Organization of the Drosophila Circadian Control Circuit

Current Biology 18, R84–R93, January 22, 2008 ª2008 Elsevier Ltd All rights reserved

DOI 10.1016/j.cub.2007.11.061

Organization of the Drosophila Circadian Control Circuit Michael N. Nitabach1* and Paul H. Taghert2*

Molecular genetics has revealed the identities of several components of the fundamental circadian molecular oscillator — an evolutionarily conserved molecular mechanism of transcription and translation that can operate in a cellautonomous manner. Therefore, it was surprising when studies of circadian rhythmic behavior in the fruit fly Drosophila suggested that the normal operations of circadian clock cells, which house the molecular oscillator, in fact depend on non-cell-autonomous effects — interactions between the clock cells themselves. Here we review several genetic analyses that broadly extend that viewpoint. They support a model whereby the approximately 150 circadian clock cells in the brain of the fly are sub-divided into functionally discrete rhythmic centers. These centers alternatively cooperate or compete to control the different episodes of rhythmic behavior that define the fly’s daily activity profile.

Introduction Circadian behavioral rhythms in Drosophila melanogaster depend on rhythmic clock gene expression that occurs within a cohort of w150 ‘clock’ neurons, among the estimated 100,000 neurons of the fly’s CNS. Light is the primary environmental cue that entrains these 150 internal clocks and brings them into register with local time. Historically, the most accepted model of the circadian clock mechanism — the fundamental circadian oscillator — has been based on the individual cell. Nevertheless, several recent observations have suggested that the purely cell-autonomous model of the circadian mechanism may need serious revision, as neuronal interactions among multiple different pacemaker neurons in the fly brain appear to play critical roles in generating rhythmic behavior. Remarkably, these interactions may even be required to maintain proper molecular oscillations within the interacting pacemaker cells in the absence of any environmental cues that mark the passage of time. Hence, increasing attention has been paid to the distinctive properties of different individual clock cells by using novel genetic methods to target and manipulate them. Here we review recent accounts that suggest the circadian neural circuit may comprise multiple, autonomous components. These studies support a model which predicts that the distinct components contribute separately, but coordinately, to control locomotor behavior at different times of day and are differentially sensitive to environmental cues such as light and temperature.

1Department

of Cellular and Molecular Physiology, Yale University School of Medicine, New Haven, Connecticut 06520, USA. 2Department of Anatomy and Neurobiology, Washington University Medical School, Saint Louis, Missouri 63110, USA. E-mail: [email protected] (M.N.N.), [email protected] (P.H.T.)

Review

Molecular Mechanisms of Circadian Timekeeping in Drosophila Several components of the clockwork in Drosophila have been identified by molecular genetics (Figure 1). To set the scene for later discussion, we shall briefly decribe the principal molecules and how they interact to produce a daily rhythm of gene expression (for more detailed overviews see [1,2]). Each day, a dedicated set of proteins brings about orchestrated changes in their own steady-state levels and that of their mRNAs. These changes create rhythmic waves of gene and protein expression, the phases of which are specific to the different molecules, but their periods approximate 24 hours [3–10]. The positive-acting (rhythm-driving) factors within this dedicated set are two basic helix-loophelix (bHLH) regulator proteins called Clock (Clk) and Cycle (cyc) and they directly activate transcription of many genes, including period (per) and timeless (tim). The PER and TIM proteins form the negative loop in the molecular cycle. These proteins and their mRNAs display a smooth increase in levels over the course of the day: RNA levels peak early in the evening and protein levels peak around daybreak. PER and TIM undergo several posttranslational modifications (mainly phosphorylation and de-phosphorylation) which critically influence their stabilities and ability to accumulate to levels that are functional. With individual time courses [11,12], PER and TIM enter the nucleus; aided by TIM (and other proteins), PER then terminates the daily actions of CLK and CYC in transcriptional activation. Subsequent degradation of TIM and PER permits resumption of CLK–CYC-mediated transcriptional activation, thus starting another daily cycle. Important roles have also been demonstrated for membrane depolarization and intracellular Ca2+ signals in permitting and regulating, respectively, this rhythmic negative transcriptional feedback loop [13,14]. The molecular mechanisms that couple these electrochemical signaling events to circadian transcriptional regulation remain unknown. Light entrainment is mediated by many cellular and molecular pathways (see below); an important one involves the photosensitive CRYPTOCHROME (CRY) protein, which imparts light sensitivity onto TIM through direct protein interactions that lead to TIM’s regulated degradation [15]. Anatomy of Drosophila Circadian Control Circuit The number and locations of circadian clock neurons in the fly brain are traditionally determined by cytological staining for clock gene products, such as the per and tim clock gene mRNAs and their protein products (for example, [16– 18]), or for per and tim promoter-driven activities [19]. Per expression is found in many tissues and cell types. Within the brain, the clock neurons have been divided into two major groups, the lateral neurons and the dorsal neurons (reviewed in [20]). The traditional view is that lateral neurons are the key pacemaker clock neurons which play critical roles in imposing circadian structure on the daily pattern of rest and activity, and which demonstrate cell-(or cluster-)autonomous oscillator function [21,22]. In contrast, dorsal neurons have been thought to play more subtle roles in modulating

Review R85

Figure 1. Overview of the Drosophila molecular circadian oscillator. This diagram illustrates many of the essential features of the current model of the Drosophila molecular oscillator that operates within individual pacemaker cells. Expression of the per and tim genes is promoted by the heterodimeric CLK–CYC transcription factors and reaches a peak late in the day. Translation of per and tim RNAs leads to the gradual accumulation and dimerization of PER and TIM proteins within the cytoplasm. The protein levels peak in the night, during which time they separately enter the nucleus to inhibit further CLK–CYC transcriptional activity as shown.

PER

PER

PER

TIM

TIM

CYC CLK per/tim

Nucleus TIM

Cytoplasm

Lateral Neurons The 15–16 lateral neurons in each brain hemisphere are divided into three commonly recognized subgroups: the large and small ventrolateral neurons (LNVs); the 5th small ventrolateral neuron (5th small LNV) [18]; and the dorsolateral neurons (LNDs). Several lines of evidence suggest that the small LNVs may be especially important as circadian pacemakers for locomotor rhythms in constant darkness (DD), and that the large LNVs are largely unimportant in DD. For example, GAL4 promoters that genetically distinguish the large and small subgroups were used to demonstrate that PER expression is sufficient in the small, but not the large, LNVs to support rhythmic activity in per mutant flies in DD [24]. Other observations, diverse and largely consistent, support a special role for the small cells [25–31] as a dedicated oscillator regulating a morning activity peak in light–dark (LD) conditions [24,32]. The function of the large LNVs is not known, but under conditions of constant darkness their molecular oscillations cease [26,27,28,33]. Collins et al. [34] and Helfrich-Fo¨rster et al. [35] speculate that large LNVs contribute to the gating of light inputs to the circadian system. Stoleru et al. [31] suggest the molecular clocks in the large LNv (and DN2) have a different phase relationship to that of the other clock cells and hence may represent elements of a different control circuit. The 5th small LNV is identified as a PER-positive neuron residing among the lateral neuron clusters, with a cell body the size of the other four small LNVs, but which does not express the neuropeptide pigment-dispersing factor (PDF) ([18]; see below). The LND and the 5th small LNV neuronal groups are critical pacemaker neurons and are considered (in part or whole) to represent an important component of an oscillator dedicated to regulate a specific phase of daily activity — the evening activity peak of LD ([24,32]; compare [36]). Dorsal Neurons Like the lateral neuron group, the w50 doral neurons are often divided for analysis into three distinct sub-groups: DN1, DN12 and DN13. DN1s define a poorly coherent cluster of 16

PER/TIM proteins

per/tim mRNAs Relative abundance

circadian rhythmicity [23]. The newer studies reviewed here present a richer conception of the relationships between the anatomical diversity of clock neurons and their functional attributes.

