights reserved.
DOI 10.1016/j.conb.2006.09.003
Introduction
All animals modify their behavior through learning, which
increases chances of survival. Insects rely heavily on
olfaction for various behaviors, including foraging, mating
and predator avoidance. The heavy reliance on olfactory
information, in part, has made insects attractive model
systems for dissecting olfactory memory formation. Significant progress was made in the last three decades of the
20th century in understanding the molecular biology and
anatomy that underlies insect olfactory memory. Many
www.sciencedirect.com
Functional optical imaging reveals multiple,
distributed cellular memory traces
Electrophysiological recordings have been widely used to
study the cellular or neuronal circuitry changes that occur
with learning in mammalian model systems. However,
such recordings are difficult to perform in insects because
of the small size of their bodies, brains and neurons.
Progress with functional optical imaging has helped to
solve this problem. In functional optical imaging, specially
engineered molecules (often fluorescent proteins) are used
to report the activity of neurons in living animals during
stimulation with sensory cues, such as an odor applied to an
insect, before or after olfactory learning has occurred.
Synthetic chemical reporters were first used for functional
optical imaging in the honeybee, Apis mellifera. Faber et al.
Current Opinion in Neurobiology 2006, 16:679–685
680 The neurobiology of behaviour
[3] applied the calcium sensitive dye, calcium-green-2AM,
through a small window cut in the bee’s head cuticle and
using the window, were able to image activity in the
antennal lobe. The investigators observed the activity in
response to odor stimuli presented before and after associative conditioning with sucrose as a rewarding stimulus.
They discovered that specific areas of the antennal lobe
exhibit a calcium signal in response to odor presented to
naı̈ve animals, and that this signal is increased for at least
30 min after conditioning. In addition, the patterns of
activation changed after learning, as evaluated by the
correlation in the pattern of activity elicited by a rewarded
odor before and after learning compared with that elicited
by an unrewarded one. Thus, olfactory conditioning of the
honeybee leads to quantitative and qualitative changes in
the calcium responses of undefined neurons that innervate
the antennal lobes. These and other changes in the
response properties of neurons described below could
underlie behavioral memory.
In Drosophila, functional optical imaging has employed
protein-based optical reporters expressed in specific neurons with the GAL4–UAS expression system. In this
binary system, one transgenic fly line carries the yeast
transcription activator GAL4 controlled by a tissue specific promoter, and a second transgenic fly line carries a
gene encoding a neuronal reporter downstream of the
binding site for GAL4, the upstream activating sequences
(UAS). Progeny from mating these two transgenic lines
express the neuronal reporter only in cells defined by the
tissue specific promoter [4]. Protein reporters that have
been used to report calcium influx include G-CaMP [5],
cameleon [6] and camgaroo [7], all of which have been
built on a green fluorescent protein (GFP) framework by
adding calcium binding motifs, such that the reporters
emit stronger fluorescence when intracellular calcium
levels are elevated. Wang et al. [5] expressed G-CaMP
in the primary olfactory sensory neurons and projection
neurons, and imaged antennal lobe calcium responses
that occurred when different odorants were presented to
the flies. They discovered that different odors generated
responses in specific sets of glomeruli that were unique to
the odor type, but conserved for each odor across flies. A
similar but less comprehensive activity map has also been
generated for the MBs using G-CaMP [8]. In addition to
calcium-sensitive reporters, a pH-sensitive reporter for
synaptic transmission (synapto-pHluorin) has also been
developed and used [9]. As synaptic vesicles fuse with the
plasma membrane to release neurotransmitter into the
synaptic cleft, the pH inside the vesicles increases from
acidic to neutral. Transgenically expressed synaptopHluorin, which resides on the lumen side of the synaptic
vesicle, reports this pH change and, therefore, the fusion
of synaptic vesicles that are releasing neurotransmitter.
Synaptic activity maps of antennal lobe neurons responding to odors have been made using this reporter, similar to
those generated using G-CaMP [9].
