dper, Clk, EG:155E2.4
transcription factor - pas domain protein - regulates the diurnal photoperiod response of adult flies via a negative feedback loop - The Cycle/Clock complex binds to the promoter activating transcription - transcription is repressed when Per protein interact directly with the Cycle/Clock complex - After the sun rises Per molecules degrade, thereby releasing the repression of the Cycle/Clock complex resulting in resumption of gene transcription
Gene model reviewed during 5.45
Forms a heterodimer with timeless (TIM); the complex then translocates into the nucleus. A proportion of the protein exists as homodimer.
Phosphorylated with a circadian rhythmicity, probably by the double-time protein (dbt). Phosphorylation could be implicated in the stability of per monomer and in the formation of heterodimer per-tim.
Contains a remarkable run of alternating Gly-Thr residues which is polymorphic in length. At least three types of Gly-Thr length exist in the natural population, (Gly-Thr)23 (shown here), and two major variants (Gly-Thr)17 and (Gly-Thr)20. This Gly-Thr stretch is implicated in the maintenance of circadian period at different temperatures. Deletion of the repeat leads to a shortening of the courtship song cycle period, and thus could be important for determining features of species-specific mating behavior.
Mutations in the PAS domain result in longer circadian rhythms and courtship song (PERL mutation) or makes the flies arrhythmic (PER01 mutation).
Click to get a list of regulatory features (enhancers, TFBS, etc.) and gene disruptions (point mutations, indels, etc.) within or overlapping Dmel\per using the Feature Mapper tool.
Expression is strongly enriched in the ventral midline at embryonic stage 16.
Expression oscillates with the circadian cycle and peaks at ZT16, with the peak becoming less pronounced as flies age. Both the relative amount of mRNA and the strength of the peak are less in bodies compared to heads.
Expressed cyclically in the adult fat body.
RT-PCR analysis indicates that the levels of the per transcript cycle when in light:dark conditions, in anti-phase with the levels of the Clk transcript. per transcript levels increase gradually from ZT02 to ZT14, and decrease after that until ZT20.
per is expressed in adult oenocytes and shows a sinusoidal expression pattern with a peak at night and a trough during the day.
per transcripts are expressed in photoreceptor cells in the eye. They are also expressed in a wide region between the optic lobe and the central brain which includes the lateral neurons. They are expressed in the same pattern as dco transcripts.
per transcript levels in adult flies are not affected by exposure to light.
Comment: circadian oscillating expression
Comment: circadian oscillating expression
Comment: circadian oscillating expression
Comment: circadian oscillating expression
Expression in photoreceptors oscillates with the circadian cycle, with the oscillation becoming less pronounced in 50-day-old flies compared to 5-day old flies. Levels of per stayed constant between young and old flies in Malpighian tubules, hindgut, and fat body.
per protein starts being expressed in the embryonic brain at stage 12, extending gradually to 130 neurons at stage 16; a hundred of those also express Scer\GAL4per.PK. per-positive neurons are found throughout the brain but with more cells located caudally. Around 20 cells co-express per and tim. These cells are located in the lateral protocerebrum, close to where the lateral and dorsomedial larval clock neurons are found. Expression of per in the ventral nerve cord starts at stage 12 in 2-3 neurons per segment close to the midline. The number of labeled cells increases until stage 16, when per is detected in 8 midline cells and 11 pairs of lateral cells per segment. Co-expression with en is found in 3 VUM interneurons and one UMI interneuron, four other midline neurons and two lateral cells per hemisegment.
per is detected in follicle cells from oogenesis stage S2 and is not detected in stage 10 or later egg chambers. It is not expressed in every cell, resulting in a patchy expression pattern.
per protein shows robust circadian cycling in the pacemaker neuron of wildtype flies.
