<<

Current Biology, Vol. 15, R714–R722, September 6, 2005, ©2005 Elsevier Ltd. All rights reserved. DOI 10.1016/j.cub.2005.08.019

The Circadian Timekeeping System of Review

Paul E. Hardin products, in particular, whether they are thought to activate , to repress transcription, to alter stability or subcellular localization, or to Daily rhythms in behavior, physiology and degrade . In the transcriptional activator cat- metabolism are controlled by endogenous circadian egory are two basic-helix-loop-helix/PAS domain clocks. At the heart of these clocks is a circadian transcription factors, Clock (Clk) and (Cyc), oscillator that keeps circadian time, is entrained by which form heterodimers to activate transcription environmental cues such as light and activates [2–4], and one basic-leucine zipper transcription rhythmic outputs at the appropriate time of day. factor, Par domain protein 1ε (PDP1ε) [5]. The Genetic and molecular analyses in Drosophila have transcriptional repressor category includes revealed important insights into the molecules and (Per), another PAS domain protein, and (Tim), mechanisms underlying circadian oscillator function which function as heterodimers to inhibit Clk–Cyc in all organisms. In this review I will describe the function [4], and the basic-leucine zipper transcrip- intracellular feedback loops that form the core of the tional repressor Vrille (Vri) [5–7]. Drosophila circadian oscillator and consider how The category of proteins that alter protein stability they are entrained by environmental light cycles, and subcellular localization includes kinases that where they operate within the and how they are destabilize proteins that control transcrip- thought to control overt rhythms in physiology and tion: the mammalian casein kinase 1ε (CK1ε) homolog behavior. I will also discuss where work remains to (Dbt), also known as Discs overgrown or be done to give a comprehensive picture of the Dco, destabilizes Per [8,9]; casein kinase 2 (CK2), circadian clock in Drosophila and likely many other which has α and β subunits, destabilizes Per and also organisms. affects its nuclear localization [10,11]; and the glucose synthase kinase 3 (GSK3) homolog Shaggy (Sgg) phosphorylates Tim to promote nuclear localization of Per–Tim heterodimers [12]. In contrast to the destabi- Introduction lizing effects of the protein kinases, protein phos- Circadian clocks regulate rhythmic phenomena in phatase 2a (PP2a), which has regulatory subunits animals, plants, fungi and even some prokaryotes. In Twins (Tws) and Widerborst (Wdb), stabilizes Per via Drosophila, these clocks control a number of rhythmic dephosphorylation [13]. The protein degradation cat- outputs, including adult emergence (eclosion), egory includes the F-Box/WD40 protein Slimb (Slmb), locomotor activity and olfactory physiology. The which targets phosphorylated Per for degradation in molecular nature of the Drosophila clock is being the proteasome [14,15]. elucidated at a rapid pace, and serves as a good Having introduced the key components of the model for clocks in other animals given that many of Drosophila clock, in the next section I will focus on the components have been conserved. Moreover, their roles within the circadian oscillator. clocks enable an organism to adapt to daily environ- mental cycles by mechanisms that are starting to be Molecular Circuitry of the Drosophila Circadian uncovered in Drosophila. After an initial description of Oscillator Drosophila clock components, I will focus on the The Drosophila circadian oscillator is composed of intracellular feedback loops that lie at the center of the two intracellular feedback loops in : a circadian oscillator and how they are entrained by Per/Tim loop and a Clk loop [16,17]. Within these light. I will then describe where these oscillators feedback loops, rhythmic transcription of particular operate in adult , the rhythmic outputs they are clock is controlled via feedback from their own known or suspected to control, and how these protein products. Post-translational mechanisms rhythms are proposed to be controlled, before control the levels and subcellular localization of clock concluding with a perspective on the direction and proteins so that transcriptional feedback occurs at the significance of future work on the Drosophila clock. appropriate time of day. These feedback loops use different mechanisms to regulate transcription in Components of the Drosophila Circadian Clock different phases of the circadian cycle, yet are Genetic analysis has revealed a number of ‘clock’ interlocked by their requirement for Clk–Cyc genes that are critical for clock function in Drosophila dependent transcription. The Per/Tim loop is required (reviewed in [1]). These genes can be divided up for the function of both loops, and will thus be roughly according to the molecular nature of their described first.

The Per/Tim Feedback Loop Department of Biology and Biochemistry, University of To initiate the Per/Tim feedback loop, Clk–Cyc Houston, 4800 Calhoun Road, Houston, Texas 77204-5001, heterodimers bind E-box regulatory elements from USA. E-mail: [email protected] mid-day through early night, thereby activating Current Biology R715

Figure 1. Model of the Per/Tim feed- P Tim back loop. Sgg Clk–Cyc heterodimers bind to E-boxes P Per and activate transcription of Per and Dbt Tim Tim. As Per is produced it is phospho- P Per Dbt rylated by Dbt and CK2, which leads Clk Cyc Per, Tim to its degradation. Tim binds to, and E-Box stabilizes, phosphorylated Per, which X P Tim P PP2a remains bound to Dbt. Per is also sta- X