Current Opinion in Neurobiology 2006, 16:679–685
Recent studies using the protein-based reporters of neuronal activity have confirmed and extended the observation made in the honeybee that memory traces form in the
antennal lobes. The growing impression from these studies is that the antennal lobe is a site for olfactory memory
formation, in addition to the longer-held view that the
function of the lobe is to encode the identity of odor type.
Yu et al. [10] expressed synapto-pHluorin in specific types
of neurons that innervated the antennal lobes, and
imaged synaptic release in antennal lobe glomeruli in
living flies through a small window in the dorsal head
cuticle. Consistent with previous studies [5,9], they
observed that different odors stimulated synaptic release
in groups of glomeruli that were unique to each odor
tested but conserved among flies. This was observed
when synaptic release was monitored from olfactory
receptor neurons, GABAergic interneurons, or the projection neurons, which have presynaptic release sites in the
antennal lobe glomeruli and postsynaptic sites that
receive information from the olfactory receptor neurons.
Surprisingly, the activation pattern of sets of projection
neurons exhibited a qualitative rather than a quantitative
change in response to one cycle of conditioning with an
odor paired with electric shock. For instance, the odor
3-octanol stimulated synaptic release from four sets of
projection neurons (glomeruli) in naı̈ve flies, but from five
sets of projection neurons after conditioning (Figure 1b,
left). Thus, the synapses of some sets of projection
neurons that were silent in the naı̈ve state became active
after conditioning, and this occurred within 3 min of
training. This synaptic memory trace persisted for only
about 5 min after training, suggesting that the antennal
lobes have a role in olfactory memory for only a few
minutes after learning. Electrophysiology and behavioral
data from locust [11] and moth [12] also support a role
for the antennal lobes in short-term olfactory memory
formation.
In addition, the antennal lobe projection neurons might
participate in long-term memory formation. Ashraf et al.
[13] constructed a reporter transgene for synaptic protein synthesis by fusing a GFP coding cassette with the 30
untranslated region (30 UTR) of the Drosophila calcium–
calmodulin protein kinase II (CaMKII) gene. The 30 UTR of
the CaMKII gene contains sequences that target its
mRNAs to the synaptic compartments of neurons for
synaptic protein synthesis. When the reporter transgene
was expressed in projection neurons, the investigators
were able to monitor changes in synaptic protein synthesis due to neural activity of these neurons by following
the expression level of GFP. Remarkably, they discovered that synaptic protein synthesis, as measured with
this reporter, was increased in some sets of projection
neurons 24 h after the flies received multiple pairings of
the odor and the electric shock in a temporally spaced
configuration (with a rest between each pairing;
Figure 1b, right). Spaced olfactory conditioning produces
www.sciencedirect.com
Insect olfactory memory in time and space Liu and Davis 681
Figure 1
Neurons with defined olfactory memory traces. (a) Olfactory-learning related structures in Drosophila. MB cells (blue circles) extend dendrites
into the calyx (C) region and send axons through peduncles (P) that branch into different lobes (a, a’, b, b’ and g). Projection neurons from the
antennal lobes (AL, green) make synaptic connections with MB cells in the calyces. Dorsal paired medial neurons (DPM, red) and dopaminergic
neurons (DA, yellow) innervate MB lobes. Memory traces in specific glomeruli of the AL, DPM and DA neurons have been detected with
functional optical imaging. (b) Short-term and long-term memory traces in the ALs. An illustration of the AL with glomeruli labeled is shown
for both short and long term traces. Left: for the short-term memory trace [10], orange indicates glomeruli responding to 3-octanol in both naı̈ve
and trained flies; magenta indicates glomeruli showing detectable synaptic responses to the odor only after training. Right: for the long-term
memory trace [13], orange indicates some of the glomeruli that were assayed for synaptic protein synthesis in response to spaced conditioning;
magenta indicates glomeruli that exhibited increased synaptic protein synthesis after spaced conditioning. Note the overlap of the responding
glomeruli between the short term (synaptic transmission) and the long term (synaptic protein synthesis) traces. (c) Time course of memory
traces in different neurons. These memory traces were detected by increased calcium influx, synaptic transmission or synaptic protein synthesis
using optical reporters for these processes. Solid lines indicate experimental results; dashed lines indicate hypothetical extensions. AL traces
(green) are from [10] and [13]; DPM (red) trace from [17]; DA trace (yellow) from [19].