Observed at ZT00, 06, 12 and 18, per protein shows oscillation in the LP neurons and DN1a neurons under a 12:12 LD cycle. per protein is found in the nucleus at ZT23. Expression of per is variable in intensity in the DN1p neurons, the DN2 neurons and DN3 neurons at ZT23.
per protein shows strong oscillations under LD conditions, with a peak in the night and a gradual decrease at lights on. per protein is not restricted to the cytoplasm, but migrates to and from the nucleus.
per is lost from within the nucleus of the l-LNvs by ZT10. It is found to be evenly distributed between the nucleus and the cytoplasm at ZT16, four hours after lights-off. It becomes predominantly nuclear by ZT19 and remains so for the rest of the dark period. In the s-LNv neurons, per is restricted to the cytoplasm until ZT16, but is seen in both the cytoplasm and nucleus by ZT18, becoming predominantly nuclear by ZT20.
per protein in the adult is expressed in the s-LNv Pdf positive, l-LNv, LNd and DN neurons. On the third instar larval stage it is expressed in the larval s-LNv Pdf neurons.
Expression levels of per cycle in Malpighian tubules in a circadian pattern.
A detailed examination of the distribution of per protein-expressing cells within the adult brain was made. Expression was detected in the ocelli, the photoreceptor cells, and throughout the central brain and the optic ganglia and the detailed locations of expressing cells were described. The most prominent immunoreactive cells are lateral cells in the cortical area between the inner margin of the medulla and the central brain neuropil which may correspond to the medullary tangential neurons. Double staining with an antibody to elav which recognizes only neurons was used to determine the nature of the per protein-expressing cells. The per protein-expressing cells in the ocelli and eyes, the lateral cells, and the dorsal-most cortex cells were shown to be neurons. The cells located at the margins of the cortex and neuropil in the optic lobes and the central brain, the cells within the lamina and central brain neuropil, and the cells in the inner chiasm are not neurons and are thought to be glia.
Comment: faint expression
Comment: Expression is observed in tissue surrounding the mouth hook.
GBrowse - Visual display of RNA-Seq signalsView Dmel\per in GBrowse 2
Please Note FlyBase no longer curates genomic clone accessions so this list may not be complete
Please Note This section lists cDNAs and ESTs that fall within the genomic extent of the gene model, which may include cDNAs and ESTs of genes within introns, or of overlapping genes. Please see GBrowse for alignment of the cDNAs and ESTs to the gene model.
For each fully sequenced cDNA the DGRC maintains various forms of the cDNA (e.g tagged or untagged) in several different host vectors for subsequent cloning and expression in Drosophila and Drosophila cell lines.
Source for identity of: per CG2647
Delayed per entry into the nucleus correlates with an increase in period length.
DNA-protein interactions: genome-wide binding profile (ChIP-chip) assayed for per protein in adult heads; see GEO_GSE32613 (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE32613).
Gene expression is increased in response to the presence of two copies of Scer\GAL4hs.PB.
Shows particularly robust cycling of transcription in adult heads, as assessed by expression analysis using high density oligonucleotide arrays with probe generated during three 12-point time course experiments over the course of 6 days.
Identified as one of 10 highest fold cycling genes as assessed by expression analysis using high density oligonucleotide arrays with probe generated from adult heads harvested over six time points over the course of a day.
per dependent molecular oscillators may have a role in the modulation of amine receptor responsiveness.
"Gene order: Overall orientation not stated: per+ CG2650- CG2658- Csat-" was stated as revision.
Light and post-transcriptional regulation play major roles in defining the temporal properties of the per and tim protein curves. The lag between mRNA and protein accumulation is unecessary for the feedback regulation of per and tim protein on per and tim transcription.
Sensitization to repeated cocaine exposures, a phenomenon also seen in humans and animal models and associated with enhanced drug craving, is eliminated in flies mutant for per, dco, Clk, and cyc but not tim.
A thermosensitive splicing event in the 3' UTR of the per mRNA plays an important role in how the circadian clock adapts to seasonally cold days. Enhanced splicing at cold temperatures advances the steady state phases of the per mRNA and protein cycles, contributing to the preferential day time activity of flies on cold days. There is a temperature-dependent switch in the molecular logic governing cycles in per mRNA levels.
Photoreceptors R1-R6 contain an autonomous circadian oscillator that can function without per mRNA cycling.
Study of per and Dpse\per fusions reveals striking phenotypic differences between transgenic flies, these difference support the idea of an intragenic coevolution between the repeat and flanking regions of the two genes.
The gene product of the dco locus regulates per protein accumulation. The function of dco may be to reduce the stability and thus the level of accumulation of monomeric per proteins, promoting the delay between per/tim transcription and the function of the per/tim protein complex, which is essential for molecular rhythmicity.