0

2

4

6

8

10

12

14

16

18

20

22

24

26

Time of day Current Biology

or 17 cells in the dorsal brain. Subsets of this group have been specifically implicated in driving oscillatory behavior in constant light (LL) conditions ([37,38]; see below). The two DN2 cell bodies lie close to the terminals of the small LNVs; their functions are not known, but the details of the molecular oscillator usually places them in opposition to the majority of circadian clock neurons [18,31]. PER expression among the 30–35 DN3s alone is not sufficient to drive locomotor rhythms under constant dark conditions, but it can direct the per-dependent evening peak of activity under LD conditions [33]. One final group is worth mention here, although it does not fall neatly into either a lateral or dorsal neuron designation. The lateral posterior neurons were originally described as cells that express TIM but not PER [19]. More recently, they were confirmed as bona fide neuronal clock cells [39,40] and, together with the DN2s, lateral posterior neurons have recently been implicated specifically in temperature entrainment [41]. In summary, the network of clock neurons in Drosophila consists of w150 neurons that fall into at least seven readily-identifiable groups. Despite the ability to identify such groups and to assign control over certain specific behavioral functions, the physiological bases of these functions, and of the interactions between groups, remain largely unknown. Neurochemistry of the Drosophila Circadian Control Circuit Glutamate and three neuropeptide transmitters have been identified as candidate signaling molecules in the fly circadian system: PDF, neuropeptide F (NPF) and neuropeptide precursor-like protein 1 (NPLP1). These are each found in largely separate subsets of clock neurons.

Current Biology Vol 18 No 2 R86

Glutamate The observation that some DN1s and DN3s in the adult brain could be immunolabeled with antibodies against the Drosophila vesicular glutamate transporter suggested that they are glutamatergic [42]. In the same study, small LNvs were found to express metabotropic glutamate receptors, and knockdown of that receptor function was found to alter LD activity patterns and lengthen the period under DD. These data suggest that certain dorsal neurons have a substantial glutamatergic influence on small LNvs. PDF PDF is expressed in a very limited pattern that includes about approximately 16 neurons in the brain and in approximately six neurons in the most caudal abdominal neuromeres [43,44]. PDF neurons in the brain are all circadian clock neurons and they represent the ventral component of the lateral neuron subgroup. As mentioned, LNVs are further divided into small and large LNVs, with the small cell group regarded as critical for daily locomotor rhythms [24,45]. Signal intensity of anti-PDF immunostaining of the small LNvs cycles in intensity as a function of time-of-day [25]. This is widely considered to be evidence for a daily, gated release of PDF (although exactly when PDF is released during the day is difficult to determine). Kula et al. [46] found that measurable PDF staining rhythms are not demonstrable in certain Drosophila strains that nevertheless display robust rhythmic behavior. As mentioned, the precise contributions of the large LNVs are not well defined. PDF is required for normal entrained behavior in a 12hr:12hr LD environment; the morning peak of activity is generally absent and the evening peak is phaseadvanced in pdf null flies [30,47]. PDF is also required for normal rhythmicity in constant darkness (DD): roughly half the mutant flies are aperiodic after a few cycles and the remainder display only weak, short-period rhythms. Over-expression of PDF in certain brain regions induces arrhythmic or complex rhythmic locomotor behavior [48]. In several respects, the phenotype of pdf mutant flies resembles the phenotype of knockout mice lacking function of the VIP or VPAC2 receptor genes [49]. The PDF receptor was recently identified by three research groups [50–52]. It is a G protein-coupled receptor most similar to family B receptors, such as those for peptides such as PACAP, secretin, and VIP. The PDF receptor appears to receive most or all of the PDF signals that support daily locomotor rhythms in flies, because severe pdfr alleles phenocopy both a pdf null allele and genetic ablation of the PDFexpressing neurons [30]. PDF receptor antisera have not produced a consensus on where the receptor is expressed [50,52]; in situ hybridization analysis places gene expression in the location of the neuroendocrine Pars Intercerebralis region of the brain [51]. This region of the insect brain is a complex neuroendocrine center with functional and developmental similarities to the endocrine hypothalamus of vertebrates [53,54]. It controls diverse physiological and behavioral functions, such as growth [55], reproduction [56] and sleep [57]. NPF NPF is structurally related to vertebrate regulatory peptides of the neuropeptide Y family. In Drosophila, it is present in midgut endocrine cells and in the CNS. Among Drosophila clock neurons, NPF is found in three neurons within the LND cluster in male brains [58]. In female brains, none of the

six LND neurons expresses the neuropeptide and it has been suggested that this neurochemical dimorphism may contribute to sexually dimorphic profiles of rest:activity behavior in flies. No NPF alleles have yet been described. The NPF receptor gene has been identified [59], but as yet it is not known where it is expressed in the adult brain or what effects perturbation of its function have on circadian rhythmic behavior. NPLP1 NPLP1 was identified following a peptidomic analysis of Drosophila larval nervous system [60]. The precursor contains multiple predicted peptides, of which two have been identified by direct sequencing methods: their short names are IPNamide and MTYamide. It is strongly expressed by the Ap-let cohort of neurons in the ventral nerve cord [61,62] and moderately expressed by many other neurons in the CNS. Within the clock neuronal complement, it is found in a subset of the approximately 16 cell DN1 cluster, specifically in the DN1 pair of cells that derives from the two larval DN1 cells [39]. All but two of the adult DN1 clusters express the transcription factor GLASS and are missing in glass mutants [33,63]. The two NPLP1-expressing DN1 cells are GLASS-negative; they are further distinguished from the others by their anterior position and by survival in the glass mutant background. From these observations, Shafer et al. [39] suggested the designations ‘anterior DN1s’ (aDN1s: GLASS-negative) versus ‘posterior DN1s’ (pDN1s: GLASSpositive). There is no information as yet concerning the identity of an NPLP1 receptor(s), its potential role in the circadian control circuit, nor any phenotypic analysis of its actions. Other Transmitter Systems In addition to this evidence for transmitter expression by the different clock neurons, there are genetic, anatomical and pharmacological data that highlight several small molecules as transmitters mediating synaptic input to the circadian neural circuitry. These include acetylcholine, histamine, serotonin and g-amino-butyric acid (GABA) [64–66]. Serotonin has specifically been implicated in pathways that underlie photoentrainment [66]. Light and Temperature Input Pathways of the Drosophila Circadian Control Circuit Notwithstanding a report that light pulses to the back of the knee can induce circadian phase shifts in humans [67], it is now well established that circadian light responses in human beings are mediated solely through ocular photoreceptors [68,69]. In contrast, entraining light inputs can reach Drosophila clock neurons by three independent pathways [15]: classical phototransduction in the compound eyes and ocelli [29]; classical and non-classical phototransduction in the extra-retinal Hofbauer-Buchner (H-B) eyelets which reside behind each compound eye [33,70]; and non-classical phototransduction by the blue-light photopigment CRY [29,71]. Light information transduced by the first two phototransduction pathways reaches clock neurons via synaptic connections. H-B eyelets are photoreceptive organs which are located within the optic lobes of the brains of adult insects and send axonal projections to the accessory medulla [72– 74]. Within the accessory medulla, the axon terminals of the H-B eyelets make contact with the small LNV pacemaker clock neurons, as assessed with light microscopy techniques [75,76]. Ultrastructural analysis of the adult blowfly