the formation of long-term, protein synthesis-dependent
behavioral memory, whereas a single training trial or
massed training with multiple trials (with no rests between
pairings) fails to produce this form of memory. Moreover,
the specific sets of projection neurons showing increased
synaptic protein synthesis after spaced conditioning were,
strikingly, overlapping with those previously described to
have increased synaptic activity after single pairing (shortterm) training [10], which infers that the mechanisms that
lead to the increased synaptic activity over the short-term
might also lead to synaptic protein synthesis for long-term
memory (Figure 1b). The regulation of synaptic protein
synthesis for normal long-term memory is apparently
under the control of the RNA-induced silencing complex
(RISC) pathway, because genetic disruptions of protein
components of the RISC pathway blocked the traininginduced changes in synaptic protein synthesis and longterm behavioral memory. Spaced conditioning induces a
prolonged activation of PKA in the honeybee antennal
lobes. A single training trial along with the release of caged
cAMP in the antennal lobes by photostimulation also
induces long-term memory [14]. This study provides
www.sciencedirect.com
further evidence that supports a role for the antennal lobes
in long-term memory formation.
Amnesiac (amn) is a Drosophila olfactory memory mutant
with a strong effect on post-training performance that
begins 15–30 min after conditioning [15]. Immunohistochemical studies revealed that the putative product of the
gene is expressed in the dorsal paired medial (DPM)
neurons, a pair of neurons in the brain sending two major
processes into the neuropil areas that house the axons of
MB cells [16]. The behavioral effect of the mutant has
been ascribed to the DPM neurons, because expression of
a wild type amn gene in these neurons is sufficient to
rescue the behavioral phenotype. The gene is regarded as
primarily influencing medium-term memory, because
there is no memory defect immediately after acquisition,
and synaptic transmission from these neurons is not
required during acquisition, as revealed by synaptic
blocking experiments (see below).
Yu et al. [17] expressed G-CaMP or synapto-pHluorin
in the DPM neurons and imaged the odor-response
Current Opinion in Neurobiology 2006, 16:679–685
682 The neurobiology of behaviour
properties of DPM neuronal process after olfactory learning. They found that pairing odor with shock increased
subsequent odor-evoked calcium influx and synaptic
release from the DPM neurons, but this increase was
delayed, appearing first at 30 min after training and persisting for at least an hour (Figure 1c, middle). Remarkably, this delayed memory trace was restricted to the
DPM processes that innervate the vertical lobes of mushroom bodies and not the horizontal lobes, and its formation required the activity of the amn gene in the DPM
neurons themselves. These functional imaging studies
have, therefore, identified a branch-specific, delayed
memory trace that forms in DPM neurons with kinetics
similar to those of behavioral medium-term memory.
The synaptic activity of dopaminergic neurons is required
for odor–electric shock learning, as shown in experiments
using a conditional block of dopaminergic synaptic transmission at the time of training (see below) [18]. Riemensperger et al. [19] expressed the optical reporter cameleon
in the dopaminergic neurons, and imaged the flies before
and after olfactory learning. Surprisingly, they detected
calcium responses in these neurons when odors were
presented to the flies, even though the dopaminergic
neurons are thought to be part of the unconditioned
stimulus (US) pathway (electric shock) and not the conditioned stimulus (CS) pathway (odor). The odor
responses of the neurons were also prolonged after training, indicating that conditioning alters the response of
these neurons to odor [19] (Figure 1c, bottom). These
data, therefore, surprisingly indicate that cellular memory
traces also form in dopaminergic neurons after training.
These recent studies suggest that cellular memory traces
form in a distributed fashion in the olfactory nervous
system after olfactory conditioning, but they do not
challenge a central role for the MBs in olfactory learning.
