Adult circadian behavior in Drosophila requires developmental expression of cycle, but not period.

Abstract

Circadian clocks have evolved as internal time keeping mechanisms that allow anticipation of daily environmental changes and organization of a daily program of physiological and behavioral rhythms. To better examine the mechanisms underlying circadian clocks in animals and to ask whether clock gene expression and function during development affected subsequent daily time keeping in the adult, we used the genetic tools available in Drosophila to conditionally manipulate the function of the CYCLE component of the positive regulator CLOCK/CYCLE (CLK/CYC) or its negative feedback inhibitor PERIOD (PER). Differential manipulation of clock function during development and in adulthood indicated that there is no developmental requirement for either a running clock mechanism or expression of per. However, conditional suppression of CLK/CYC activity either via per over-expression or cyc depletion during metamorphosis resulted in persistent arrhythmic behavior in the adult. Two distinct mechanisms were identified that may contribute to this developmental function of CLK/CYC and both involve the ventral lateral clock neurons (LN(v)s) that are crucial to circadian control of locomotor behavior: (1) selective depletion of cyc expression in the LN(v)s resulted in abnormal peptidergic small-LN(v) dorsal projections, and (2) PER expression rhythms in the adult LN(v)s appeared to be affected by developmental inhibition of CLK/CYC activity. Given the conservation of clock genes and circuits among animals, this study provides a rationale for investigating a possible similar developmental role of the homologous mammalian CLOCK/BMAL1 complex.

Full text

Page 1

Adult Circadian Behavior in Drosophila Requires
Developmental Expression of cycle, But Not period
Tadahiro Goda1., Karolina Mirowska1,2., Jake Currie1., Min-Ho Kim1, Neethi Varadaraja Rao1,2, Gloribel
Bonilla1, Herman Wijnen1*
1Department of Biology, University of Virginia, Charlottesville, Virginia, United States of America, 2PhD Program in Biology, University of Virginia, Charlottesville, Virginia,
United States of America
Abstract
Circadian clocks have evolved as internal time keeping mechanisms that allow anticipation of daily environmental changes
and organization of a daily program of physiological and behavioral rhythms. To better examine the mechanisms
underlying circadian clocks in animals and to ask whether clock gene expression and function during development affected
subsequent daily time keeping in the adult, we used the genetic tools available in Drosophila to conditionally manipulate
the function of the CYCLE component of the positive regulator CLOCK/CYCLE (CLK/CYC) or its negative feedback inhibitor
PERIOD (PER). Differential manipulation of clock function during development and in adulthood indicated that there is no
developmental requirement for either a running clock mechanism or expression of per. However, conditional suppression of
CLK/CYC activity either via per over-expression or cyc depletion during metamorphosis resulted in persistent arrhythmic
behavior in the adult. Two distinct mechanisms were identified that may contribute to this developmental function of CLK/
CYC and both involve the ventral lateral clock neurons (LNvs) that are crucial to circadian control of locomotor behavior: (1)
selective depletion of cyc expression in the LNvs resulted in abnormal peptidergic small-LNvdorsal projections, and (2) PER
expression rhythms in the adult LNvs appeared to be affected by developmental inhibition of CLK/CYC activity. Given the
conservation of clock genes and circuits among animals, this study provides a rationale for investigating a possible similar