Review R87

nervous system has revealed direct synaptic contacts between the homologues of the H-B eyelet and PDF-expressing LNVs [77]. The larval visual organ, the Bolwig organ, contacts the larval PDF-expressing LNs, and is thought to be the developmental precursor to the H-B eyelet [75,76]. The Bolwig organ projection to the larval LNs is important for the larval light avoidance response, and this response is gated circadianly by the LNs [78]. Environmental light information is also transmitted to the circadian clock via the compound eyes and the ocelli, but the complete anatomical pathways for receipt of these light signals remain a mystery. In addition to the synaptic transfer of light information from classical photoreceptive organs to the circadian control circuit, light information is transduced by CRY blue-light photopigment cell-autonomously in various clock neurons themselves. Loss-of-function cry b mutant flies exhibit severely damped phase responses to brief light pulses, and CRY over-expression increases phase response, suggesting that CRY protein mediates circadian light responses [29,71]. However, cry b mutant flies are still able to entrain to LD cycles [29]. Transgenic expression of wild-type CRY protein solely in the PDF-positive LNVs of cry b mutant flies leads to substantial rescue of circadian light responses, with essentially complete rescue when cry is expressed in all clock neurons. CRY expression in the photoreceptors of the compound eye is ineffective at rescuing circadian light responses of the cry b mutant [79]. Together, these studies suggest that CRY functions autonomously in clock neurons to transduce light signals into circadian phase information, and that there are additional, CRY-independent, pathways that are sufficient for entrainment. Recent studies have begun to explore the functional relationships between the anatomically and biochemically distinct light input pathways of Drosophila. Rieger et al. [80] reported the entrainment deficits from genetically disabling individual pathways. For example, cry appears more important for entrainment to short days than to long days, while the compound eyes are more important for the normal masking effects of light. Double-mutant flies have been equally informative: norpAP41; cry b double mutants, which in addition to the loss of CRY function have an impaired phototransduction cascade in the compound eye and ocelli, still entrain, albeit poorly, to LD cycles [29,70]. This suggests the existence of yet another functional entrainment pathway, one independent of both the CRY and the compound eye/ocellar pathways. Mutant glass60j (gl60j) flies lack all classical photoreceptors — those of the compound eyes and ocelli, as well as those of the H-B eyelets [70]; gl60j; cry b double-mutant flies completely lack all circadian responses to light, including entrainment [70]. Blocking synaptic outputs of the H-B eyelet in norpAP41; cry b double-mutant flies increases the deficits in entrainment [81]. These results show that the H-B eyelets provide the CRYand compound eye/ocelli-independent circadian light input pathway that persists in norpAP41; cry b double-mutant flies. Most recently, a cry null allele (cry 0) was produced by homologous recombination [82], and analysis of the null mutant phenotype suggest that the mutant CRY-B protein has some residual function. In addition, the double mutant norpAP41; cry 0 is said to exhibit more severe deficits of photoentrainment than the norpAP41; cry b double-mutant. Surprisingly, cry 0 flies display rhythmic patterns of eclosion under LD, DD and LL conditions [82], suggesting the existence of a distinct photoreceptor(s) that can entrain that rhythmic behavioral output.

Drosophila circadian rhythms can also be entrained by cyclic changes in temperature, even when light conditions are otherwise constant [83]. At a molecular level, one postulated mechanism involves temperature-dependent alternative splicing of the per transcript [84]. The phospholipase C encoded by the norpA gene also contributes to thermal entrainment [85], which suggests that a receptor-coupled transduction cascade is an additional key to the molecular entrainment mechanism. The anatomical position(s) of the precise cells which contain circadian thermosensors has not yet been mapped. There is a peripheral themosensory organ in the fly antenna which is responsible for acute orienting responses to temperature [86]. By analogy to the multiple light input pathways to the circadian clock, sites through which the clock obtains temperature input may likewise be very distributed. This possibility is underscored by demonstrations that circadian rhythms of gene expression in isolated explants of several fly tissues can be entrained by temperature cycles [85]. Hence there may exist cell-(or at least tissue-)autonomous circadian thermosensors that play roles analogous to that of CRY for circadian photosensitivity. There have been several studies of whether particular clock neuron groups in the brain play differential roles in temperature versus light entrainment of locomotor rhythms, producing mixed answers. One group found that experimentally induced weakening of molecular rhythms of the PDF-positive LNvs, by forcing constitutive per expression, greatly accelerates temperature entrainment [87]. This suggests that the (light-sensitive) LNvs also act normally to retard temperature entrainment of dorsal neurons, perhaps in favor of their entrainment by light. They also obtained evidence that temperature cycles may operate on a dedicated subset of the circadian neurons (distinct from those termed M and E cells; see below) to entrain rhythmic behavior. Another recent study, however, observed molecular oscillations in various cell groups under conditions where light and temperature cycles were synchronized (warmer in the light, cooler in the dark) or were placed out-of-phase by six hours [41]. Molecular oscillations of lateral clock neurons (LNVs and LNDs) were phase-locked to the light–dark cycle, regardless of whether the light-dark cycle was synchronized with the temperature cycle. In contrast, molecular oscillations of dorsal clock neurons (DN1, DN2, DN3 and the lateral posterior neurons) were phase shifted when the LD and temperature cycles were out-of-phase. The authors concluded that light entrainment proceeds mainly via the lateral clock neurons, while temperature entrainment proceeds mainly via the dorsal neurons. Synchronization of Multiple Oscillators: Complex Rhythms At least one important role for PDF signaling in the circadian control circuit appears to be synchronization of multiple autonomous cellular oscillators. Evidence for this comes from a variety of genetic manipulations that modify normal spatio-temporal patterns of PDF signaling in the fly brain. PER molecular cycles are normally well-synchronized among the four small LNVs; in pdf01 flies, however, they display phasedispersal over several days in constant darkness, while the LNDs display synchronized phase advancement [28]. Ectopic expression of PDF by non-clock neurons that project to the dorsomedial protocerebrum induces arrhythmia, or complex behavioral rhythmicity — simultaneous shortperiod and long-period rhythms of free-running locomotor

Current Biology Vol 18 No 2 R88

DN1

DN2

DN3 LND l-LNv 5th s-LNv LPN

s-LNv

Wild type

LD

DD rhythmic

DD

hypothesis #1

hypothesis #2 Current Biology

Figure 2. Circadian oscillators driving in LD and DD behavioral rhythms. Top: a map of the positions and number of circadian clock neurons found in the adult fly brain. Black circles represent individual cell bodies; cluster identity indicated by acronyms (see text); gray structure represents the outline of the brain. Bottom: brain maps indicate clock neurons that are thought especially important to drive the rhythmic behaviors indicated by shadings in the actograms present to the side. Blue neurons indicate E cells as indicated by two different groups [24,32]. The precise compositions of E cells are different between the two groups, based on which genetic drivers they used. This diagram presents the larger representation. Red neurons (small LNvs) indicate M cells [24,32], which are necessary for the morning peak of activity. In DD, the predominant activity peak is marked red because the M cells are necessary and sufficient for its production, although other pacemakers also normally contribute: restriction of pacemaking activity to just M cells leads to an altered phase of DD activity [24,32]. The same actogram is represented with different colors to indicate alternative hypotheses (#1 versus #2) concerning the E/M phenomena. Hypothesis #1 posits that the LD evening activity peak is singularly controlled by E cells and hence is shaded blue [24,32]. Hypothesis #2 posits that the M cells contribute to the regulation of the LD evening peak and hence it is shaded purple to symbolize the combination of blue and red regulation [91]. See text for further details.