The neurons that form these memory traces (antennal
lobe projection neurons, DPM neurons and dopaminergic
neurons) are all thought to synapse onto the MB neurons,
potentially suggesting that these cellular memory traces
ultimately work through the MB neurons. Nevertheless,
MBs must not be viewed as the only anatomical structure
in which olfactory memories form. Future studies will
undoubtedly reveal the mechanisms underlying these
newly discovered memory traces and the general
issue of whether they are dependent or independent of
one another.
Molecular genetic approaches to dissect
behavioral memory in time and space
Functional optical imaging experiments, described
above, have provided evidence that olfactory memory
traces can form at different times and in different places
in the insect brain. Molecular genetic experiments have
also allowed for time and space dissections of behavioral
memory.
Current Opinion in Neurobiology 2006, 16:679–685
A central question for any mutant with memory deficits is
whether the loss of function of a specific gene alters the
development of the nervous system so as to perturb
memory formation or whether it alters the cellular
mechanisms underlying memory formation. To address
this important issue, systems that offer the experimenter
control of transgene expression in both time and space
have been developed. The traditional GAL4–UAS system relies on the inherent temporal expression pattern of
a tissue-specific promoter to drive the GAL4 transgene.
Two recently developed systems, named Gene-Switch
and TARGET, provide extensions of the GAL4–UAS
system for temporally controlling expression [20]. The
TARGET system uses a transgene encoding a temperature-sensitive inhibitor of GAL4, named GAL80ts, in
addition to the GAL4 and UAS components. At low
temperatures (18 8C), GAL80ts represses GAL4 activity
to inhibit expression of the UAS-transgene. At high
temperatures (32 8C), the repression is released and
GAL4 activates expression from the UAS component
in the spatial pattern defined by the promoter driving
GAL4. In Gene-Switch, a GAL4-progesterone receptor
chimera replaced the GAL4 component, which functions
as a transcriptional activator only in the presence of
mifepristone (RU486). The tissue-specific promoter,
therefore, confers the pattern of expression, and administration of RU486 through feeding provides temporal
control.
McGuire et al. [21] used the TARGET system to express
a wild type rut transgene in rut null mutant flies. They
discovered that expression of this transgene in the MBs
only during adulthood was sufficient to rescue the olfactory learning phenotype. This demonstrates that the rut
learning phenotype is due to the loss of rut activity in the
adult brain after development is completed. The same
conclusion was reached in experiments using GeneSwitch to provide temporal control [22]. Ferris et al.
[23] disrupted olfactory learning by expressing pertussis
toxin, an inhibitor of heterotrimeric G(o) proteins, in
adult fly mushroom bodies using both TARGET and
Gene-Switch systems to prove that G(o)-based signaling
is required for learning.
One useful division of memory is formed on the basis of
its persistence, such as short-, medium- and long-term
memory. But memory can also be divided operationally
into acquisition, stabilization, consolidation and retrieval.
A major challenge in the functional dissection of memory
is to assign the specific phases and operations of memory
to different brain regions, neurons, cellular processes and
genes. New and old techniques and reagents have begun
to make this possible in Drosophila.
Structural brain mutants have been studied for decades
for their effects on memory formation in an attempt
to dissect the underlying anatomy. Pascual et al. [24]
www.sciencedirect.com
Insect olfactory memory in time and space Liu and Davis 683
identified a new structural brain mutant that causes the
variable loss of the vertical and/or the horizontal lobes of
the MBs in different animals. Although the mutant flies
exhibit normal performance at all times after training
when evaluated all together, if analyzed as a subgroup,
those flies missing the vertical lobes exhibit no long-term
but normal short-term memory. Animals missing the
horizontal lobes have no phenotype. These structural
studies, therefore, suggest that long-term memory is
either formed within the vertical lobes of the MBs or
distributed from these lobes.