Heat pulses at all times in a daily cycle elicited dramatic and rapid decreases in the levels of per and tim proteins, the proteins can be independently degraded by heat pulses. These two modalities produce markedly different long term effects on the circadian time-keeping mechanism. Heat-induced phase delays in behavioural rhythms are accompanied by long-term delays in the per and tim biochemical oscillations. Heat pulses in the late night elicit transient and rapid increases in the speed of the per-tim cycles. The timing of per and tim mRNA cycles is perturbed by heat pulses in a manner consistent with the direction and magnitude of the behavioural phase shift.
per protein expression in the pupal prothoracic gland has been studied in vivo and in ring glands cultured in vitro.
A circadian transcriptional enhancer within a 69bp DNA fragment is identified in per upstream sequences. The enhancer drives high-amplitude mRNA cycling under light-dark-cycling or constant-dark conditions and this activity is per protein-dependent. An E-box sequence within this fragment is necessary for high-level expression, but not for rhythmic expression, indicating per mediates circadian transcription through other sequences in this fragment.
Analysis of the rhythmic expression of a per reporter construct suggests that the Malpighian tubules contain a circadian pacemaker that functions independently of the brain.
The temporal regulation of per protein and RNA products is used to evaluate the status of the oscillatory mechanism in Pka-C1 mutants and to determine the site of action on the circadian timing system that is affected by reduced levels of Pka-C1.
Quantifying rates of protein sequence divergence within and between species reveals that the Drosophila genome harbors a substantial proportion of genes with a very high divergence rate.
Nuclear run-on assays for fly heads and the in vivo transcription rate of per and tim suggest that there is an important circadian regulation at a post-transcriptional level. Results suggest this additional regulatory mode serves to ensure proper circadian fluctuations of clock gene expression.
A new regulatory element necessary for the correct temporal expression of per is identified by monitoring real-tie per expression in living individual flies carrying two different per-Ppyr\LUC transgenes.
Circadian rhythms are clearly exhibited in constant darkness even in flies reared in constant light and constant darkness as well as flies reared in light-dark cycles, but the freerunning period differs. Results suggest that the circadian clock is assembled without any cyclical photic information and light influences the developing circadian clock to alter the freerunning period.
Photic stimuli perturb the timing of the per protein and mRNA cycles in a manner consistent with the direction and magnitude of the phase shift. The tim protein interacts with per in vivo, and the association is rapidly decreased by light. The disruption of the per-tim complex in the cytoplasm is accompanied by a delay in per phosphorylation and nuclear entry and disruption in the nucleus by an advance in per phosphorylation and disappearance.
Light inhibits the level and phosphorylation status of per and tim proteins, this then delays the negative feedback circuit (in which per or the per-tim complex participates) and extends the RNA profiles. This light-mediated delay compensates for the accelerated RNA increase in per mutant strains and restores rhythms to wild-type like periodicities.
The tim and per proteins physically interact and the timing of their association and nuclear localization promotes cycles of per and tim transcription through an autoregulatory feedback loop. tim couples the per-tim pacemaker to the environment: the tim gene product is rapidly degraded on exposure to light.
Studies of per mRNA cycling and locomotor activity rhythms under different light/dark cycling regimes indicate that the per feedback loop uses lights-off as a phase reference point and suggest that per mRNA cycling is not regulated via simple negative feedback from the per protein.
The complete sequence of the Thr-Gly region is examined and reveals that the region has evolved largely by the action of deletions/duplications and point mutations. Polymorphic sites upstream and downstream of the Thr-Gly region are also examined.
The identification of a period-altering tim allele provides further evidence that tim is a major component of the clock, and the allele-specific interactions with per provide evidence that the per/tim heterodimer is a unit of circadian function.
tim and per accumulate in the cytoplasm when independently expressed in S2 cells and move to the nucleus when coexpressed. Domains of per and tim have been identified that block nuclear localisation of the monomeric proteins. In vitro protein interaction studies indicate that the sequence inhibiting the nuclear accumulation of per forms a binding site for tim. Results indicate a mechanism for controlled nuclear localisation in which suppression of cytoplasmic localisation is accomplished by direct interaction of per and tim.
The conformation of (Thr-Gly)n peptides (the conserved (Thr-Gly)n repeat in the per gene may have an important role in the temperature compensation of the circadian clock) has been analysed.