P P P Per P bilized by PP2a, which removes phos- Tim X phates that were added to Per. The P Per P Clk Tim–Per–Dbt complexes are phospho- Dbt P P Per Clk Cyc rylated by Sgg which, in concert with P P Tim by CK2, promotes P Per, Tim their transport into the nucleus. E-Box Tim–Per–Dbt complexes then bind to Per X P Per Clk–Cyc, thereby removing Clk–Cyc from the E-box and inhibiting Per and Clk Cyc CK2 Tim transcription. Per and Clk are then Dbt Per, Tim destabilized, via Dbt phosphorylation, E-Box and degraded, whereas Tim degrada- Clk Cyc tion (at least in response to light) is Nucleus Cytoplasm triggered by tyrosine phosphorylation. Current Biology The accumulation of non-phosphory- lated (or hypophosphorylated) Clk leads to heterodimerization with Cyc and another cycle of Per and Tim transcription. Solid lines with arrow, sequential steps in the feed- back loop; blocked line, inhibitory interaction; wavy line, Per and Tim mRNA; double arrow line, reversible phosphorylation; dashed lines, proteasomal degradation; black X, degraded proteins; P, protein phosphorylation; double line, nuclear membrane. transcription of the Per and Tim genes (Figure 1) Clk mRNA (see below) and activates E-box dependent [4,18–20]. The levels of Per and Tim transcripts peak transcription, thus starting the next transcriptional early in the night, whereas Per and Tim proteins do cycle (Figure 1). In addition to activating Per and Tim, not accumulate to peak levels until late evening Clk–Cyc directly activates Vri and Pdp1ε within the Clk [16,21–26]. This delay is the result of phosphorylation loop and a subset of clock output genes (see below). dependent destabilization of Per by Dbt, and possibly also CK2, followed by stabilization of phosphorylated The Clk Feedback Loop Per by Tim binding [8,9,11,27]. Per is also stabilized by In the Clk feedback loop, Clk–Cyc binds E-boxes to PP2a, which is thought to remove the phosphates activate high levels of Vri and Pdp1ε expression added by Dbt and CK2 [13]. during the late day and early night (Figure 1) [5–7]. Vri Dbt remains bound to Per to form a Per–Tim–Dbt accumulates in phase with its mRNA and binds complex, and the entire complex (or possibly just Tim) Vri/PDP1ε box (V/P box) regulatory elements to inhibit is translocated into the nucleus upon Sgg-dependent Clk transcription [5,7]. Consequently, Clk mRNA Tim phosphorylation and CK2-dependent Per cycles in the opposite phase as Clk–Cyc/E-box regu- phosphorylation [10–12,28–30]. Once in the nucleus, lated transcripts [5,7]. PDP1ε accumulates to high Per continues to be phosphorylated by Dbt, and this levels during the mid to late evening and activates Clk phosphorylation potentiates Per’s ability to repress transcription [5]. In vitro experiments showed that transcription [27]. Per appears to be a more potent PDP1ε can compete with Vri for binding to V/P-boxes, inhibitor of Clk–Cyc dependent transcription than suggesting a model in which increasing levels of Per–Tim [29,31], consistent with the observation that PDP1ε displace Vri from V/P-boxes and activate Clk Tim falls to low levels several hours before Per [26]. transcription [5]. Though attractive, this model does From the results of in vitro experiments, Per is not explain the constant peak levels of Clk expression thought to repress Clk–Cyc dependent transcription in non-functional ClkJrk and cyc01 mutants [17], which by binding to Clk and inhibiting the DNA binding virtually eliminate Vri and Pdp1ε expression [5]. To activity of Clk–Cyc dimers [32]. Recent in vivo analysis accommodate the ClkJrk and cyc01 results, a clock has not only confirmed this mode of regulation, but independent activator may drive constitutive Clk also suggests that Dbt dependent phosphorylation transcription, which is then rhythmically modulated by destabilizes Clk, explaining the coincidence between Vri and PDP1ε (Figure 2). phospho-Per and phospho-Clk levels (W. Yu, personal A common feature of the Per/Tim and Clk feedback communication). In addition, a more stringent extrac- loops is the activation of rhythmic transcription by tion procedure revealed that hypophosphorylated Clk Clk–Cyc. As rhythmic transcription of Clk–Cyc accumulates in antiphase to hyperphosphorylated Clk; activated genes requires feedback by Per–Tim, this thus, hypophosphorylated Clk accumulates in phase implies that the Per/Tim loop is also required for the with Per, Tim and other E-box/Clk-Cyc dependent Clk loop. Indeed, per01 and tim01 mutants abolish transcripts (W. Yu, personal communication). This transcriptional rhythms in both loops [5,7,16,17]. suggests a model in which hypophosphorylated, and Rhythms in the levels of Per and Tim persist if either of thus stable, Clk accumulates from declining levels of their respective mRNAs is constitutively expressed Review R716

Figure 2. Model of the Clk feedback loop. Clk–Cyc heterodimers bind to E-boxes and activate Vri and Pdp1ε transcription. Vri accumulates in parallel with its mRNA, binds to V/P boxes and inhibits Clk tran- Act Vri scription. PDP1ε accumulates in a Clk V/P-Box delayed fashion and supplants Vri from Vri V/P boxes to derepress Clk transcription. A clock independent activator (Act) constitutively activates Clk transcription in PDP1ε Act PDP1ε the absence of Vri, which would explain Clk the high levels of Clk mRNA in the V/P-Box absence of Clk or Cyc. Accumulation of non-phosphorylated (or hypophosphory- Clk lated) Clk leads to heterodimerization with Cyc Cyc and another cycle of Vri and Pdp1ε transcription. Solid lines with arrow, sequential steps in the feedback loop; wavy lines, Vri and Pdp1ε or Clk mRNA; double line, nuclear membrane. Clk Cyc Vri, Pdp1 E-Box

Nucleus Cytoplasm

Current Biology

[33,34], but eliminating cycling of both Per and Tim cycle to constant darkness (DD), light pulses applied mRNAs severely disrupts molecular and behavioral at times when lights would have been on (subjective rhythms [34], though not to the extent seen in non- day) produce little or no effect on oscillator phase, but functional per01 or tim01 mutants. In contrast, driving light pulses applied soon after lights would have gone Clk mRNA in antiphase has little effect on Clk off (early subjective night) produce a phase delay, and phosphorylation or behavioral rhythms ([35] and W. light pulses applied just before lights would have Yu, personal communication), which suggests that Clk come on (late subjective night) produce a phase mRNA cycling is not necessary for circadian oscillator advance [44–46]. Substantial headway has been made function. in identifying the molecules and mechanisms that The Clk loop also controls rhythmic transcription of mediate the light dependent phase shifting — (Cry), which encodes a circadian entrainment — of the Drosophila circadian oscillator. photoreceptor that also functions as a clock compo- In general, light entrains a circadian oscillator by nent in some tissues [36–39]. In fly heads, Cry protein activating a photoreceptor, which then directly or levels accumulate in the dark and decline in the light indirectly alters the level or activity of an oscillator [37,40]; Cry abundance is thus driven by environmen- component. In Drosophila, several photoreceptors are tal light–dark (LD) cycles rather than Cry mRNA cycles. activated by light, and these photoreceptors then Moreover, Cry photoreceptor function can be rescued trigger the ubiquitin-proteasome dependent degrada- by constant levels of Cry mRNA [41,42]. As Clk and tion of Tim via tyrosine phosphorylation [47]. The light Cry mRNA cycling are not necessary for clock func- dependent degradation of Tim is thought to produce tion, perhaps a major function of the Clk loop is to phase advances and delays depending on the levels control rhythmic transcription of genes required for of Tim mRNA [24–26,48]. Early in the dark phase, Tim circadian behavior, physiology and metabolism. This levels can rebound after light induced degradation possibility could be tested genetically using mutants because of the high levels of tim mRNA, replaying a that eliminate Vri and/or Pdp1ε, but unfortunately such few hours of the circadian cycle and producing a mutants are developmental lethals [5,43]. Reduction phase delay. Late in the dark phase, Tim levels cannot or elimination of Vri and Pdp1ε via RNA interference rebound after light induced degradation because of (RNAi) in adult clock cells may be a fruitful method to the low levels of Tim mRNA, fast forwarding to the test Clk feedback loop function. next phase of the circadian cycle and advancing the phase of the clock. No phase shifting is seen during Entraining the Oscillator to Light the subjective day, as there is little or no Tim to The circadian oscillator must maintain synchrony with replenish. The light dependent loss of Tim leads to a environmental cycles to drive behavioral, physiologi- delay in Per nuclear localization and phosphorylation cal and metabolic outputs at the appropriate time of during the early night, and an advanced degradation day. Daily environmental cycles of light, temperature, of Per during the late night, consistent with the food and social interactions are all capable of entrain- changes in gene expression and behavior due to light ing circadian oscillators, but light is generally consid- applied at these times [48]. ered to be the strongest and most pervasive factor. Genetic analysis has revealed that external Light shifts the phase of the circadian oscillator in a photoreceptors in compound eyes and ocelli, internal predictable manner. If a fly is transferred from an LD photoreceptors in the Hofbauer-Buchner eyelet, and Current Biology R717