The mutant shibirets (shits) causes a rapid and reversible
block in synaptic transmission at high temperatures
(>29 8C), owing to the function of the gene product in
replenishing the neurotransmitter vesicle population at
synapses. This mutant gene was placed under UAS control (UAS-shits) in a transgene as a new tool to evaluate the
importance of synaptic transmission from specific neurons. Two different groups expressed this transgene in
the MBs at elevated temperatures at different time points
during olfactory learning. They showed that synaptic
transmission from these neurons is only required at the
time of retrieval [25,26]. Blocking synaptic transmission
from these neurons at the time of acquisition or during the
subsequent consolidation phase was without effect on
performance, which was tested at times up to 3 h after
training. These studies, therefore, indicate that early
memories are formed in the MB or upstream neurons,
because blocking MB output had no effect on acquisition
or stability for the first few hours. Keene et al. [27]
performed similar experiments to block synaptic transmission from the DPM neurons between 0 and 2 h after
odor–electric shock training, a time window during which
consolidation is thought to occur. When certain odor
combinations were used for training, this blockade
impaired performance when tested at 3 h after training,
suggesting that synaptic output onto the MB neurons
during this period is mechanistically part of the consolidation process. Blocking synaptic transmission during the
acquisition or retrieval phases is without effect on memory tested 3 hours after training. Similar results have been
obtained using odor–sucrose reward training, indicating
that the mechanisms provided by DPM synaptic transmission onto the following neurons are used for consolidating both aversive and appetitive unconditioned
stimuli [28]. Finally, UAS-shits was used to demonstrate
that synaptic release from dopaminergic neurons is
required during acquisition for normal memory formation
[18].
Olfactory memory formation across
extended space: crossmodal interactions
for olfactory learning
Learning is generally a complex process that integrates
different sensory cues. Olfactory memory, visual memory,
taste memory and other forms of memories represent
www.sciencedirect.com
unimodal memories within the whole of ‘memory space’,
and they can interact extensively with each other in this
space. To dissect how olfactory memory interacts with
other forms of memories within this extended space,
crossmodal learning paradigms have been invented.
Guo and Guo [29] combined the flight simulator with an
odor delivery system to study these interactions. They
used a noxious heat stimulus as a negative reinforcer to
train flies suspended by a thin wire to fly towards different
shapes, odors or their combinations in an artificial flight
space. They first defined the discriminatory threshold of
the visual cue (horizontal bars presented at different
vertical positions relative to the horizon) required for
learning an association between one visual cue and the
heat reinforcer, and independently the odor concentration threshold necessary for learning to discriminate a
punished and an unpunished odor. When presented
with subthreshold visual and olfactory cues together,
Drosophila display significant learning to either cue,
although each cue by itself is inadequate to support
learning. Moreover, when flies were pre-exposed to both
visual and olfactory cues simultaneously without reinforcement, and then trained to associate only one of the cues
with heat punishment, they were able to respond to the
cue representing the other sensory modality during testing. This suggests that the association of visual and
olfactory cues during preconditioning provided the
cross-modal memory space to transfer subsequent
reinforced associations across modalities.
Crossmodal learning studies have also been performed in
honeybees. To test if honeybees were capable of complex
learning tasks, Giurfa et al. [30] passed honeybees through
series of training chambers, each of which contained one
of two odors. The bees were rewarded with sucrose if they
entered the final chamber with the same odor as that
experienced in the first chamber. In some trials, the
honeybees were put through a similar series of chambers
except that the two odor cues were replaced by two
different colors. They preferred the final chamber with
the same color as the first chamber. The opposite was
observed if the honeybees were trained to choose the odor
that is different from the one experienced in the first
chamber. These experiments show that honeybees can
transfer the concept of ‘sameness’ and ‘difference’ across
olfactory and visual memories. Reinhard et al. [31] tested
crossmodal interactions in honeybees with a setting closely mimicking the natural environment. They trained
bees to forage at sugar water feeders associated with
different odors and colors that were placed at different
outdoor locations. Presenting a specific odor into the hive
prompted the bees to locate the feeders previously associated with that odor, even though the odor was not
present during testing. Thus, memory of a specific odor
triggers complex navigational behavior in the honeybee
that engages visual memory.
Current Opinion in Neurobiology 2006, 16:679–685
684 The neurobiology of behaviour
Conclusions
Olfactory memory formation is a complicated process
that is being dissected with many different techniques.
The dissections of when and where memory traces are
formed in the brain, of when and where gene products are
required in the brain for normal learning, and of when
and where synaptic transmission is required to support
memory formation are among the most fundamental
issues in the field of learning and memory. New techniques and tools are starting to help unravel these issues.