developmental role of the homologous mammalian CLOCK/BMAL1 complex.
Citation: Goda T, Mirowska K, Currie J, Kim M-H, Rao NV, et al. (2011) Adult Circadian Behavior in Drosophila Requires Developmental Expression of cycle, But Not
period. PLoS Genet 7(7): e1002167. doi:10.1371/journal.pgen.1002167
Editor: Joseph S. Takahashi, University of Texas Southwestern Medical Center, Howard Hughes Medical Institute, United States of America
Received February 21, 2011; Accepted May 19, 2011; Published July 7, 2011
Copyright: ? 2011 Goda et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: TG, KM, JC, M-HK, NVR, and HW were supported by grant R01 GM78339 from the National Institutes of Health (www.nih.gov) to HW and by a
Distinguished Young Investigator award from the University of Virginia (www.virginia.edu) to HW. The funders had no role in study design, data collection and
analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: hw9u@virginia.edu
. These authors contributed equally to this work.
Introduction
Circadian clocks are internal daily time keeping mechanisms that
allow organisms to anticipate daily environmental rhythms as well
as efficiently organize behavioral and physiological functions in a
daily schedule. The molecular mechanisms that form the basis for
circadian rhythmicity in animals involve interlocked feedback loops
controlling gene expression as well as post-translational activities
[1,2]. In both insects and mammals a circadian transcription
complex of two basic helix-loop-helix PAS domain transcription
factors promotes the rhythmic expression of several of its negative
feedback regulators. The fruit fly Drosophila melanogaster has emerged
as a model system for animal circadian clocks that is both successful
andrepresentative.Intheclock-bearingcellsofDrosophila CLOCK/
CYCLE (CLK/CYC) acts as the central circadian transcription
complex and induces peak expression of a set of transcripts
including those for the negative feedback regulators period (per),
timeless (tim), vrille (vri), and clock work orange (cwo) just after dusk [3–9].
PER and TIM proteins form a complex with the casein kinase 1e
ortholog DOUBLETIME (DBT), in which TIM helps protect PER
from destabilization by DBT-mediated phosphorylation [10–12].
PER-containing complexes enter the nucleus around midnight and
trigger repression of CLK/CYC activity [5,13–16], VRI acts as a
transcriptional repressor for the Clk gene [9,17], and CWO reduces
CLK/CYC activity by competitively binding CLK/CYC-regulated
promoter elements [4,7,8].
The circadian clock circuits are linked to synchronizing input
pathways as well as output pathways that signal time-of-day
information to downstream biological functions. The extensive
interconnectedness of the molecular circadian cycle complicates
identification of the order of its events. We reasoned that the
development of transgenic flies with conditional circadian clock
function, in which the circadian cycle could be arrested or started
at will, would help distinguish direct from indirect effects and
determine sequential steps in circadian pathways. Moreover,
transgenic flies with conditionally titratable transcription of a clock
component would allow molecular, cellular, and behavioral
circadian phenotypes to be determined over a range of expression
levels. Finally, flies with conditionally controlled clock function
would allow separation of developmental and adult functions of
clock genes. Based on these arguments we created conditionally
rhythmic transgenic Drosophila strains.
In the present study, we describe the generation of transgenic
flies in which clock function becomes conditional on account of
PLoS Genetics | www.plosgenetics.org1 July 2011 | Volume 7 | Issue 7 | e1002167