Key pacemaker neurons of the brain were examined to determine the changes of subcellular distribution of per with the time of day, per accumulates in the cytoplasm for several hours before entering the nucleus during a narrow time window. Long-period mutations cause a delay in the timing of nuclear translocation and a further delay at elevated temperature. This data indicates that regulation of per nuclear entry is critical for circadian oscillations by providing a necessary temporal delay between per synthesis and its effect on transcription.
Identification: as a protein that interacts specifically with per protein in a yeast two-hybrid assay.
The entire arborisation pattern of per-containing pacemaker cells is revealed by immunostaining with crustacean pigment-dispersing hormone (PDH) antiserum. The arborisations of these neurons are appropriate for the modulation of the activity of many neurons and they might interact with per containing glial cells.
Temperature compensation of circadian period may be due in part to temperature-independent PER activity, which is based on competition between inter- and intramolecular interactions with similar temperature coefficients.
tim and per interact and both are required for production of circadian rhythms. Absence of tim sequence similarity to the PER dimerization motif (PAS) indicates that direct interaction between the per and tim products would require a heterotypic protein association.
Mutations of tim lead to loss of circadian rhythms. tim suppresses circadian oscillations of per abundance and phosphorylation in light/dark cycles, depresses levels of per and blocks nuclear localisation of a per reporter gene due to a primary affect on per expression at the post-transcriptional level. Constant light has no effect on per in tim flies.
A linear relationship between temperature compensation and per Thr-Gly repeat length has been demonstrated. The major natural variants differ by units of (Thr-Gly)3, and rarer variants whose lengths fall out of phase with this pattern show more erratic temperature compensation, providing a correlation between behaviour and protein structure. Interspecific comparisons reveal a co-evolutionary process between the Thr-Gly region and flanking regions.
The tim RNA oscillations are dependent on the presence of per and tim proteins. The cyclic expression of tim appears to dictate the timing of per protein accumulation and nuclear localization, suggesting that tim promotes circadian rhythms of per and tim transcription by restricting per RNA and protein accumulation to separate times of day.
DNA binding coimmunoprecipitation assays studying the interaction of human Arnt with other PAS proteins demonstrates human Arnt forms heterodimers with per and sim, by means of the PAS domain, in a cooperative way.
Expression of per under the control of the gl promoter confers both behavioral and molecular rhythmicity. Expression in a few central brain cells producing gl and per product are capable of generating biological rhythms.
The per protein undergoes daily oscillations in apparent molecular mass and abundance. The mobility changes are exclusively due to multiple phosphorylation events, phosphorylation is an important determinant in the clock time keeping mechanism.
A hypothesis has been suggested that a controlled chaotic attractor may provide the central oscillator responsible for the generation of circadian and ultradian rhythms. this can be tested by seeking chaotic dynamics in systems where the controls have been disrupted.
The process of reinitiation of a 24 hour rhythmicity in individual flies reared in constant darkness (DD) is studied.
D.melanogaster females do not discriminate between males carrying D.melanogaster or D.simulans per genes, indicating that the per locus may only make a small contribution to total premating isolation between the two species.
per gene does not influence an observable locomotor behavioural phenotype in the larval stage of development.
Sequences homologous to per demonstrated in the genome of the mole rat, Spalax ehrenbergi.
Protein-protein interactions mediated through the PAS domain may be a crucial aspect of per biochemical function.
The per gene contains an amino acid motif of approximately 270 residues, termed PAS, whose function is unknown. PAS is also present in sim gene product, and in the two subunits of the mammalian dioxin receptor. Coprecipitation experiments showed that PAS functions in vitro as a novel protein dimerization motif which can mediate interactions between different members of the PAS protein family. Missense mutations in the PAS domain, including the original perL1, decrease the dimerization efficiency. In vivo experiments using transformants with tagged per coding regions suggest that dimerization also occurs in vivo.
A 1.9 kb region of per has been compared in D.melanogaster, D.simulans, D.sechellia and D.mauritiana, and reveals a complex history. D.simulans appears to be a parent species to D.sechellia and D.mauritiana, but the order of appearance of the two species remains unclear. Whereas D.simulans and D.mauritiana share a large number of polymorphisms, D.sechellia shows very little variation.