Current and future research will enable not only the
dissection of olfactory memory formation in both time
and space but will also incorporate this information into a
general multi-sensory learning program. Deeper insights
are anticipated using combined molecular–genetic and
systems neuroscience approaches to study olfactory
learning.
References and recommended reading
Papers of particular interest, published within the annual period of
review, have been highlighted as:
of special interest
of outstanding interest
1.
McGuire SE, Deshazer M, Davis RL: Thirty years of olfactory
learning and memory research in Drosophila melanogaster.
Prog Neurobiol 2005, 76:328-347.
2.
Davis RL: Olfactory memory formation in Drosophila: from
molecular to systems neuroscience. Annu Rev Neurosci 2005,
28:275-302.
in primary olfactory networks. Proc Natl Acad Sci USA 2004,
101:10476-10481.
13. Ashraf SI, McLoon AL, Sclarsic SM, Kunes S: Synaptic protein
synthesis associated with memory is regulated by the RISC
pathway in Drosophila. Cell 2006, 124:191-205.
Synaptic protein synthesis is crucial for long-term memory. The authors
used a new reporter containing the 30 UTR of the CaMKII gene to monitor
synaptic protein synthesis in projection neurons of the Drosophila antennal lobes after long-term olfactory associative learning. They found that
the mRNA reporter was directed to synaptic regions and translated in
specific sets of projection neurons after spaced training. The authors also
found that this process is regulated by the RISC pathway.
14. Muller U: Prolonged activation of cAMP-dependent protein
kinase during conditioning induces long-term memory in
honeybees. Neuron 2000, 27:159-168.
15. Quinn WG, Sziber PP, Booker R: The Drosophila memory mutant
amnesiac. Nature 1979, 277:212-214.
16. Waddell S, Armstrong JD, Kitamoto T, Kaiser K, Quinn WG:
The amnesiac gene product is expressed in two neurons in the
Drosophila brain that are critical for memory. Cell 2000,
103:805-813.
17. Yu D, Keene AC, Srivatsan A, Waddell S, Davis RL: Drosophila
DPM neurons form a delayed and branch-specific memory
trace after olfactory classical conditioning. Cell 2005,
123:945-957.
Using G-CaMP and synapto-pHluorin as reporters, the authors imaged
the activity and synaptic release of DPM neuronal processes after
olfactory conditioning. They detected a memory trace registered as
increased calcium influx and increased synaptic release at 30 minutes
after training but not before, and that persisted for at least 1 hour. This
memory trace is restricted to the DPM neuronal processes that innervate
the vertical lobes of MBs but not the horizontal lobes. Formation of the
memory trace also requires normal amn gene function in the DPM
neurons. These observations are consistent with behavioral data indicating a role for the DPM neurons in medium-term memory.
18. Schwaerzel M, Monastirioti M, Scholz H, Friggi-Grelin F,
Birman S, Heisenberg M: Dopamine and octopamine
differentiate between aversive and appetitive olfactory
memories in Drosophila. J Neurosci 2003,
23:10495-10502.
3.
Faber T, Joerges J, Menzel R: Associative learning modifies
neural representations of odors in the insect brain.
Nat Neurosci 1999, 2:74-78.
4.
Duffy JB: GAL4 system in Drosophila: a fly geneticist’s Swiss
army knife. Genesis 2002, 34:1-15.
19. Riemensperger T, Voller T, Stock P, Buchner E, Fiala A:
Punishment prediction by dopaminergic neurons in
Drosophila. Curr Biol 2005, 15:1953-1960.
5.
Wang JW, Wong AM, Flores J, Vosshall LB, Axel R: Two-photon
calcium imaging reveals an odor-evoked map of activity in the
fly brain. Cell 2003, 112:271-282.
20. McGuire SE, Roman G, Davis RL: Gene expression systems in
Drosophila: a synthesis of time and space. Trends Genet 2004,
20:384-391.
6.