Page 2

temperature-dependent rescue of the per01or cyc01mutations or
temperature-dependent mis-expression of per. Moreover, we made
use of these flies to experimentally determine the developmental
requirements for a functional circadian clock as well as the
individual clock components PER and CYC. We confirmed and
extended previously published observations [18,19] indicating that
developmental rescue of arrhythmia in per01mutants is not needed
for restoration of circadian rhythms in adults. However, devel-
opmental mis-expression of per or failure to developmentally rescue
the cyc01mutation led to persistent adult arrhythmia. In particular,
CLK/CYC function during the pupal and pharate adult stages
was associated with adult clock function. Our results suggest
two distinct mechanisms underlying the developmental require-
ment for CLK/CYC function: (1) cyc expression contributes cell-
type-autonomously in the ventral lateral neurons (LNvs) to the
formation of peptidergic dorsal projections containing the neuro-
peptide PIGMENT DISPERSING FACTOR (PDF), which are
thought to be important for adult circadian behavior and (2)
CLK/CYC activity during development enables normal clock
gene expression rhythms in the adult LNvs.
Results
Transgenic flies with conditional rescue of per01in
clock-bearing cells
We made use of the temporal and regional gene expression
targeting (TARGET) system [20] to create transgenic flies in
which the essential clock components CYC and PER were
expressed conditionally in relevant spatiotemporal patterns. The
TARGET system combines the binary GAL4/UAS system [21]
that allows transgenic expression to be directed spatiotemporally
by a promoter of interest via the intermediate regulator GAL4
with a ubiquitously expressed GAL80tsgene, which encodes a
temperature sensitive inhibitor of GAL4. As a result, the
TARGET system permits GAL4-mediated transgenic expression
at high temperatures (e.g., 29uC), but progressively restricts it
at lower temperatures. First, we generated transgenic flies that
conditionally rescued the arrhythmic per01phenotype [22] by
introducing a GAL4-driver transgene directing expression in all
clock-bearing cells (tim(UAS)-Gal4) [9] along with a GAL4-
responsive per cDNA expression construct (UAS-per) [23] and a
transgene ubiquitously expressing GAL80ts(tubP-Gal80ts) [20] in a
per01genetic background [22] (see Figure 1A). The resulting
genotype is abbreviated, here, as per01[timP.per]ts. As expected,
clock-controlled phenotypes such as behavioral rhythmicity,
relative rhythmic power and period length were readily and
significantly modulated by environmental temperature in these
flies (Figure 1B–1D, Figures S1 and S2). Robust circadian rhythms
in locomotor activity were virtually absent at a restrictive
temperature (18uC), but rescued to varying degrees over a range
of higher temperatures (21–29uC) (Figure 1B–1D, Figures S1 and
S2). The circadian period length observed at 29uC was
significantly longer than those at 25uC, 27uC, and 28uC for
females and those at 23uC, 25uC and 28uC for males (Figure 1D,
Figure S2A, S2C; Welch test and post-hoc Games-Howell
analysis). It is noteworthy that the decrease in rhythmicity and
relative rhythmic power and the increase in circadian period
length found at the highest experimental temperature of transgenic
induction (29uC) were also observed as a result of transgenic per
over-expression in a wild-type background (see below). In
comparison with wild-type controls per01[timP.per]tsflies were
much less rhythmic at 18uC (or 29uC), but at 25uC both genotypes
showed comparable percentages of rhythmic, weakly rhythmic,
and arrhythmic flies (Figure S3). At permissive temperatures the