the blue light photoreceptor Cry all contribute to light dependent entrainment of behavioral rhythms in X Drosophila [49]. The mechanism by which external Cry and internal photoreceptors trigger light dependent Cry degradation of Tim in neurons that control behavioral rhythms has not been characterized, and it is unlikely that these photoreceptors entrain oscillator cells else- where in the fly head and body (see below). In con- Tyr Cry kinase trast, Cry is expressed in oscillator cells throughout Cry the head and body ([50] and H. Zheng, personal com- Tim munication), and recent work has revealed important +/Ð insights into how Cry triggers Tim degradation in X Complex response to light (Figure 3). Cry contains a conserved Tim P photolyase domain and a unique carboxy-terminal Tim +/Ð domain [51]. On stimulation by light, the carboxy-ter- +/Ð Complex minal domain of Cry is thought to shift position or Complex release an inhibitor to reveal a Tim binding site Current Biology [41,42,52]. Cry then binds Tim [53], which is associ- ated with the Per–Dbt or Per–Dbt–Clk–Cyc complex, Figure 3. Model for the light dependent entrainment of the and triggers its tyrosine phosphorylation and degra- Drosophila oscillator. dation by the proteasome [47]. Light also promotes Stimulation of CRY by light alters the carboxy-terminal domain Cry degradation, albeit more slowly, consistent with either structurally or by releasing an inhibitor, thereby the light dependent rhythm in Cry levels [37,40]. permitting binding to Tim. Cry binds to either Tim or a Tim complex — a Dbt–Per–Tim complex in the cytoplasm or a Dbt–Per–Tim–Clk–Cyc complex in the nucleus — and promotes Organization of the Drosophila Circadian System phosphorylation of Tim by a tyrosine kinase. Phosphorylated The spatial distribution of clock gene expression has Tim is then committed to rapid degradation in the proteasome. been used to infer the presence of circadian Prolonged light stimulation leads to the eventual degradation of oscillators in Drosophila. Analysis of reporter gene Cry in the proteasome. Solid lines with arrow, sequential steps expression driven by the per promoter and Per in the light response pathway; dashed lines, proteasomal degra- dation; black X, degraded proteins; P, protein phosphorylation. immunolocalization revealed expression in a variety of neuronal and non-neuronal tissues in fly heads and bodies [54–56]. In heads, Per is expressed in whole animals or cultured tissues over several days photoreceptors of the compound eye (which [55,64,65]. Individual cultured tissues such as anten- accounts for ~75% of all per expression in heads), nae, wings, probosci and legs rhythmically express antennae, the proboscis, ocelli, the esophagus, fat Per-luc under LD and DD conditions in the same cir- bodies, brain glia and six clusters of brain neurons cadian phase [55]. Particularly prevalent among [54–57] (Figure 4). These clusters of Per expressing tissues with circadian oscillators are those with brain neurons have been classified according to size sensory function, such as the photosensory ocelli and and position as small ventral lateral neurons (sLNVs), compound eyes, chemosensory tissues of the third large ventral lateral neurons (lLNVs), dorsal lateral antennal segment, maxillary palps, wings, legs and neurons (LNDs), dorsal neuron 1s (DN1s), dorsal proboscus, and mechanosensory tissues of the neuron 2s (DN2s) and dorsal neuron 3s (DN3s) [57] second antennal segment (Figure 4A). This suggests (Figure 4B). In the fly body, Per is expressed in the that the circadian clock controls some aspect of gut, the cardia, salivary glands, Malpighian tubules, sensory function; indeed, at least one chemosensory the rectum, legs, wings, fat bodies, ovaries and function, olfaction, is under clock control in flies [66]. testes [55,56]. Remarkably, oscillators in cultured Drosophila Essentially all of these tissues express Per gene tissues, whether overtly sensory or not, can be products rhythmically [21,55,58,59], indicating that directly entrained by light, indicating that they they have circadian oscillators. Although circadian operate in a tissue autonomous manner [55]. Pre- oscillator cells have been defined based on the sumably light entrains oscillators in peripheral expression of Per gene products, other feedback loop tissues via Cry, but this has been difficult to ascer- components such as Tim and Clk are, as expected, tain as Cry is important for oscillator function per se also made in these cells [24,25,60–62]. Unlike Per and in many cultured peripheral tissues [38,39,67]. Tim, which appear to be expressed almost exclusively Although the fly is generally viewed as a collection of in oscillator cells, Clk is also expressed in numerous autonomous, light-entrainable circadian oscillators, non-oscillator cells [60]. One tissue that does not current data cannot eliminate the possibility that express Per rhythmically is the ovary [58], yet even in there may be communication between oscillators. this tissue Tim is coproduced with Per, though not in Indeed, there is evidence that communication a Clk–Cyc dependent manner [63]. between oscillator neurons within the brain is impor- The relationship between oscillators in different tant for locomotor activity rhythms under LD and DD tissues has been addressed using luciferase reporter conditions [68–70], and that interaction between the genes driven by the per promoter (Per-luc), which prothoracic gland and lateral neurons is important have the advantage of allowing real-time monitoring of for rhythms in eclosion [71]. Review R718