Validity of experiments that led to the conclusions that strong differences in intercellular coupling distinguished per genotypes have been called into question. A reinvestigation of dye coupling concluded that no consistent differences in intercellular coupling in salivary glands can be attributed to mutations at the per lous (Flint, Rosbash and Hall, and Spray and Siwicki, unpublished information).
Mutations altering the structure of an approximately 20 amino acid segment, surrounding the location of the pers mutation, confer short per phenotypes. Loss or lowered function may dramatically increase measured level of protein activity.
Mutants exhibit defective courtship song.
Effects of per mutant alleles on visual pigment, sensitivity and rhabdomere size in 12hr light/12hr dark cycles was measured.
Minisatellite region encoding thr-gly repeat is polymorphic in length in natural populations. Geographical analysis of this polymorphism cloned by PCR reveals a robust clinal pattern observed along a north-south axis: the higher the latitude the more likely a population is to have thr-gly20 and less likely to have thr-gly17. Length polymorphism may be maintained by thermal selection, because thr-gly region has been shown to provide thermostability to the circadian phenotype.
A 13.2kb construct cloned from Clk-bearing flies will rescue arrhythmicity of per mutants. Transformant flies have shorter than normal rhythms demonstrating that the Clk mutation is within the 13.2b per fragment.
Internally marked mosaics determined that the pacemaker location is in the brain but not exclusively in the eyes, optic lobe or the ocelli. Although the pacemaker may be paired, the function of one of them is sufficient for rhythmicity. Glial expression is sufficient for some, albeit weak, rhythmicity.
Courtship song rhythms and locomotor activity rhythms assayed in D.melanogaster carrying Dsim\per or D.melanogaster/D.simulans per gene fusions. In all cases the circadian periodicities were slightly longer than for wild type, suggesting that the 13.2kb per fragment used to make the transgenic constructs is slightly inadequate at performing wild type per function.
Phases of the evening peaks of activity under LD conditions are correspondingly earlier than normal for the short-period mutants and later than normal for long cycle durations. The morning peaks, however, move minimally under the influence of a given per variant.
Fluctuations in per mRNA are primarily controlled by fluctuations in per transcription: per mRNA has a relatively short half life, and sequences sufficient to drive per mRNA cycling are present in 1.3kb of 5' flanking sequences.
Mutagenesis of the per gene product reveals that the only the presence of a serine at residue 589 gives a 24h period but mutation of position 589 gives short period mutants. These short period mutants give rise to pacemakers with elements that fail to respond to the negative feedback loops of circadian oscillators.
Associative learning of per mutants is assayed using the classical conditioning procedure of Tully (FBrf0043081). Results lend little support to the possibility that a short-cycle oscillation plays an integral role in learning processes; neither do they indicate any general effect on learning attributable to an abnormally lengthened circadian rhythm.
Chimeric per gene constructs from D.melanogaster and D.simulans have been used to map the genetic control of their courtship song rhythm difference to a small segment of the amino acid encoding information within per.
Ectopic expression of per demonstrates that per gene action during preimaginal stages is neither necessary or sufficient for locomotor activity rhythms to be expressed in the adult. per gene function appears to be necessary for pacemaker functioning itself. By analysing the singing of D.simulans and D.melanogaster reciprocally hybrid males the genetic etiology of the song rhythm difference is due to 1-4 amino acid replacements that have occurred over evolutionary time.
per mRNA levels in the fly head undergo circadian fluctuations during both 12 hour light/12 hour dark cycles and constant darkness.
The rhythmic components of male courtship songs have been spectral analysed.
The per gene product is thought to be involved in the regulation of the cell cycle as per mutant development time differs from wild type. A role of per in timers required for conditioning is suggested as mutations can slow down the biological timers. Transformation experiments involving the Thr-Gly encoding region of per (Yu, Nature 326:765 ) suggest that the Thr-Gly residues may play a role in determining song cycles, per has 17 to 23 Thr-Gly pairs and sings with 55 second song cycles.
Fluctuations in per protein expression in the visual system and central brain of adult flies at different times of the day have been studied.
per has an indirect influence on the levels of CG2650 transcript via its own regulation of eclosion rhythm.