most consistent difference in the behavior of per01[timP.per]tsflies
relative to wild-type controls was a significantly longer circadian
period length increased by 2 h or more (Figure S3B–S3D).
Next, we examined clock-controlled molecular responses
in per01[timP.per]tsflies released from restrictive (17 or 18uC) to
permissive conditions (25uC). As expected, transgenic per expres-
sion was strongly induced in adult fly heads following this
transition (Figure S4A, S4B). In addition, the Clk, tim, vri, cwo,
Par-domain Protein 1 (Pdp1), and Slow-poke binding protein (Slob) clock-
controlled transcripts showed relative expression responses that
appeared consistent with their circadian phase relationships in
wild-type heads. Nevertheless, the amplitude of the observed
expression responses in clock-controlled genes was reduced
relative to previously reported amplitudes of circadian oscillation
in wild-type heads [4,7–9,24–27]. Thus, upon transfer to
permissive conditions, rescue of molecular circadian oscillations
in adult per01[timP.per]ts
heads, unlike behavioral rhythms,
appeared to be incomplete. This discrepancy might be explained
by a selective restoration of high-amplitude clock gene expression
rhythms in clock neurons. We, therefore, examined circadian
transcript responses in dissected adult brains of per01[timP.per]ts
flies released under permissive conditions. However, molecular
amplitudes in adult brains were comparable to those previously
seen in adult heads (cf Figure S4A and S4C) suggesting incomplete
restoration of molecular circadian rhythms in both peripheral
clocks and the neural clock circuit. Since different time points in
the Northern and Quantitative Reverse Transcriptase PCR (qRT-
PCR) experiments of Figure S4 come from different samples of
individual flies incomplete synchrony in the experimental
population may have also contributed to the detection of relatively
shallow transcript rhythms.
Adult circadian locomotor behavior does not require a
functioning circadian clock or rescue of per01during prior
development
To test if developmental expression of per in clock-bearing cells
was required for adult clock function, we raised per01[timP.per]ts
Author Summary
The fruit fly Drosophila melanogaster is an excellent model
system for studying the internal circadian clocks that
animals use for daily time keeping. Since clocks exist and
function in animals not only in adults, but also during prior
development, the question arises if and how adult cir-
cadian rhythms depend on developmental clock circuits
and components. To address this question we created
transgenic flies in which the essential clock components
CLOCK/CYCLE (CLK/CYC) and PERIOD (PER) can be manip-
ulated via environmental temperature. Stopping the clock
during development by depleting the negative regulator
PER did not prevent restoration of circadian time keeping
in the adult. However, a developmental arrest of the clock
due to either depletion of the positive regulator CYC or
overproduction of PER resulted in a persistent loss of
clock-controlled behavior function in adults. Taken togeth-
er, these observations indicate that adult clock function
developmentally requires activity of the CLK/CYC transcrip-
tion complex rather than a ticking clock. Based on the
behavioral, molecular, and anatomical consequences of
inhibiting CLK/CYC in circadian pacemaker neurons, we
propose that the developmental requirement maps to
these cells. It will be interesting to determine whether there
is a comparable developmental requirement for the
equivalent clock genes in humans.
The Adult Clock Requires CLK/CYC Developmentally
PLoS Genetics | www.plosgenetics.org2July 2011 | Volume 7 | Issue 7 | e1002167