These transcripts are under clock control, as their A OC rhythmic expression is abolished in arrhythmic Clk flies [72,73,75,76]. In addition, the levels of many non- rhythmic transcripts are increased or decreased in ClkJrk flies, which suggests that Clk regulates processes independently of the circadian clock [75]. AN2 AN2 Such clock independent regulation could reflect the action of Clk in non-oscillator cells [60]. To understand how clock controlled transcripts regulate rhythmic outputs, it is important to determine AN3 how the oscillator regulates cycling of these transcripts and to identify the behavioral, physiologi- CE CE cal and metabolic processes these transcripts control. Several clock controlled transcripts were found to be direct targets of Clk–Cyc in S2 cell culture assays [75], but direct targets of PDP1ε or Vri have not been identified. Determining what rhythmic processes these MP MP transcripts control is a major challenge, as it depends on making various behavioral, physiological or PR metabolic measurements over circadian time. Such B analyses are nevertheless essential for determining FB DN1 the biological impact of the clock. Circadian rhythms DN3 DN2 in several processes have been identified in adults, including locomotor activity, mating receptivity, LN CB D oviposition and olfaction [66,77–79]. Although no lLN V clock output pathway has been characterized in sLN ES OL V OL detail, progress has been made in understanding how locomotor activity and olfaction rhythms are regulated. FB Current Biology Regulation of Locomotor Activity Rhythms Figure 4. Circadian oscillators in Drosophila heads. In Drosophila, locomotor activity under DD conditions (A) External structures containing circadian oscillators. A frontal peaks during subjective dusk [78]. As a first step view of a Drosophila head is shown. OC, ocelli; CE, compound towards understanding how this rhythm is controlled, eyes; AN2, second antennal segment; AN3, third antennal the oscillator cells responsible for this rhythm were segment; MP, maxillary palps; PR, proboscis. (B) Oscillator identified. Evidence from flies in which large and small cells within and surrounding the brain. A frontal section through LNVs were ablated by expressing proapoptotic genes, a Drosophila brain and surrounding tissues is shown. CB, or electrically silenced by expressing constitutively central brain; OL, optic lobes; FB, fat body; ES, esophagus; + LN , dorsal lateral neurons; lLN , large ventral lateral neurons; active mutant K channels showed that these neurons D V are necessary for free-running (DD) locomotor activity sLNV, small ventral lateral neurons; DN1, dorsal neuron 1s; DN2, dorsal neuron 2s; DN3, dorsal neuron 3s. Glia containing rhythms [80,81]. Surprisingly, electrically silencing oscillators (not shown) are found within the central brain and large and small LNVs eliminates molecular oscillations optic lobes. of Per and Tim, suggesting that electrical activity is required for circadian oscillator function in these cells Regulating Rhythmic Outputs [80]. Restoring Per function in large and small LNVs of A fundamental question in circadian biology is how per01 flies rescues both circadian oscillator function in circadian oscillators regulate rhythms in behavior, these neurons and free-running locomotor activity physiology and metabolism. Given that the circadian rhythms [70]. As lLNVs do not sustain oscillator func- oscillator is composed of transcriptional feedback tion in DD [44,70], sLNVs alone appear to be sufficient loops, regulation of rhythmic outputs is likely to occur for this rhythmic output. In contrast, restoring Cyc 01 via clock dependent transcription of genes that function in large and small LNVs of cyc flies rescues directly control behavioral, physiological and molecular rhythms, but not free-running locomotor metabolic processes. Such regulation could also be activity rhythms [68]. It is possible that this failure of indirect; rhythmically expressed phosphatases, such Cyc to rescue locomotor activity rhythms occurs as PP2a, and/or kinases could mediate rhythms in the because of developmental defects in sLNV projections activity of proteins that the control output processes. in cyc01 flies [82]. Taken together, these data indicate The most efficient and comprehensive method of that sLNVs are necessary and sufficient for free- identifying transcripts that cycle in abundance has running locomotor activity rhythms. been through microarray analysis. About 150 cycling The sLNVs send projections into the dorsal brain, transcripts, representing genes that regulate close to the cell bodies of DN2 and DN3 clock processes such as protein degradation, detoxifica- neurons and Kenyon cells, which are required for tion, immunity, phototransduction and neurotrans- olfactory learning [57,83]. These projections contain a mission, have been identified in this way [72–76]. neuropeptide called pigment dispersing factor (PDF), Current Biology R719

which is required for free running locomotor activity oscillators, including olfactory receptor neurons, can rhythms in flies [81,84]. Circadian fluctuations in PDF be entrained directly by light [55]. Localization of the content within the termini of sLNV projections suggest EAG oscillator to olfactory receptor neurons suggests that PDF is rhythmically released [82]. These apparent that components of the olfactory rhythms in PDF release are sensitive to clock gene pathway, such as odorant receptors and G proteins, mutations, indicating they are under clock control [82]. may be targets for clock regulation. Identifying com- Moreover, expressing constant levels of PDF in the ponents of the olfactory signal transduction pathway vicinity of sLNV terminals disrupts locomotor activity that are rhythmically controlled may reveal clock rhythms, suggesting that rhythms in PDF release are outputs that effect EAG rhythms. important for free-running locomotor activity rhythms [85]. As constant levels of PDF expressed in brain Conclusion and Perspectives areas outside the sLNV projections were found not to Substantial progress has been made in defining the affect behavior [85], the site of PDF action appears to molecular mechanisms underlying circadian clock be in the dorsal brain. The presumptive G protein function in Drosophila. Input pathways have been coupled receptor for PDF has not been identified, but identified, and the mechanisms by which they entrain it is almost certainly among the ~20 orphan G protein the oscillator are being revealed at an ever-increasing coupled neuropeptide receptors in the Drosophila pace. The molecular feedback loop model of circadian genome [86]. Once the PDF receptor is in hand, PDF oscillator function has not only withstood the test of target cells and their projections can be defined to time, but is being elaborated in fine detail. These oscil- reveal the next piece of the neural circuit controlling lators are present in a variety of tissues, where they free-running locomotor activity rhythms. largely operate as autonomous clocks. A large Locomotor activity during LD cycles differs from number of clock controlled transcripts have been that in DD in that there are two activity peaks: one in identified, and progress is being made in defining the morning and one in the evening. Early studies with oscillators and, in the case of locomotor activity, Drosophila pseudoobscura and more contemporary molecules that mediate rhythmic outputs. But despite work with indicate that the the substantial progress that has been made in our morning and evening activity peaks are controlled by understanding of the Drosophila clock, many impor- separate oscillators [87,88]. Recently, these oscillators tant questions remain. were identified by targeting expression of clock genes One of the least understood aspects of circadian to specific sets of LNs and DNs via Gal4 activation clock function is how oscillators regulate rhythmic and/or Gal80 inhibition [69,70]. These experiments outputs. There are many oscillators in the fly, yet few showed that the morning activity peak is driven by the rhythmic outputs have been characterized. Rhythms LNVs, whereas the evening peak is driven by the LNDs in locomotor activity have been the most extensively and possibly a subset of DN1s. As the LNVs drive characterized output in flies. Studies focusing on the morning activity during LD cycles and are sufficient for clock regulation of PDF release and identifying the activity rhythms in DD, it is intriguing that free-running PDF receptor should begin to define the molecular activity peaks during the subjective evening in wild- output mechanism and reveal the underlying neural type flies. Communication among these oscillator circuitry. Now that oscillator cells controlling olfaction neurons, possibly by PDF or some other neurotrans- rhythms have been identified, the mechanisms by mitter, may enhance the robustness and modulate the which they control rhythms in olfactory physiology can phase of locomotor activity rhythms [69,70]. Such be defined. communication may facilitate adjustments in the The small number of clock outputs is a reflection of phase of locomotor activity by environmental factors the difficulty in developing or performing assays that such as temperature and photoperiod [89]. can be used to measure circadian rhythms in metabolism, physiology and behavior. The presence Regulation of Olfaction Rhythms of circadian oscillators in a host of sensory tissues Another circadian output that has been characterized nevertheless suggests a number of quantitative in flies is a rhythm in olfaction. This rhythm is behavioural and physiological assays. In addition, the measured by assaying the magnitude of odor-induced spatial expression and predicted function of electrophysiological responses in the antennae called rhythmically expressed mRNAs may highlight specific electroantennagrams (EAGs). EAG responses to the processes to target for circadian monitoring. As more food odorant ethyl acetate show a robust rhythm in clock outputs are discovered, their clock dependent wild-type flies under both LD and DD conditions [66]. regulation must be defined so that we can begin to These rhythms are abolished in per01 and tim01 flies, understand the various mechanisms by which the confirming that they are under clock control [66]. clock imposes its control on behavior, physiology and Rhythms in EAG responses were known to require the metabolism. function of oscillators in peripheral tissues [66], and In addition to controlling daily rhythms, the recent studies in which oscillator function was either Drosophila circadian clock also controls annual disrupted or rescued in specific cell types have shown rhythms in various phenomena, including the phase of that olfactory receptor neurons in the antenna are locomotor activity and reproduction [91,92]. In the both necessary and sufficient for the EAG rhythms case of locomotor activity, flies are most active just [90]. These olfactory receptor neuron oscillators before dawn and after dusk during the summer, but behave as self-contained clocks because antennal after dawn and before dusk during the winter [90]. The Review R720