Comparison of the predicted protein sequence of the Thr-Gly repeat region of different D.melanogaster strains reveals a high degree of polymorphism and evolutionary plasticity. Deletion derivatives of the per gene lacking all the perfect Thr-Gly repeats indicates that the Thr-Gly region may have an important function in the courtship song phenotype.
per01 is complemented by l(1)3A and l(1)3B mutants nearby (Young and Judd, 1978; Smith and Konopka, 1981). The mutant per01 of D.melanogaster can be rescued (i.e. made to show rhythmic behavior) by transformation with a hybrid gene carrying the coding region of the D.pseudoobscura.pseudoobscura per gene (Peterson et al., 1988).
The per gene is essential for biological clock functions and determines the period length of circadian and ultradian rhythms. The per mutants are characterized by aberrant rhythms involving eclosion and locomotor activity (Konopka and Benzer, 1971) and may change the rhythmic component of the male courtship song (Crossley, 1988; Ewing, 1988; Kyriacou and Hall, 1980, 1986, 1988). These mutants also affect the rhythm of the larval heartbeat (Dowse, Ringo and Kyriacou) (Livingstone, 1981), the level of tyrosine decarboxylase (Livingstone and Tempel, 1983) and fluctuations in membrane potentials in larval salivary glands (Weitzel and Rensing, 1981), modulate intercellular junctional communication (Bargiello et al., 1987), and alter the location of neural secretory cells in the brain (Konopka and Wells, 1980). In wild-type flies the period length is about 24 hr. In general, increases in per+ dosage lead to shortened circadian rhythms and decreases lead to lengthened circadian rhythms (Baylies et al., 1987; Cote and Brody, 1986; Hamblen et al., 1986; Smith and Konopka, 1981; Smith and Konopka, 1982; Young et al., 1985). Females heterozygous for per+ and a deletion of the locus or a per01 allele show longer-than-normal periods. per flies can be classified on the basis of their circadian rhythms as: (1) Cryptic period mutants (per01, per-) which have a 10-15 hr (ultradian) period and appear arrhythmic except in special algorhythmic tests (Dowse, Hall and Ringo, 1987); (2) Long period mutants (perL), 29 hr; (3) Long-period variable mutants (perLvar), which in homozygotes or heterozygotes are arrhythmic but in combination with certain partial deletions of the per locus result in a 30-34 hr period (Konopka, 1987); (4) Short period mutants (pers), 19 hr; (5) Short period variable mutants (persvar), some flies having a 20 hr period and the others a normal 24 hr period for locomotor activity. In temperature-change experiments on pers and perL1, the locomotor activity periods were found to be nearer to 24 hr at low temperatures, but to diverge further from normal upon heating (Konopka, Pittendrigh, and Orr, 1989; Hamblen, Ewer and Hall). perL2 shows lengthening of the periods at high temperatures. The mutant types affecting circadian rhythms (per01, perL1 and pers) may cause similar kinds of changes in the rhythmic fluctuations in courtship song interpulse intervals (IPIs) of the male (Crossley, 1988; Ewing, 1988; Kyriacou and Hall, 1980; Kyriacou and Hall, 1986; Kyriacou and Hall, 1988). per01 mutants show nonrhythmic variations in the interval between pulses of wing vibration. Neural studies show that transplantation of pers brains into per01 adult hosts causes some of the hosts to be 'rescued'; i.e. to show short-period circadian rhythms for locomotor activity (Handler and Konopka, 1979). Octopamine synthesis occurs at subnormal rates in per01 brains, with a corresponding decrease in the enzyme tyrosine decarboxylase (Livingstone and Tempel, 1983); less severe decrements in tyrosine decarboxylase are found in pers and perL1 flies. Physiological studies that claimed that per mutations can affect the level of gap junctional communication among cells in a tissue (Bargiello et al., 1987). In salivary glands and that the per01 and perL1 mutations cause a lowering of the level of junctional communication have been withdrawn (Saez, Young, Baylies, Gasic, Bargiello and Spray, 1992). Mosaic analysis of pers mutants indicates that the gene influences the brain with respect to aberrant locomotor rhythms (Konopka, Wells and Lee, 1983); per01 and per02 (and, to a lesser degree, pers) are said to cause anomalous photonegative behavior in light-response tests (Palmer, Kendrick, and Hotchkiss, 1985), but in general are not defective in visual responses (phototaxis tests, optomotor behavior and electroretinogram) according to Dushay and Hall.