activity in the same animal [48]. Hyperexcitation of the LNVs by expression of a slowly inactivating bacterial Na+ channel renders constant the normally rhythmic accumulation of PDF in the terminals of the small LNVs, and induces complex behavioral rhythms and a phase advance of molecular oscillations in dorsal neurons [88]. Many of these features are also observed in flies deficient in PDF receptor expression [50,51]. Taken together, these results suggest that appropriate spatio-temporal patterns of PDF release and signaling are important for proper coordination of molecular oscillations between clock neurons in the circadian control circuit. Anatomical Loci of Morning and Evening Oscillators Almost thirty years ago, it was theorized that some nocturnal rodent circadian rhythms are controlled by independent, but coupled, morning (M) and evening (E) oscillators which respond differentially to light onset and offset [36]. Drosophila are crepuscular under normal laboratory conditions, with

morning- and evening-associated peaks of activity in LD conditions (Figure 2). This activity profile raised the possibility of exploiting the fly’s genetic tractability to test the hypothesis that anatomically distinct subsets of clock neurons might embody M and E oscillators. Several groups have approached this problem and their combined results are highly congruent. Nevertheless, the different studies reveal some conclusions enough at odds to support consideration of two alternative hypotheses, which are illustrated in Figure 2. In support of hypothesis #1, two independent studies used related genetic mosaic techniques. Rosbash and colleagues [32] used cell-specific drivers to express cell-death genes in three distinct cellular patterns: first, in a relatively wide population of clock neurons, including the PDF-expressing LNVs, LNDs, and some DN1s as well as other neurons and glia; second, solely in the 22 PDF-expressing neurons that include the LNVs; and third, solely in those cells of the relatively wide population other than the PDF-expressing LNVs. Killing

Review R89

the first population severely disrupted both the morning and evening peaks of activity. Killing the second population — just the PDF-expressing LNVs — prevented display of the morning peak of anticipatory activity, while preserving a (phase-advanced) evening peak [30,32]. Killing the third population — just the non-PDF-expressing clock neurons — most severely disrupts the evening peak of activity, with a much weaker effect on the morning peak. These results suggested that a sufficient M oscillator resides in the PDFexpressing LNVs and a minimally defined E oscillator resides in some combination of the non-PDF-expressing LNDs, the 5th small LNv and perhaps also two DN1s (Figure 2, hypothesis #1). Rouyer and colleagues [24] used cell-specific drivers to express wild-type PER protein in distinct cellular patterns in the per0 arrhythmic mutant genetic background, which lacks both morning and evening anticipatory peaks. PER expression restricted to PDF-expressing neurons (including the small and large LNVs) rescued the morning peak, but not the evening peak; PER expression in small LNVs and LNDs rescued both morning and evening peaks. In addition, pdf01 flies lacking PDF in the LNVs lack morning anticipation, but exhibit very robust, albeit phase-advanced, evening anticipation [30,32]. These results indicate that the small LNVs likely represent a sufficient M oscillator and the LNDs an E oscillator (Figure 2, hypothesis #1). The work of Rosbash and colleagues [32] was also notable for introducing the first use of the GAL80 method in fly circadian studies. This transgenic technique is designed to refine the manipulation of gene expression by restricting the spatial pattern of transgene expression. The yeast GAL80 protein directly inhibits the transactivating property of the yeast GAL4 transcription pattern. Therefore, a target UAS transgene is subject to GAL4 regulation only in the absence of GAL80 [89]. For the case of Drosophila circadian analysis, the authors [32] used the promoter of the pdf neuropeptide gene to drive GAL80 in just the w20 PDF-expressing neurons. When flies transgenic for pdf-GAL80 are combined with those transgenic for a ‘pan-clock cell’ promoter driving GAL4 (such as timeless-GAL4) they were able to restrict GAL4 activation of a third (responder) transgene to all clock cells except the PDF-expressing LNvs. This refined method of gene manipulation was instrumental in providing observations leading to that group’s definition of an E cell oscillator. Hypothesis #2 derives from two additional studies [90,91], which have considered the application of the M and E hypothesis to Drosophila by analyzing the state of molecular oscillators under low LL in either a wild type or cry b mutant background. Normally, LL drives wild-type fly locomotor activity into near arrhythmia by providing a constant stimulation to the TIM degradation signaling pathway (reviewed in [15]). The cry b mutation was previously shown to circumvent this obligate LL arrhythmia, presumably because it causes a defect in light-stimulated TIM degradation [79]. Yoshii et al. [90] and Rieger et al. [91] found that under low level LL, cry b flies, and even some wild-type flies, displayed complex activity rhythms, exhibiting both a fast (w22 hour) rhythm and a slow (w25 hour) rhythm simultaneously. As with the genetic mosaic experiments, these results are best explained by the actions of two distinct brain oscillators, similar to the M (light-decelerated) and E (light-accelerated) oscillators, respectively, that were previously hypothesized by Pittendrigh and Daan [36]. Normally the peak of PER protein oscillation in all Drosophila circadian pacemakers occurs

roughly 12 hours out of phase with the main (E) peak of activity [92]. Using this 12 hour phase angle rule as a basis for correlation, Rieger et al. [91] observed that M activity correlated with small LNV molecular oscillations, while E activity correlated with molecular oscillations in LNDs and the 5th small LNV. However, they suggested the designation M should more properly refer to a main oscillator, and not simply a morning oscillator, because the M cell rhythm (oscillation in the small LNV) appeared to drive both activity rhythms. Evidence for this comes from inspection of the activity records. For example, upon transition from LD to LL conditions, both fast (M-driven) and slow (E-driven) rhythms emerged from the ‘evening’ peak of LD activity. Secondly, upon transition from LD to DD, the durable sustained activity rhythm, which is completely dependent on the PDF-expressing small LNVs, likewise derives from the evening activity peak of LD. Hence the M oscillator may contribute to both the morning and evening activity peaks of normal LD conditions (Figure 2, hypothesis #2). In this regard, the E and M oscillators of the fly may be more similar to those of the rodent in organizing a single large block of activity — the E peak — which is durable in constant darkness and may formally be analogized to the single block of rodent nocturnal activity. In summary, several studies have recently re-examined the fly’s circadian neural architecture under different environmental conditions and in genetic variants. The general conclusions suggest the proposition that the Drosophila neuronal clock cohort is best explained by models that feature multiple, interacting rhythmic centers. ‘Pacemaking’ under Different Environmental Conditions: DD versus LL In DD, Drosophila exhibits persistent circadian rhythms of locomotor activity. Several studies suggest that DD rhythmicity is driven by the PDF-expressing small LNVs (for example [13,25,30,32,45,47]). As mentioned above, the opposite constant condition, LL, normally produces aperiodic locomotor patterns; but genetic variants like cry b can display LL rhythmicity (Figure 3). This past year, two studies [37,38] have used genetic mosaics to help define which clock cell groups are responsible for such LL rhythmicity. Murad et al. [37] found that mis-expression in an otherwise wildtype background of the proteins MORGUE (MOR), an F box/ubiquitin conjugase, or PER produces strong LL behavioral rhythms (Figure 3). Limiting MOR mis-expression further, to only non-PDF-expressing clock cells, produced the same rhythmic behavior. Significantly, limiting the mis-expression to just the PDF neurons did not have this effect. This indicates that the relevant LL cellular oscillator(s) was present in non-PDF-expressing clock cells. Robust molecular oscillations in LL were only present in a subset (six or seven) of the w17 DN1 cells (green shaded cells in Figure 3). But this effect was still partially dependent on PDF signaling, suggesting a permissive contribution by M cells to this lightresistant circadian oscillator. The authors suggested MOR or PER mis-expression may act by disrupting photosensitivity in non-M cells, with an effect similar to that of a mutation within the cry gene itself. Similarly, Stoleru et al. [38] reported that mis-expression of the clock-relevant kinase SHAGGY (SGG) within non-M clock cells (using the GAL80 method), but not within M cells, also produced strong LL rhythmicity. They also found molecular oscillations limited in these experimental situations to only a subset of DN1 neurons. As above, the subtleties of the