Page 3

Figure 1. Conditional transgenic rescue of per01arrhythmic behavior. Introduction of tim(UAS)-Gal4, UAS-per, and tubP-Gal80tstransgenes in
a per01genetic background resulted in conditional circadian clock function, with virtually no rescue of circadian locomotor behavior at the restrictive
temperature (18uC) and approximation of wild-type circadian behavior at permissive temperatures (23–29uC). (A) The lack of rescue at 18uC is
The Adult Clock Requires CLK/CYC Developmentally
PLoS Genetics | www.plosgenetics.org3July 2011 | Volume 7 | Issue 7 | e1002167

Page 4

flies at 17uC in constant light (LL) until adulthood and examined
behavioral rhythms at restrictive (17uC) and subsequent permissive
(25uC) conditions in constant darkness (DD). Consistent with the
hypothesis that rescue of circadian clock function in per01flies can
be achieved when per expression is restricted to clock-bearing cells
in the adult, we observed restoration of circadian locomotor
rhythms immediately following transition to permissive conditions
(Figure 2A, Figure S5). Although adult per01[timP.per]tsflies failed
to show strongly rhythmic locomotor behavior at the restrictive
temperature in DD, a subset of individual flies did exhibit residual
weak rhythms under these conditions (Figure S2). Nevertheless, we
do not believe that the observed behavioral rescue in adults
depends on residual clock function during the prior exposure to
the restrictive temperature for two reasons: (1) The phase of the
restored rhythms is determined by the phase of the prior switch
from restrictive to permissive conditions rather than the phase of
the light/dark transition associated with the start of the behavioral
experiment (Figure 2) and (2) our experiments included develop-
mental exposure to LL, which is associated with both behavioral
and molecular arrhythmia as well as severely reduced PER
expression levels [13,28]. Therefore, we conclude that there is
no developmental requirement for either a functioning clock
mechanism or expression of per in the clock-bearing cells in order
to allow circadian clock function in adult flies.
Over-expression of per during metamorphosis disrupts
adult circadian behavior
Given the ability to restore adult clock function from a circadian
cycle arrest due to PER depletion, we were wondering whether
circadian cycle arrests associated with excess PER expression were
equally reversible. This question was addressed experimentally with
the help of transgenic flies, in which per was conditionally over-
expressedtohighlevelsinclock-bearingcellsduetotheintroduction
of the tim(UAS)-Gal4, UAS-per, and tubP-Gal80tstransgenes in the
presenceofa wild-typepergene.Flieshomozygousfortheautosomal
tim(UAS)-Gal4 and UAS-per insertions with a single X-chromosomal
tubP-Gal80tstransgene (abbreviated as [timP.per]ts) showed condi-
tional clock function with robust rhythms, relative rhythmic power,
Figure 2. Adult circadian behavior does not require developmental rescue of per01. Circadian locomotor behavior was readily restored in
per01[timP.per]tsflies raised under restrictive conditions (17uC LL) upon transfer of adults from restrictive (17uC DD) to permissive conditions (25uC
DD). Moreover, the phase of circadian behavior was determined by the phase of transfer. (A) Double-plotted actograms representing average
locomotor activity at 17uC DD and subsequent 25uC DD. The left panel represents data for 19 female flies transferred to the permissive condition at
reference phase hour 12.5, whereas the right panel corresponds to the activity of 14 female flies switched at reference phase 3.25. Note the phase
relationship between subsequent behavioral rhythms and the time of transfer (marked by a yellow triangle). (B) The phase of the offset of circadian
locomotor activity at 25uC is plotted separately for median data from 6 groups of male and 6 groups of female flies as a function of the phase of the
17uC to 25uC switch along with trend lines and associated correlation coefficients (2-tailed test significance: p,0.005 for females, p,0.003 for males).
doi:10.1371/journal.pgen.1002167.g002
explained by inhibition of GAL4-mediated expression of transgenic per mediated by the temperature sensitive GAL4 repressor GAL80ts. At permissive
temperatures the modulating effect of GAL80tsis reduced allowing GAL4 expressed in clock-bearing cells to induce per and rescue circadian clock
function. (B,C) Flies with a y per01w; tim(UAS)-Gal4/tubPGal80ts; UAS-per/+ genotype, abbreviated as per01[timP.per]ts, raised at ambient temperature
(,23uC) were monitored for adult locomotor activity sequentially at restrictive and permissive conditions (see Materials and Methods). The large
diagrams in (B) and (C) are double-plotted actograms representing the median locomotor activity for 8 female flies. Yellow triangles indicate the time
of the temperature shifts. The small diagrams are chi-square periodograms based on the first 5 or 6 full days at the first (top) or second (bottom)
experimental condition. Circadian period lengths detected at the permissive conditions are indicated in large blue type-face. Note the absence of
strong circadian rhythms at 18uC. The considerably lengthened period and progressive weakening of rhythms observed at 29uC are likely attributable
to excessive per expression. (D) Temperature-dependent conditional rescue of the percentage of rhythmic flies, relative rhythmic power, and period
length in per01[timP.per]tsflies. Chi-square periodogram analysis of circadian locomotor behavior in DD was performed for 5-day intervals at the
indicated temperatures. The three panels show changes as a function of environmental temperature in the percentage of rhythmic flies, relative
rhythmic power (across both rhythmic and weakly rhythmic flies), and circadian period length for rhythmic flies. Error bars represent the Standard
Error of the Mean (SEM).
doi:10.1371/journal.pgen.1002167.g001
The Adult Clock Requires CLK/CYC Developmentally
PLoS Genetics | www.plosgenetics.org4July 2011 | Volume 7 | Issue 7 | e1002167