Fiala A, Spall T, Diegelmann S, Eisermann B, Sachse S,
Devaud J-M, Buchner E, Galizia CG: Genetically expressed
cameleon in Drosophila melanogaster is used to visualize
olfactory information in projection neurons. Curr Biol 2002,
12:1877-1884.
21. McGuire SE, Le PT, Osborn AJ, Matsumoto K, Davis RL:
Spatiotemporal rescue of memory dysfunction in Drosophila.
Science 2003, 302:1765-1768.
7.
Yu D, Baird GS, Tsien RY, Davis RL: Detection of calcium
transients in Drosophila mushroom body neurons with
camgaroo reporters. J Neurosci 2003, 23:64-72.
8.
Wang Y, Guo HF, Pologruto TA, Hannan F, Hakker I, Svoboda K,
Zhong Y: Stereotyped odor-evoked activity in the mushroom
body of Drosophila revealed by green fluorescent proteinbased Ca2+ imaging. J Neurosci 2004, 24:6507-6514.
9.
Ng M, Roorda RD, Lima SQ, Zemelman BV, Morcillo P,
Miesenbock G: Transmission of olfactory information between
three populations of neurons in the antennal lobe of the fly.
Neuron 2002, 36:463-474.
10. Yu D, Ponomarev A, Davis RL: Altered representation of the
spatial code for odors after olfactory classical conditioning:
memory trace formation by synaptic recruitment. Neuron 2004,
42:437-449.
11. Bazhenov M, Stopfer M, Sejnowski TJ, Laurent G: Fast odor
learning improves reliability of odor responses in the locust
antennal lobe. Neuron 2005, 46:483-492.
12. Daly KC, Christensen TA, Lei H, Smith BH, Hildebrand JG:
Learning modulates the ensemble representations for odors
Current Opinion in Neurobiology 2006, 16:679–685
22. Mao Z, Roman G, Zong L, Davis RL: Pharmacogenetic
rescue in time and space of the rutabaga memory impairment
by using Gene-Switch. Proc Natl Acad Sci USA 2004,
101:198-203.
23. Ferris J, Ge H, Liu L, Roman G: G(o) signaling is required
for Drosophila associative learning. Nat Neurosci 2006,
9:1036-1040.
24. Pascual A, Preat T: Localization of long-term memory
within the Drosophila mushroom body. Science 2001,
294:1115-1117.
25. McGuire SE, Le PT, Davis RL: The role of Drosophila mushroom
body signaling in olfactory memory. Science 2001,
293:1330-1333.
26. Dubnau J, Grady L, Kitamoto T, Tully T: Disruption of
neurotransmission in Drosophila mushroom body blocks
retrieval but not acquisition of memory. Nature 2001,
411:476-480.
27. Keene AC, Stratmann M, Keller A, Perrat PN, Vosshall LB,
Waddell S: Diverse odor-conditioned memories require
uniquely timed dorsal paired medial neuron output.
Neuron 2004, 44:521-533.
www.sciencedirect.com
Insect olfactory memory in time and space Liu and Davis 685
28. Keene AC, Krashes MJ, Leung B, Bernard JA, Waddell S:
Drosophila dorsal paired medial neurons provide a general
mechanism for memory consolidation. Curr Biol 2006,
16:1524-1530.
29. Guo J, Guo A: Crossmodal interactions between olfactory and
visual learning in Drosophila. Science 2005, 309:307-310.
The authors trained flies with both olfactory and visual cues in the flight
simulator. They found subthreshold visual and olfactory cues elicited
significant learning when presented simultaneously (crossmodal
enhancement). They also found visual or olfactory memory is transferable
www.sciencedirect.com
across modalities if both stimuli are presented together before training
(crossmodal transfer).
30. Giurfa M, Zhang S, Jenett A, Menzel R, Srinivasan MV:
The concepts of ‘sameness’ and ‘difference’ in an insect.
Nature 2001, 410:930-933.
31. Reinhard J, Srinivasan MV, Guez D, Zhang SW: Floral scents
induce recall of navigational and visual memories in
honeybees. J Exp Biol 2004, 207:4371-4381.
Current Opinion in Neurobiology 2006, 16:679–685