Current Biology Vol 18 No 2 R90

DN1

DN2

DN3 LND l-LNv 5th s-LNv LPN

s-LNv

LL Wild type

LL arrhythmic

LL

cryptochrome mutant

Figure 3. Circadian oscillators driving LD and LL behavioral rhythms. Top: a map of the positions and number of circadian clock neurons found in the adult fly brain, as in Figure 1. Bottom: representative brains of three different genetic strains under constant light conditions (LL). In control (wild type) brains, LL produces arrhythmia in most flies and molecular rhythms in clock neurons are largely disrupted. In cryptochrome mutants, one or more rhythms of behavioral activity are recorded. Analysis reveals a long period rhythm (w25 hr) is correlated with molecular cycles in the blue pacemakers (putative E cells), while the short period activity rhythm (w22 hr) is correlated with molecular cycles in the red pacemakers (M cells). More robust examples of split activity rhythms in LL are found in [40,91]. A different set of clock neurons is implicated in flies that are LLrhythmic due to mis-expression of mor, or per, or sgg [37,38]. The durable rhythmic activity in LL here is shaded green to correspond to sustained molecular cycles in a subset of DN1 neurons. It is not known if the precise DN1 subsets are the same between the two studies. The M cells are shown with slight shading to highlight their notable contribution to this LL rhythmicity (not sufficient, but necessary in part or whole). See text for further details.

LL rhythmic LL

circadian neuronal control circuit are apparent: LL rhythmicity produced by over-expression of SGG within non-PDF-expressing cells was completely dependent on PDF produced by signals from PDF-expressing cells. The conclusion therefore follows that certain non-PDF-expressing cells drive the LL rhythm, but with a permissive (non-pacemaking) contribution from the PDF-expressing cells. These are landmark studies because they prove the fly brain contains robust, behaviorally relevant circadian oscillators that are distinct from the PDF-expressing small LNVs. These additional oscillators reside among DN1 neurons, but unlike the small LNVs, their ability to serve as wholly autonomous pacemakers is still uncertain. Drosophila do not normally experience constant light conditions, so what does this demonstration of cellular rhythmicity in subsets of the DN1 cluster tell us about how the circadian neuronal control circuit normally operates? Here are two possibilities. First, the genetic mosaic studies suggest that different clock neurons have intrinsically different mechanisms of circadian photosensitivity. These physiological differences could

translate into distinct contributions by the separate oscillators over the course morgue or sgg of a single day in normal LD conditions. mis-expression Second, the results lend support to the novel hypothesis that the separate subsets of clock cells may differentially direct the orchestration of rhythmic LL behavior under the differing photorhythmic periods that are experienced in different seasons. Thus, Stoleru et al. [38] Current Biology suggest that seasonal variations in the daily profile of rhythmic activity may be explained by light-driven hierarchical alternations between M and E cells, each competing and potentially able to claim responsibility for setting behavioral pace at different times of year. Open Issues/Future Directions As we consider the emerging picture of multiple oscillators operating within the network of the Drosophila circadian circuitry, we see evidence for simplicity, but also for complexity. The simplicity is inherent in a model that features two peaks of activity and two alternating oscillator centers — the E and M cell groups. Here, the power of Drosophila genetics and the cellular resolution offered by its relatively smaller nervous system combine to produce a compelling hypothesis. This model has captured the imagination and focus of the Drosophila circadian community and generated many innovative experimental designs. But the complexities presented by these studies are equally intriguing, and they make clear there is still much to learn. At an important level, the devil remains (as always) in the details; for example,

Review R91

precisely which cells produce the E oscillation? While M cells are defined with clarity and enjoy nearly unanimous consensus, E cell constituents are not well-enumerated because we lack a robust E cell promoter that is comparable to the highly restrictive pdf gene (M cell) promoter. Additionally, the E cell function may in fact be distributed across several anatomically different cell groups, more than one of which is able to pass genetic tests of ‘‘E oscillator’’ sufficiency. Much more information is needed to understand the relationship between, on the one hand, the LND and the 5th small LNV (thought capable of representing an/the E oscillator) and the newly identified DN1 subset which constitute a behaviorally relevant oscillator in LL. Likewise, the assignment of specific clock groups as having exclusive provenance over certain peaks of activity (for example, E cells as controllers of the evening peak) may require re-consideration with future experiments. As described above, there is evidence that M cells control the morning peak, but they also seem to contribute substantially to the evening peak, and are indispensable for robust LL rhythmicity despite taking a back seat in this context to pacemakers among the DN1 cell group. In addition, E cells can generate morning activity in the absence of a clock in the M cells [32]. Thus, different experimental conditions reveal flexibility in the relative drive and hierarchical order by which these neuronal groups organize their outputs. These results in Drosophila are reminiscent of earlier work demonstrating flexibility in the phasing of per molecular oscillations in left versus right halves of the mammalian suprachiasmatic nucleus [93]. In all, this suggests that a network of circadian pacemakers may be adaptive because its constituent neurons can produce alternative synaptic relationships to promote multiple and distinct patterned outputs. This fundamental hypothesis of extensive network flexibility was first elaborated from studies of modulation in the physiology of the crustacean stomatogastric nervous system (reviewed in [94]). We are just beginning to understand the environmental and physiological factors that determine which of these coordinated functional relationships dominates the fly’s circadian circuit in any given situation. The application of sophisticated molecular genetic tools, such as the mosaic analyses featured here, has provided solid experimental evidence to reveal the existence of multiple, independent circadian oscillators in the Drosophila neural clock circuit. Additional technical approaches will offer useful complementary information about the properties of the various clock neurons. For example, how do these pacemakers differ with respect to patterns of action potential generation and when do they become most active? Here application of traditional micropipette physiology to identified Drosophila clock neurons (for example [95]) could help describe electrogenic variations that may occur over the course of the day and which may be coupled to a clock neuron’s molecular oscillation. Likewise, at what times of day do clock neurons receive synaptic and hormonal inputs that modulate or trigger their intrinsic activity? Here the application of modern imaging methods will be useful to capture the dynamics of intracellular signaling events in real-time (for example, calcium dynamics [96]) to reveal receipt of patterned inputs from other neurons or endocrine centers. The neurobiology of the Drosophila circadian system has clearly turned a corner in defining key issues for future exploration. A consensus now exists that multiple oscillators, displaying flexible interactions, are essential features of a robust

circadian control circuit. It is also now apparent that the prevalent model of the essential, self-sustaining circadian oscillator — relying solely on autonomous molecular rhythms of gene expression in single neurons — is incomplete. Studies in both mammalian and Drosophila neurons indicate that neuronal membrane activity and intercellular communication are essential for robust molecular rhythms ([13,28,97–100], but see also [101]). It thus may be that the relevant and essential unit of self-sustaining oscillations in vivo is not a molecular rhythm of gene expression in a single neuron, but is rather an integrated, cellular (or perhaps even multi-cellular or network) phenomenon. With these things established, further progress will now require attacking the intrinsic cellular physiology of the different clock cell classes and their physiological interactions. Simply put, we need a better understanding of how these w150 diverse pacemaker neurons operate as neurons. Acknowledgments Work in the laboratory of M.N.N. was supported in part by the Whitehall Foundation and NINDS Grants R01NS055035 and R01NS056443. Work in the laboratory of P.H.T. is supported in part by NINDS Grant R01NS21749 and NIMH Grant R01MH67122. References 1.

Hall, J.C. (2005). Systems approaches to biological rhythms in Drosophila. Methods Enzymol. 393, 61–185.

2.

Taghert, P.H., and Lin, Y. (2005). Tick-talk: The cellular and molecular biology of Drosophila circadian rhythms. In Comprehensive Molecular Insect Science, vol 4, L.I. Gilbert, K. Iatrou, and S.S. Gill, eds. (Elsevier BV), pp. 357–395.