clock adapts to these seasonal changes in the envi- ocelli and/or the Hofbauer-Buchner eyelet can ronment through the thermosensitive splicing of an mediate light entrainment. As the mammalian SCN is intron in the 3′ untranslated region of Per mRNA; also entrained by photoreceptors residing in other enhanced splicing in cold temperatures and short tissues, understanding how light information is trans- photoperiods advances oscillator phase and conse- mitted to sLNVs and triggers a phase shift in flies is of quently locomotor activity, whereas reduced splicing great interest. in longer photoperiods and warm temperatures delay Analysis of the Drosophila circadian system has led oscillator phase and locomotor activity [89,93,94]. The to many breakthroughs in our understanding of circa- molecular mechanism that regulates this splicing dian clock function. The studies outlined above should event is not clear, though light, clock factors and a continue that tradition by providing the mechanistic phospholipase C are all involved [93,94]. Understand- detail needed to understand how the clock functions ing how these factors effect splicing of this intron will as an integrated set of components to drive daily provide significant insight into how the clock adapts rhythms in behavior, physiology and metabolism. to a changing environment. The current model of transcriptional regulation Acknowledgments within the Clk feedback loop is necessarily I would like to thank Nick Glossop and Wangjie Yu for incomplete. First, it is surprising that Clk and Cry are comments on the manuscript. This work was supported by rhythmically transcribed, even though Clk and Cry grants from the NIH. levels are controlled post-transcriptionally. Perhaps the primary function of Vri and PDP1ε is to control References 1. Hall, J.C. (2003). Genetics and molecular biology of rhythms in rhythmic transcription of output genes in phase with Drosophila and other . Adv. Genet. 48, 1–280. Clk and Cry. This possibility could be addressed by 2. Rutila, J.E., Suri, V., Le, M., So, W.V., Rosbash, M., and Hall, J.C. tissue-specific elimination or reduction of Vri and (1998). CYCLE is a second bHLH-PAS clock protein essential for ε circadian rhythmicity and transcription of Drosophila period and Pdp1 function by RNAi or expressing dominant neg- timeless. Cell 93, 805–814. ative forms of the gene products. 3. Allada, R., White, N.E., So, W.V., Hall, J.C., and Rosbash, M. (1998). Secondly, although Vri and Pdp1ε are both A mutant Drosophila homolog of mammalian Clock disrupts circa- activated by Clk–Cyc, Pdp1ε mRNA accumulation is dian rhythms and transcription of period and timeless. Cell 93, 791–804. delayed compared to that of Vri [5]. This delay pro- 4. Darlington, T.K., Wager-Smith, K., Ceriani, M.F., Staknis, D., duces a concomitant delay in PDP1ε accumulation, Gekakis, N., Steeves, T.D., Weitz, C.J., Takahashi, J.S., and Kay, which counteracts Vri dependent repression and con- S.A. (1998). Closing the circadian loop: CLOCK-induced transcrip- sequently activates Clk and Cry. Understanding the tion of its own inhibitors per and tim. Science 280, 1599–1603. ε 5. Cyran, S.A., Buchsbaum, A.M., Reddy, K.L., Lin, M.C., Glossop, mechanism that governs Pdp1 mRNA accumulation N.R., Hardin, P.E., Young, M.W., Storti, R.V., and Blau, J. (2003). will be important for understanding how rhythmic tran- vrille, Pdp1, and dClock form a second feedback loop in the scription within and downstream of the Clk loop is Drosophila circadian clock. Cell 112, 329–341. 6. Blau, J., and Young, M.W. (1999). Cycling vrille expression is controlled. required for a functional Drosophila clock. Cell 99, 661–671. Jrk Lastly, the high levels of Clk mRNA in Clk and 7. Glossop, N.R., Houl, J.H., Zheng, H., Ng, F.S., Dudek, S.M., and cyc01 flies suggest that Clk is constitutively activated. Hardin, P.E. (2003). VRILLE feeds back to control circadian tran- As ectopic expression of Clk is capable of inducing scription of Clock in the Drosophila circadian oscillator. Neuron 37, 249–261. circadian clock function [95], it is possible that the 8. Price, J.L., Blau, J., Rothenfluh, A., Abodeely, M., Kloss, B., and constitutive activator is the same one that activates Young, M.W. (1998). double-time is a novel Drosophila clock gene Clk during development. Characterizing Clk develop- that regulates PERIOD protein accumulation. Cell 94, 83–95. mental expression and regulation might be a prof- 9. Kloss, B., Price, J.L., Saez, L., Blau, J., Rothenfluh, A., Wesley, C.S., and Young, M.W. (1998). The Drosophila clock gene double-time itable way to resolve this question. encodes a protein closely related to casein kinase Iepsilon. Recent studies have greatly enhanced our Cell 94, 97–107. understanding of how Cry entrains the circadian 10. Lin, J.M., Kilman, V.L., Keegan, K., Paddock, B., Emery-Le, M., Rosbash, M., and Allada, R. (2002). A role for casein kinase 2alpha oscillator in response to light. Once Cry is exposed to in the Drosophila circadian clock. Nature 420, 816–820. light it binds to Tim, thereby committing Tim to 11. Akten, B., Jauch, E., Genova, G.K., Kim, E.Y., Edery, I., Raabe, T., degradation by the proteasome. Tim degradation is and Jackson, F.R. (2003). A role for CK2 in the Drosophila circadian preceded by tyrosine phosphorylation, which oscillator. Nat. Neurosci. 6, 251–257. 12. Martinek, S., Inonog, S., Manoukian, A.S., and Young, M.W. (2001). presumably occurs after Tim is bound by activated A role for the segment polarity gene shaggy/GSK-3 in the Cry. Identifying the relevant tyrosine kinase will enable Drosophila circadian clock. Cell 105, 769–779. a detailed analysis of the interactions that take place 13. Sathyanarayanan, S., Zheng, X., Xiao, R., and Sehgal, A. (2004). to destabilize Tim. Posttranslational regulation of Drosophila PERIOD protein by protein phosphatase 2A. Cell 116, 603–615. In contrast to Cry’s role in light entrainment, how it 14. Ko, H.W., Jiang, J., and Edery, I. (2002). Role for Slimb in the degra- contributes to oscillator function in peripheral tissues dation of Drosophila Period protein phosphorylated by Doubletime. is poorly understood. In , Cryptochrome Nature 420, 673–678. binds to Period homologs to inhibit Clock–Bmal1 15. Grima, B., Lamouroux, A., Chelot, E., Papin, C., Limbourg-Bouchon, B., and Rouyer, F. (2002). The F-box protein slimb controls the dependent transcription [96]. Whether a similar situa- levels of clock proteins period and timeless. Nature 420, 178–182. tion exists in flies can now be addressed using newly 16. Hardin, P.E., Hall, J.C., and Rosbash, M. (1990). Feedback of the developed reagents for detecting Cry subcellular Drosophila period gene product on circadian cycling of its messen- localization and protein interactions. The sLN ger RNA levels. Nature 343, 536–540. V 17. Glossop, N.R., Lyons, L.C., and Hardin, P.E. (1999). Interlocked oscillator can be entrained to light independently of feedback loops within the Drosophila circadian oscillator. Science Cry, indicating that photoreceptors in compound eyes, 286, 766–768. Current Biology R721