3.

Claridge-Chang, A., Wijnen, H., Naef, F., Boothroyd, C., Rajewsky, N., and Young, M.W. (2001). Circadian regulation of gene expression systems in the Drosophila head. Neuron 32, 657–671.

4.

McDonald, M.J., and Rosbash, M. (2001). Microarray analysis and organization of circadian gene expression in Drosophila. Cell 107, 567–578.

5.

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. Neurosci. 22, 9305–9319.

6.

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.

7.

Ueda, H.R., Matsumoto, A., Kawamura, M., Iino, M., Tanimura, T., and Hashimoto, S. (2002). Genome-wide transcriptional orchestration of circadian rhythms in Drosophila. J. Biol. Chem. 277, 14048–14052.

8.

Wijnen, H., Naef, F., Boothroyd, C., Claridge-Chang, A., and Young, M.W. (2006). Control of daily transcript oscillations in Drosophila by light and the circadian clock. PLoS Genet. 2, e39.

9.

Boothroyd, C.E., Wijnen, H., Naef, F., Saez, L., and Young, M.W. (2007). Integration of light and temperature in the regulation of circadian gene expression in Drosophila. PLoS Genet. 3, e54.

10.

Keegan, K.P., Pradhan, S., Wang, J.P., and Allada, R. (2007). Meta-Analysis of Drosophila circadian microarray studies identifies a novel set of rhythmically expressed genes. PLoS Comput. Biol. 3, e208.

11.

Shafer, O.T., Rosbash, M., and Truman, J.W. (2002). Sequential nuclear accumulation of the clock proteins period and timeless in the pacemaker neurons of Drosophila melanogaster. J. Neurosci. 22, 5946–5954.

12.

Meyer, P., Saez, L., and Young, M.W. (2006). PER-TIM interactions in living Drosophila cells: an interval timer for the circadian clock. Science 311, 226–229.

13.

Nitabach, M.N., Blau, J., and Holmes, T.C. (2002). Electrical silencing of Drosophila pacemaker neurons stops the free-running circadian clock. Cell 109, 485–495.

14.

Harrisingh, M.C., Wu, Y., Lnenicka, G.A., and Nitabach, M.N. (2007). Intracellular Ca2+ regulates free-running circadian clock oscillation in vivo. J. Neurosci. 27, 12489–12499.

15.

Ashmore, L.J., and Sehgal, A. (2003). A fly’s eye view of circadian entrainment. J. Biol. Rhythms 18, 206–216.

16.

Zerr, D.M., Hall, J.C., Rosbash, M., and Siwicki, K.K. (1990). Circadian rhythms of period protein immunoreactivity in the CNS and the visual system of Drosophila. J. Neurosci. 10, 2749–2762.

17.

Ewer, J., Frisch, B., Hamblen-Coyle, M.J., Rosbash, M., and Hall, J.C. (1992). Expression of the period clock gene within different cell types in

Current Biology Vol 18 No 2 R92

the brain of Drosophila adults and mosaic analysis of these cells’ influence on circadian behavioral rhythms. J. Neurosci. 12, 3321–3349. 18.

19.

Kaneko, M., Helfrich-Fo¨rster, C., and Hall, J.C. (1997). Spatial and temporal expression of the period and timeless genes in the developing nervous system of Drosophila, newly identified pacemaker candidates and novel features of clock gene product cycling. J. Neurosci. 17, 6745–6760. Kaneko, M., and Hall, J.C. (2000). Neuroanatomy of cells expressing clock genes in Drosophila, transgenic manipulation of the period and timeless genes to mark the perikarya of circadian pacemaker neurons and their projections. J. Comp. Neurol. 422, 66–94.

20.

Taghert, P.H., and Shafer, O.T. (2006). Mechanisms of clock output in the Drosophila circadian pacemaker system. J. Biol. Rhythms 21, 445–457.

21.

Frisch, B., Hardin, P.E., Hamblen-Coyle, M.J., Rosbash, M., and Hall, J.C. (1994). A promoter-less period gene mediates behavioral rhythmicity and cyclical per expression in a restricted subset of the Drosophila nervous system. Neuron 12, 555–570.

42.

Hamasaka, Y., Rieger, D., Parmentier, M.L., Grau, Y., Helfrich-Fo¨rster, C., and Na¨ssel, D.R. (2007). Glutamate and its metabotropic receptor in Drosophila clock neuron circuits. J. Comp. Neurol. 505, 32–45.

43.

Helfrich-Fo¨rster, 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.

44.

Park, J.H., and Hall, J.C. (1998). Isolation and chronobiological analysis of a neuropeptide pigment-dispersing factor gene in Drosophila melanogaster. J. Biol. Rhythms 13, 219–228.

45.

Helfrich-Fo¨rster, C. (1998). Robust circadian rhythmicity of Drosophila melanogaster requires the presence of lateral neurons, a brain-behavioral study of disconnected mutants. J. Comp. Physiol. [A] 182, 435–453.

46.

Kula, E., Levitan, E.S., Pyza, E., and Rosbash, M. (2006). PDF cycling in the dorsal protocerebrum of the Drosophila brain is not necessary for circadian clock function. J. Biol. Rhythms 21, 104–117.

47.

Blanchardon, E., Grima, B., Klarsfeld, A., Chelot, E., Hardin, P.E., Preat, T., and Rouyer, F. (2001). Defining the role of Drosophila lateral neurons in the control of circadian rhythms in motor activity and eclosion by targeted genetic ablation and PERIOD protein overexpression. Eur. J. Neuro. 13, 871–888.

48.

Helfrich-Fo¨rster, 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.

49.

Aton, S.J., Colwell, C.S., Harmar, A.J., Waschek, J., and Herzog, E.D. (2005). Vasoactive intestinal polypeptide mediates circadian rhythmicity and synchrony in mammalian clock neurons. Nat. Neurosci. 8, 476–483.

50.

Hyun, S., Lee, Y., Hong, S.T., Bang, S., Paik, D., Kang, J., Shin, J., Lee, J., Jeon, K., Hwang, S., et al. (2005). Drosophila GPCR Han is a receptor for the circadian clock neuropeptide PDF. Neuron 48, 267–278.

51.

Lear, B.C., Merrill, C.E., Lin, J.M., Schroeder, A., Zhang, L., and Allada, R. (2005). A G protein-coupled receptor, groom-of-PDF, is required for PDF neuron action in circadian behavior. Neuron 48, 221–227.

52.

Mertens, I., Vandingenen, A., Johnson, E.C., Shafer, O.T., Li, W., Trigg, J.S., De Loof, A., Schoofs, L., and Taghert, P.H. (2005). PDF receptor signaling in Drosophila contributes to both circadian and geotactic behaviors. Neuron 48, 213–219.

22.

Vosshall, L.B., and Young, M.W. (1995). Circadian rhythms in Drosophila can be driven by period expression in a restricted group of central brain cells. Neuron 15, 345–360.

23.

Dushay, M.S., Rosbash, M., and Hall, J.C. (1989). The disconnected visual system mutations in Drosophila melanogaster drastically disrupt circadian rhythms. J. Biol. Rhythms 4, 1–27.

24.

Grima, B., Chelot, E., Xia, R., and Rouyer, F. (2004). Morning and evening peaks of activity rely on different clock neurons of the Drosophila brain. Nature 431, 869–873.

25.

Park, J.H., Helfrich-Fo¨rster, 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.

26.

Yang, Z., and Sehgal, A. (2001). Role of molecular oscillations in generating behavioral rhythms in Drosophila. Neuron 29, 453–467.

27.

Lear, B.C., Lin, J.M., Keath, J.R., McGill, J.J., Raman, I.M., and Allada, R. (2005). The ion channel narrow abdomen is critical for neural output of the Drosophila circadian pacemaker. Neuron 48, 965–976.

28.

Lin, Y., Stormo, G.D., and Taghert, P.H. (2004). The neuropeptide pigmentdispersing factor coordinates pacemaker interactions in the Drosophila circadian system. J. Neurosci. 24, 7951–7957.

29.

Stanewsky, R., Kaneko, M., Emery, P., Beretta, B., Wager-Smith, K., Kay, S.A., Rosbash, M., and Hall, J.C. (1998). The cryb mutation identifies cryptochrome as a circadian photoreceptor in Drosophila. Cell 95, 681–692.

53.

Renn, S.C., Park, J.H., Rosbash, M., Hall, J.C., and Taghert, P.H. (1999). A pdf neuropeptide gene mutation and ablation of PDF neurons each cause severe abnormalities of behavioral circadian rhythms in Drosophila. Cell 99, 791–802.

Scharrer, B., and Scharrer, E. (1944). Neurosecretion VI. A comparison between the intercerebralis-cadiacum-allatum system of insects and the hypothalamo-hypophysial system of vertebrates. Biol. Bull. 87, 242–251.

54.

Hartenstein, V. (2006). The neuroendocrine system of invertebrates: a developmental and evolutionary perspective. J. Endocrinol. 190, 555–570.

55.

Rulifson, E.J., Kim, S.K., and Nusse, R. (2002). Ablation of insulin-producing neurons in flies: growth and diabetic phenotypes. Science 296, 1118–1120.

30.

31.

Stoleru, D., Peng, Y., Nawathean, P., and Rosbash, M. (2005). A resetting signal between Drosophila pacemakers synchronizes morning and evening activity. Nature 438, 238–242.

56.

32.

Stoleru, D., Peng, Y., Agosto, J., and Rosbash, M. (2004). Coupled oscillators control morning and evening locomotor behaviour of Drosophila. Nature 431, 862–868.

Terhzaz, S., Rosay, P., Goodwin, S.F., and Veenstra, J.A. (2007). The neuropeptide SIFamide modulates sexual behavior in Drosophila. Biochem. Biophys. Res. Commun. 352, 305–310.

57.

33.

Veleri, S., Brandes, C., Helfrich-Fo¨rster, C., Hall, J.C., and Stanewsky, R. (2003). A self-sustaining, light-entrainable circadian oscillator in the Drosophila brain. Curr. Biol. 13, 1758–1767.

Foltenyi, K., Greenspan, R.J., and Newport, J.W. (2007). Activation of EGFR and ERK by rhomboid signaling regulates the consolidation and maintenance of sleep in Drosophila. Nat. Neurosci. 10, 1160–1167.

58.

34.

Collins, B.H., Dissel, S., Gaten, E., Rosato, E., and Kyriacou, C.P. (2005). Disruption of Cryptochrome partially restores circadian rhythmicity to the arrhythmic period mutant of Drosophila. Proc. Natl. Acad. Sci. USA 102, 19021–19026.

Lee, G., Bahn, J.H., and Park, J.H. (2006). Sex- and clock-controlled expression of the neuropeptide F gene in Drosophila. Proc. Natl. Acad. Sci. USA 103, 12580–12585.

59.

Garczynski, S.F., Brown, M.R., Shen, P., Murray, T.F., and Crim, J.W. (2002). Characterization of a functional neuropeptide F receptor from Drosophila melanogaster. Peptides 23, 773–780.

60.

Baggerman, G., Cerstiaens, A., De Loof, A., and Schoofs, L. (2002). Peptidomics of the larval Drosophila melanogaster central nervous system. J. Biol. Chem. 277, 40368–40374.

61.

Verleyen, P., Baggerman, G., Wiehart, U., Schoeters, E., Van Lommel, A., De Loof, A., and Schoofs, L. (2004). Expression of a novel neuropeptide, NVGTLARDFQLPIPNamide, in the larval and adult brain of Drosophila melanogaster. J. Neurochem. 88, 311–319.

35.

Helfrich-Fo¨rster, C., Shafer, O.T., Wu¨lbeck, C., Grieshaber, E., Rieger, D., and Taghert, P.H. (2007). Development and morphology of the clockgene-expressing Lateral Neurons of Drosophila melanogaster. J. Comp. Neurol. 500, 47–70.

36.

Pittendrigh, C.S., and Daan, S. (1976). A functional analysis of circadian pacemakers in nocturnal rodents. V. Pacemaker structure: a clock for all seasons. J. Comp. Physiol. [A] 106, 333–355.

37.

Murad, A., Emery-Le, M., and Emery, P. (2007). A subset of dorsal neurons modulates circadian behavior and light responses in Drosophila. Neuron 53, 689–701.

62.

38.

Stoleru, D., Nawathean, P., Fernandez Mde, L., Menet, J.S., Ceriani, M.F., and Rosbash, M. (2007). The Drosophila circadian network is a seasonal timer. Cell 129, 207–219.

Baumgardt, M., Miguel-Aliaga, I., Karlsson, D., Ekman, H., and Thor, S. (2007). Specification of neuronal identities by feedforward combinatorial coding. PLoS Biol. 5, e37.

63.

39.

Shafer, O.T., Helfrich-Fo¨rster, C., Renn, S.C.P., and Taghert, P.H. (2006). Re-evaluation of Drosophila melanogaster’s neuronal circadian pacemakers reveals new neuronal classes and inter-class neurochemical interactions. J. Comp. Neurol. 498, 180–193.

Klarsfeld, A., Malpel, S., Michard-Vanhee, C., Picot, M., Chelot, E., and Rouyer, F. (2004). Novel features of cryptochrome-mediated photoreception in the brain circadian clock of Drosophila. J. Neurosci. 24, 1468–1477.

64.

Wegener, C., Hamasaka, Y., and Na¨ssel, D.R. (2004). Acetylcholine increases intracellular Ca2+ via nicotinic receptors in cultured PDFcontaining clock neurons of Drosophila. J. Neurophysiol. 91, 912–923.

40.

Yoshii, T., Heshiki, Y., Ibuki-Ishibashi, T., Matsumoto, A., Tanimura, T., and Tomioka, K. (2005). Temperature cycles drive Drosophila circadian oscillation in constant light that otherwise induces behavioural arrhythmicity. Eur. J. Neurosci. 22, 1176–1184.

65.

Hamasaka, Y., and Na¨ssel, D.R. (2006). Mapping of serotonin, dopamine, and histamine in relation to different clock neurons in the brain of Drosophila. J. Comp. Neurol. 494, 314–330.

66.

41.

Miyasako, Y., Umezaki, Y., and Tomioka, K. (2007). Separate sets of cerebral clock neurons are responsible for light and temperature entrainment of Drosophila circadian locomotor rhythms. J. Biol. Rhythms. 22, 115–126.

Yuan, Q., Lin, F., Zheng, X., and Sehgal, A. (2005). Serotonin modulates circadian entrainment in Drosophila. Neuron 47, 115–127.

67.

Campbell, S.S., and Murphy, P.J. (1998). Extraocular circadian phototransduction in humans. Science 279, 396–399.

Review R93

68.

Ruger, M., Gordijn, M.C., Beersma, D.G., de Vries, B., and Daan, S. (2003). Acute and phase-shifting effects of ocular and extraocular light in human circadian physiology. J. Biol. Rhythms 18, 409–419.

69.

Wright, K.P., Jr., and Czeisler, C.A. (2002). Absence of circadian phase resetting in response to bright light behind the knees. Science 297, 571.

70.

Helfrich-Fo¨rster, C., Winter, C., Hofbauer, A., Hall, J.C., and Stanewsky, R. (2001). The circadian clock of fruit flies is blind after elimination of all known photoreceptors. Neuron 30, 249–261.