18. Hao, H., Allen, D.L., and Hardin, P.E. (1997). A circadian enhancer 43. George, H., and Terracol, R. (1997). The vrille gene of Drosophila is mediates PER-dependent mRNA cycling in Drosophila a maternal enhancer of decapentaplegic and encodes a new melanogaster. Mol. Cell. Biol. 17, 3687–3693. member of the bZIP family of transcription factors. Genetics 146, 19. McDonald, M.J., Rosbash, M., and Emery, P. (2001). Wild-type cir- 1345–1363. cadian rhythmicity is dependent on closely spaced E boxes in the 44. Yang, Z., Emerson, M., Su, H.S., and Sehgal, A. (1998). Response of Drosophila timeless promoter. Mol. Cell. Biol. 21, 1207–1217. the timeless protein to light correlates with behavioral entrainment 20. Wang, G.K., Ousley, A., Darlington, T.K., Chen, D., Chen, Y., Fu, W., and suggests a nonvisual pathway for circadian photoreception. Hickman, L.J., Kay, S.A., and Sehgal, A. (2001). Regulation of the Neuron 21, 215–223. cycling of timeless (tim) RNA. J. Neurobiol. 47, 161–175. 45. Suri, V., Qian, Z., Hall, J.C., and Rosbash, M. (1998). Evidence that 21. Zerr, D.M., Hall, J.C., Rosbash, M., and Siwicki, K.K. (1990). Circa- the TIM light response is relevant to light-induced phase shifts in dian fluctuations of period protein immunoreactivity in the CNS and Drosophila melanogaster. Neuron 21, 225–234. the visual system of Drosophila. J. Neurosci. 10, 2749–2762. 46. Konopka, R.J. (1979). Genetic dissection of the Drosophila circadian 22. Edery, I., Zwiebel, L.J., Dembinska, M.E., and Rosbash, M. (1994). system. Fed. Proc. 38, 2602–2605. Temporal phosphorylation of the Drosophila period protein. Proc. 47. Naidoo, N., Song, W., Hunter-Ensor, M., and Sehgal, A. (1999). A Natl. Acad. Sci. USA 91, 2260–2264. role for the proteasome in the light response of the timeless clock 23. Sehgal, A., Rothenfluh-Hilfiker, A., Hunter-Ensor, M., Chen, Y., protein. Science 285, 1737–1741. Myers, M.P., and Young, M.W. (1995). Rhythmic expression of time- 48. Lee, C., Parikh, V., Itsukaichi, T., Bae, K., and Edery, I. (1996). Reset- less: a basis for promoting circadian cycles in period gene autoreg- ting the Drosophila clock by photic regulation of PER and a PER- ulation. Science 270, 808–810. TIM complex. Science 271, 1740–1744. 24. Hunter-Ensor, M., Ousley, A., and Sehgal, A. (1996). Regulation of 49. Helfrich-Forster, C., Winter, C., Hofbauer, A., Hall, J.C., and the Drosophila protein timeless suggests a mechanism for resetting Stanewsky, R. (2001). The circadian clock of fruit flies is blind after the circadian clock by light. Cell 84, 677–685. elimination of all known photoreceptors. Neuron 30, 249–261. 25. Myers, M.P., Wager-Smith, K., Rothenfluh-Hilfiker, A., and Young, 50. Klarsfeld, A., Malpel, S., Michard-Vanhee, C., Picot, M., Chelot, E., M.W. (1996). Light-induced degradation of TIMELESS and entrain- and Rouyer, F. (2004). Novel features of cryptochrome-mediated ment of the Drosophila circadian clock. Science 271, 1736–1740. photoreception in the brain circadian clock of Drosophila. J. Neu- 26. Zeng, H., Qian, Z., Myers, M.P., and Rosbash, M. (1996). A light- rosci. 24, 1468–1477. entrainment mechanism for the Drosophila circadian clock. Nature 51. Cashmore, A.R., Jarillo, J.A., Wu, Y.J., and Liu, D. (1999). Cryp- 380, 129–135. tochromes: blue light receptors for plants and animals. Science 284, 27. Nawathean, P., and Rosbash, M. (2004). The doubletime and CKII 760–765. kinases collaborate to potentiate Drosophila PER transcriptional 52. Rosato, E., Codd, V., Mazzotta, G., Piccin, A., Zordan, M., Costa, R., repressor activity. Mol. Cell 13, 213–223. and Kyriacou, C.P. (2001). Light-dependent interaction between 28. Kloss, B., Rothenfluh, A., Young, M.W., and Saez, L. (2001). Phos- Drosophila CRY and the clock protein PER mediated by the phorylation of period is influenced by cycling physical associations carboxy terminus of CRY. Curr. Biol. 11, 909–917. of double-time, period, and timeless in the Drosophila clock. 53. Ceriani, M.F., Darlington, T.K., Staknis, D., Mas, P., Petti, A.A., Neuron 30, 699–706. Weitz, C.J., and Kay, S.A. (1999). Light-dependent sequestration of 29. Ashmore, L.J., Sathyanarayanan, S., Silvestre, D.W., Emerson, TIMELESS by CRYPTOCHROME. Science 285, 553–556. M.M., Schotland, P., and Sehgal, A. (2003). Novel insights into the 54. Siwicki, K.K., Eastman, C., Petersen, G., Rosbash, M., and Hall, J.C. regulation of the timeless protein. J. Neurosci. 23, 7810–7819. (1988). Antibodies to the period gene product of Drosophila reveal 30. Shafer, O.T., Rosbash, M., and Truman, J.W. (2002). Sequential diverse tissue distribution and rhythmic changes in the visual nuclear accumulation of the clock proteins period and timeless in system. Neuron 1, 141–150. the pacemaker neurons of Drosophila melanogaster. J. Neurosci. 