96.

71.

Emery, P., So, W.V., Kaneko, M., Hall, J.C., and Rosbash, M. (1998). CRY, a Drosophila clock and light-regulated cryptochrome, is a major contributor to circadian rhythm resetting and photosensitivity. Cell 95, 669–679.

97.

72.

Hofbauer, A., and Buchner, E. (1989). Does Drosophila Have 7 Eyes? Naturwissenschaften 76, 335–336.

73.

Fleissner, G., and Frisch, B. (1993). A new type of putative non-visual photoreceptors in the optic lobe of beetles. Cell Tissue Res. 273, 435–445.

74.

Yasuyama, K., and Meinertzhagen, I.A. (1999). Extraretinal photoreceptors at the compound eye’s posterior margin in Drosophila melanogaster. J. Comp. Neurol. 412, 193–202.

99.

75.

Helfrich-Fo¨rster, C., Edwards, T., Yasuyama, K., Wisotzki, B., Schneuwly, S., Stanewsky, R., Meinertzhagen, I.A., and Hofbauer, A. (2002). The extraretinal eyelet of Drosophila: development, ultrastructure, and putative circadian function. J. Neurosci. 22, 9255–9266.

100.

76.

Malpel, S., Klarsfeld, A., and Rouyer, F. (2002). Larval optic nerve and adult extra-retinal photoreceptors sequentially associate with clock neurons during Drosophila brain development. Development 129, 1443–1453.

77.

Yasuyama, K., Okada, Y., Hamanaka, Y., and Shiga, S. (2006). Synaptic connections between eyelet photoreceptors and pigment dispersing factor-immunoreactive neurons of the blowfly Protophormia terraenovae. J. Comp. Neurol. 494, 331–344.

78.

Mazzoni, E.O., Desplan, C., and Blau, J. (2005). Circadian pacemaker neurons transmit and modulate visual information to control a rapid behavioral response. Neuron 45, 293–300.

79.

Emery, P., Stanewsky, R., Helfrich-Fo¨rster, C., Emery-Le, M., Hall, J.C., and Rosbash, M. (2000). Drosophila CRY is a deep brain circadian photoreceptor. Neuron 26, 493–504.

80.

Rieger, D., Stanewsky, R., and Helfrich-Fo¨rster, C. (2003). Cryptochrome, compound eyes, Hofbauer-Buchner eyelets, and ocelli play different roles in the entrainment and masking pathway of the locomotor activity rhythm in the fruit fly Drosophila melanogaster. J. Biol. Rhythms 18, 377–391.

81.

Veleri, S., Rieger, D., Helfrich-Fo¨rster, C., and Stanewsky, R. (2007). Hofbauer-Buchner eyelet affects circadian photosensitivity and coordinates TIM and PER expression in Drosophila clock neurons. J. Biol. Rhythms 22, 29–42.

82.

Dolezelova, E., Dolezel, D., and Hall, J.C. (2007). Rhythm defects caused by newly engineered null mutations in Drosophila’s cryptochrome gene. Genetics 177, 329–345.

83.

Wheeler, D.A., Hamblen-Coyle, M.J., Dushay, M.S., and Hall, J.C. (1993). Behavior in light-dark cycles of Drosophila mutants that are arrhythmic, blind, or both. J. Biol. Rhythms 8, 67–94.

84.

Majercak, J., Chen, W.F., and Edery, I. (2004). Splicing of the period gene 3’-terminal intron is regulated by light, circadian clock factors, and phospholipase C. Mol. Cell Biol. 24, 3359–3372.

85.

Glaser, F.T., and Stanewsky, R. (2005). Temperature synchronization of the Drosophila circadian clock. Curr. Biol. 15, 1352–1363.

86.

Sayeed, O., and Benzer, S. (1996). Behavioral genetics of thermosensation and hygrosensation in Drosophila. Proc. Natl. Acad. Sci. USA 93, 6079– 6084.

87.

Busza, A., Murad, A., and Emery, P. (2007). Interactions between circadian neurons control temperature synchronization of Drosophila behavior. J. Neurosci. 27, 10722–10733.

88.

Nitabach, M.N., Wu, Y., Sheeba, V., Lemon, W.C., Strumbos, J., Zelensky, P.K., White, B.H., and Holmes, T.C. (2006). Electrical hyperexcitation of lateral ventral pacemaker neurons desynchronizes downstream circadian oscillators in the fly circadian circuit and induces multiple behavioral periods. J. Neurosci. 26, 479–489.

89.

McGuire, S.E., Mao, Z., and Davis, R.L. (2004). Spatiotemporal gene expression targeting with the TARGET and gene-switch systems in Drosophila. Sci. STKE. 2004, pl6.

90.

Yoshii, T., Funada, Y., Ibuki-Ishibashi, T., Matsumoto, A., Tanimura, T., and Tomioka, K. (2004). Drosophila cry b mutation reveals two circadian clocks that drive locomotor rhythm and have different responsiveness to light. J. Insect Physiol. 50, 479–488.

91.

Rieger, D., Shafer, O.T., Tomioka, K., and Helfrich-Fo¨rster, C. (2006). Functional analysis of circadian pacemaker neurons in Drosophila melanogaster. J. Neurosci. 26, 2531–2543.

92.

Zerr, D.M., Hall, J.C., Rosbash, M., and Siwicki, K.K. (1990). Circadian fluctuations of period protein immunoreactivity in the CNS and the visual system of Drosophila. J. Neurosci. 10, 2749–2762.

93.

de la Iglesia, H.O., Meyer, J., Carpinou, A., Jr., and Schwartz, W.J. (2000). Antiphase oscillation of the left and right suprachiasmatic nuclei. Science 290, 799–801.

94. 95.

98.

101.

Marder, E., and Calabrese, R.L. (1996). Principles of rhythmic motor pattern generation. Physiol. Rev. 76, 687–717. Park, D., and Griffith, L.C. (2006). Electrophysiological and anatomical characterization of PDF-positive clock neurons in the intact adult Drosophila brain. J. Neurophys. 95, 3955–3960. Kim, Y.J., Zitnan, D., Galizia, C.G., Cho, K.H., and Adams, M.E. (2006). A command chemical triggers an innate behavior by sequential activation of multiple peptidergic ensembles. Curr. Biol. 16, 1395–1407. Harmar, A.J., Marston, H.M., Shen, S., Spratt, C., West, K.M., Sheward, W.J., Morrison, C.F., Dorin, J.R., Piggins, H.D., Reubi, J.C., et al. (2002). The VPAC(2) receptor is essential for circadian function in the mouse suprachiasmatic nuclei. Cell 109, 497–508. Maywood, E.S., Reddy, A.B., Wong, G.K., O’Neill, J.S., O’Brien, J.A., McMahon, D.G., Harmar, A.J., Okamura, H., and Hastings, M.H. (2006). Synchronization and maintenance of timekeeping in suprachiasmatic circadian clock cells by neuropeptidergic signaling. Curr. Biol. 16, 599–605. Yamaguchi, S., Isejima, H., Matsuo, T., Okura, R., Yagita, K., Kobayashi, M., and Okamura, H. (2003). Synchronization of cellular clocks in the suprachiasmatic nucleus. Science 302, 1408–1412. Peng, Y., Stoleru, D., Levine, J.D., Hall, J.C., and Rosbash, M. (2003). Drosophila free-running rhythms require intercellular communication. PLoS Biol. 1, E13. Welsh, D.K., Logothetis, D.E., Meister, M., and Reppert, S.M. (1995). Individual neurons dissociated from rat suprachiasmatic nucleus express independently phased circadian firing rhythms. Neuron 14, 697–706.