55. Plautz, J.D., Kaneko, M., Hall, J.C., and Kay, S.A. (1997). Indepen- 22, 5946–5954. dent photoreceptive circadian clocks throughout Drosophila. 31. Rothenfluh, A., Young, M.W., and Saez, L. (2000). A TIMELESS- Science 278, 1632–1635. independent function for PERIOD proteins in the Drosophila clock. 56. Liu, X., Lorenz, L., Yu, Q.N., Hall, J.C., and Rosbash, M. (1988). Neuron 26, 505–514. Spatial and temporal expression of the period gene in Drosophila 32. Lee, C., Bae, K., and Edery, I. (1999). PER and TIM inhibit the DNA melanogaster. Genes Dev. 2, 228–238. binding activity of a Drosophila CLOCK-CYC/dBMAL1 heterodimer 57. Helfrich-Forster, C. (2003). The neuroarchitecture of the circadian without disrupting formation of the heterodimer: a basis for circa- clock in the brain of Drosophila melanogaster. Microsc. Res. Tech. dian transcription. Mol. Cell. Biol. 19, 5316–5325. 62, 94–102. 33. Cheng, Y., and Hardin, P.E. (1998). Drosophila photoreceptors 58. Hardin, P.E. (1994). Analysis of period mRNA cycling in Drosophila contain an autonomous circadian oscillator that can function head and body tissues indicates that body oscillators behave dif- without period mRNA cycling. J. Neurosci. 18, 741–750. ferently from head oscillators. Mol. Cell. Biol. 14, 7211–7218. 34. Yang, Z., and Sehgal, A. (2001). Role of molecular oscillations in 59. Hege, D.M., Stanewsky, R., Hall, J.C., and Giebultowicz, J.M. (1997). generating behavioral rhythms in Drosophila. Neuron 29, 453–467. Rhythmic expression of a PER-reporter in the Malpighian tubules of 35. Kim, E.Y., Bae, K., Ng, F.S., Glossop, N.R., Hardin, P.E., and Edery, decapitated Drosophila: evidence for a brain-independent circadian I. (2002). Drosophila CLOCK protein is under posttranscriptional clock. J. Biol. Rhythms 12, 300–308. control and influences light-induced activity. Neuron 34, 69–81. 60. Houl, J.H., Yu, W., Dudek, S.M., and Hardin, P.E. (2005). Drosophila 36. Stanewsky, R., Kaneko, M., Emery, P., Beretta, B., Wager-Smith, K., CLOCK is constitutively expressed in circadian oscillator and non- Kay, S.A., Rosbash, M., and Hall, J.C. (1998). The cryb mutation oscillator cells. Submitted. identifies cryptochrome as a circadian photoreceptor in Drosophila. 61. Giebultowicz, J.M., Stanewsky, R., Hall, J.C., and Hege, D.M. (2000). Cell 95, 681–692. Transplanted Drosophila excretory tubules maintain circadian clock 37. Emery, P., So, W.V., Kaneko, M., Hall, J.C., and Rosbash, M. (1998). cycling out of phase with the host. Curr. Biol. 10, 107–110. CRY, a Drosophila clock and light-regulated cryptochrome, is a 62. Kaneko, M., and Hall, J.C. (2000). Neuroanatomy of cells express- major contributor to resetting and photosensitiv- ing clock genes in Drosophila: transgenic manipulation of the period ity. Cell 95, 669–679. and timeless genes to mark the perikarya of circadian pacemaker 38. Krishnan, B., Levine, J.D., Lynch, M.K., Dowse, H.B., Funes, P., Hall, neurons and their projections. J. Comp. Neurol. 422, 66–94. J.C., Hardin, P.E., and Dryer, S.E. (2001). A new role for cryp- 63. Beaver, L.M., Rush, B.L., Gvakharia, B.O., and Giebultowicz, J.M. tochrome in a Drosophila circadian oscillator. Nature 411, 313–317. (2003). Noncircadian regulation and function of clock genes period 39. Ivanchenko, M., Stanewsky, R., and Giebultowicz, J.M. (2001). Cir- and timeless in oogenesis of Drosophila melanogaster. J. Biol. cadian photoreception in Drosophila: functions of cryptochrome in Rhythms 18, 463–472. peripheral and central clocks. J. Biol. Rhythms 16, 205–215. 64. Stanewsky, R., Jamison, C.F., Plautz, J.D., Kay, S.A., and Hall, J.C. 40. Lin, F.J., Song, W., Meyer-Bernstein, E., Naidoo, N., and Sehgal, A. (1997). Multiple circadian-regulated elements contribute to cycling (2001). Photic signaling by cryptochrome in the Drosophila circa- period gene expression in Drosophila. EMBO J. 16, 5006–5018. dian system. Mol. Cell. Biol. 21, 7287–7294. 65. Brandes, C., Plautz, J.D., Stanewsky, R., Jamison, C.F., Straume, 41. Dissel, S., Codd, V., Fedic, R., Garner, K.J., Costa, R., Kyriacou, M., Wood, K.V., Kay, S.A., and Hall, J.C. (1996). Novel features of C.P., and Rosato, E. (2004). A constitutively active cryptochrome in Drosophila period Transcription revealed by real-time luciferase Drosophila melanogaster. Nat. Neurosci. 7, 834–840. reporting. Neuron 16, 687–692. 42. Busza, A., Emery-Le, M., Rosbash, M., and Emery, P. (2004). Roles 66. Krishnan, B., Dryer, S.E., and Hardin, P.E. (1999). Circadian rhythms of the two Drosophila CRYPTOCHROME structural domains in cir- in olfactory responses of Drosophila melanogaster. Nature 400, cadian photoreception. Science 304, 1503–1506. 375–378. Review R722

67. Levine, J.D., Funes, P., Dowse, H.B., and Hall, J.C. (2002). 91. Saunders, D.S. (1997). circadian rhythms and photoperi- Advanced analysis of a cryptochrome mutation’s effects on the odism. Invert. Neurosci. 3, 155–164. robustness and phase of molecular cycles in isolated peripheral 92. Sweeney, B.M., and Hastings, J.W. (1960). Effects of temperature tissues of Drosophila. BMC Neurosci. 3, 5. upon diurnal rhythms. Cold Spring Harb. Symp. Quant. Biol. 25, 68. Peng, Y., Stoleru, D., Levine, J.D., Hall, J.C., and Rosbash, M. 87–104. (2003). Drosophila free-running rhythms require intercellular com- 93. Collins, B.H., Rosato, E., and Kyriacou, C.P. (2004). Seasonal munication. PLoS Biol. 1, E13. behavior in Drosophila melanogaster requires the photoreceptors, 69. Stoleru, D., Peng, Y., Agosto, J., and Rosbash, M. (2004). Coupled the circadian clock, and phospholipase C. Proc. Natl. Acad. Sci. oscillators control morning and evening locomotor behaviour of USA 101, 1945–1950. Drosophila. Nature 431, 862–868. 94. Majercak, J., Chen, W.F., and Edery, I. (2004). Splicing of the period 70. Grima, B., Chelot, E., Xia, R., and Rouyer, F. (2004). Morning and gene 3′-terminal intron is regulated by light, circadian clock factors, evening peaks of activity rely on different clock neurons of the and phospholipase C. Mol. Cell. Biol. 24, 3359–3372. Drosophila brain. Nature 431, 869–873. 95. Zhao, J., Kilman, V.L., Keegan, K.P., Peng, Y., Emery, P., Rosbash, 71. Myers, E.M., Yu, J., and Sehgal, A. (2003). Circadian control of eclo- M., and Allada, R. (2003). Drosophila clock can generate ectopic cir- sion: interaction between a central and peripheral clock in cadian clocks. Cell 113, 755–766. Drosophila melanogaster. Curr. Biol. 13, 526–533. 96. Hardin, P.E. (2004). Transcription regulation within the circadian 72. Claridge-Chang, A., Wijnen, H., Naef, F., Boothroyd, C., Rajewsky, clock: the E-box and beyond. J. Biol. Rhythms 19, 348–360. N., and Young, M.W. (2001). Circadian regulation of gene expres- sion systems in the Drosophila head. Neuron 32, 657–671. 73. Ceriani, M.F., Hogenesch, J.B., Yanovsky, M., Panda, S., Straume, M., and Kay, S.A. (2002). Genome-wide expression analysis in Drosophila reveals genes controlling circadian behavior. J. Neu- rosci. 22, 9305–9319. 74. Lin, Y., Han, M., Shimada, B., Wang, L., Gibler, T.M., Amarakone, A., Awad, T.A., Stormo, G.D., Van Gelder, R.N., and Taghert, P.H. (2002). Influence of the period-dependent circadian clock on diurnal, circadian, and aperiodic gene expression in Drosophila melanogaster. Proc. Natl. Acad. Sci. USA 99, 9562–9567. 75. McDonald, M.J., and Rosbash, M. (2001). Microarray analysis and organization of circadian gene expression in Drosophila. Cell 107, 567–578. 76. Ueda, H.R., Matsumoto, A., Kawamura, M., Iino, M., Tanimura, T., and Hashimoto, S. (2002). Genome-wide transcriptional orchestra- tion of circadian rhythms in Drosophila. J. Biol. Chem. 277, 14048–14052. 77. McCabe, C., and Birley, A. (1998). Oviposition in the period geno- types of Drosophila melanogaster. Chronobiol. Int. 15, 119–133. 78. Konopka, R.J., and Benzer, S. (1971). Clock mutants of Drosophila melanogaster. Proc. Natl. Acad. Sci. USA 68, 2112–2116. 79. Sakai, T., Tamura, T., Kitamoto, T., and Kidokoro, Y. (2004). A clock gene, period, plays a key role in long-term memory formation in Drosophila. Proc. Natl. Acad. Sci. USA 101, 16058–16063. 80. Nitabach, M.N., Blau, J., and Holmes, T.C. (2002). Electrical silenc- ing of Drosophila pacemaker neurons stops the free-running circa- dian clock. Cell 109, 485–495. 81. Renn, S.C., Park, J.H., Rosbash, M., Hall, J.C., and Taghert, P.H. (1999). A neuropeptide gene mutation and ablation of PDF neurons each cause severe abnormalities of behavioral circadian rhythms in Drosophila. Cell 99, 791–802. 82. Park, J.H., Helfrich-Forster, C., Lee, G., Liu, L., Rosbash, M., and Hall, J.C. (2000). Differential regulation of circadian pacemaker output by separate clock genes in Drosophila. Proc. Natl. Acad. Sci. USA 97, 3608–3613. 83. de Belle, J.S., and Heisenberg, M. (1994). Associative odor learning in Drosophila abolished by chemical ablation of mushroom bodies. Science 263, 692–695. 84. Helfrich-Forster, C. (1995). The period clock gene is expressed in central nervous system neurons which also produce a neuropeptide that reveals the projections of circadian pacemaker cells within the brain of Drosophila melanogaster. Proc. Natl. Acad. Sci. USA 92, 612–616. 85. Helfrich-Forster, C., Tauber, M., Park, J.H., Muhlig-Versen, M., Schneuwly, S., and Hofbauer, A. (2000). Ectopic expression of the neuropeptide pigment-dispersing factor alters behavioral rhythms in Drosophila melanogaster. J. Neurosci. 20, 3339–3353. 86. Johnson, E.C., Bohn, L.M., Barak, L.S., Birse, R.T., Nassel, D.R., Caron, M.G., and Taghert, P.H. (2003). Identification of Drosophila neuropeptide receptors by G protein-coupled receptors-beta- arrestin2 interactions. J. Biol. Chem. 278, 52172–52178. 87. Helfrich-Forster, C. (2000). Differential control of morning and evening components in the activity rhythm of Drosophila melanogaster–sex-specific differences suggest a different quality of activity. J. Biol. Rhythms 15, 135–154. 88. Pittendrigh, C.S., and Daan, S. (1976). A functional analysis of pace- makers in nocturnal rodents. V. Pacemaker structure: A clock for all seasons. J. Comp. Physiol. 106, 333–355. 89. Majercak, J., Sidote, D., Hardin, P.E., and Edery, I. (1999). How a cir- cadian clock adapts to seasonal decreases in temperature and day length. Neuron 24, 219–230. 90. Tanoue, S., Krishnan, P., Krishnan, B., Dryer, S.E., and Hardin, P.E. (2004). Circadian clocks in antennal neurons are necessary and suf- ficient for olfaction rhythms in Drosophila. Curr. Biol. 14, 638–649.