Page 5

and only marginally increased period lengths at permissive (17uC)
conditions and behavioral arrhythmia (females) or dramatically
reduced rhythms (males) at the restrictive (29uC) conditions
(Figure 3, Figure 4, Figures S6 and S7). Loss of behavioral rhythms
during prolonged exposure of adults to the restrictive condition
could be reversed by returning the flies to the permissive (17uC)
condition (Figure 3B, Figure S7). However, comparable exposure to
restrictive conditions during development resulted in irreversible
adult arrhythmia (Figure 3B, Figure 4, Figures S6 and S7) for both
genders. To identify the developmental phase of sensitivity to PER
over-expression flies were transferred from a permissive ambient
temperature (,23uC) to 29uC or vice versa at different points
during development and then analyzed for behavioral rhythmicity
as adults. When exposure to restrictive conditions occurred prior to
the pupal stage it did not obviously affect the percentages of flies
exhibiting rhythmic, weakly rhythmic or arrhythmic adult behavior
or the relative power of the detected rhythms (Figure 4, Figure S6).
However, when flies were exposed to the restrictive temperature
throughout the pupal and pharate adult stages, adult locomotor
rhythms were clearly inhibited (Figure 4, Figure S6). Therefore, it
appears that per over-expression in pupal/pharate adult clock cells
irreversibly affects adult circadian behavior. One possible explana-
tion for the observed effect of developmental per mis-expression on
adult behavior might be the persistence of abnormally high levels of
PER protein into adulthood. However, PER is known to be an
unstable protein and even a 7-d exposure to 12-h light/12-h dark/
(LD) cycles at the permissive (17uC) temperature did not allow
subsequent restoration of behavioral rhythms in DD. Moreover,
immunofluorescence analyses of clock neurons exposed to devel-
opmental PER over-expression did not reveal a continued increase
in adult PER expression. Instead, the persistent behavioral
arrhythmia of [timP.per]tsflies raised at 29uC and exposed to
17uC LD for 7 d as adults appeared to be matched by blunted
circadian rhythms of PER expression in the PDF-expressing ventral
lateral neurons (Figure 5). No gross morphological defects in clock
neurons (including LNv, LNd, and DN cell bodies and LNv
projections) were apparent in these experiments (see Figure S8).
Thus, our results indicate that excess PER activity in clock cells
during metamorphosis negatively affects both adult circadian
locomotor activity and molecular rhythms in adult clock neurons.
Depletion of cyc expression during metamorphosis
disrupts adult circadian behavior
Based on PER’s known function as a negative regulator of
CLK/CYC circadian transcription complexes the adult pheno-
types associated with developmental PER over-expression are
likely attributable to inhibition of CLK/CYC activity. We tested
this hypothesis by determining whether adult circadian locomotor
behavior required prior developmental expression of the essential
clock component CYC. To this aim we generated transgenic flies
that conditionally expressed cyc in postmitotic neurons by
combining the elavC155::Gal4 driver element [29] with UAS-cyc
[30], and tubP-Gal80tstransgenes in a cyc01background. The
resulting flies, here referred to as cyc01[elav.cyc]ts, showed
conditional rescue of rhythmic adult locomotor activity when
raised at the permissive temperature for cyc01rescue (29uC)
(Figure 6, Figure S9). Ambient temperature (,23uC), which acted
as a mostly permissive condition for per01[timP.per]tsflies (see
Figure 1D, Figure S2, above) represented a restrictive condition
for the cyc01
[elav.cyc]ts
strain. This discrepancy is likely
attributable to differences either in the amount of GAL4 protein
produced in the relevant clock neurons in each of these strains or
the level of GAL4-directed transgenic expression that is required
to achieve behavioral rescue. Exposure of cyc01[elav.cyc]tsflies to
the restrictive temperature during metamorphosis, severely
affected adult behavioral rhythms at the permissive temperature
(Figure 6B, Figure 7, Figure S10). Therefore, depletion of cyc
expression during metamorphosis, indeed, phenocopies the adult
behavioral defects of per mis-expression during metamorphosis.
Selective inhibition of cyc01rescue in the PDF-expressing
clock neurons disrupts adult locomotor rhythms as well
as cell-type-specific neuro-anatomy and molecular
rhythms
To further explore the role of CLK/CYC expression in the
PDF-positive LNvs in ensuring normal adult circadian behavior
and neuro-anatomy we created flies in which rescue of cyc01in
postmitotic neurons (by elavC155::Gal4 and UAS-cyc) was selectively
blocked in the PDF-expressing LNvs with the help of a Pdf-Gal80
transgene [31] that expresses the GAL4 inhibitor GAL80
specifically in these cells. The behavioral phenotype of the
resulting transgenic flies, indicated as cyc01(elav-Pdf).cyc, consists
of an altered daily locomotor activity profile in the presence of
light/dark cycles that includes extended activity in anticipation of
lights-on, but reduced activity in anticipation of lights-off
(Figure 8A, Figure S11A) and a loss of sustained rhythmicity in
constant darkness (Figure 8A–8C; Figure S11). In contrast, control
cyc01rescue flies lacking the Pdf-Gal80 element (cyc01elav.cyc)
showed strong behavioral rhythms in constant darkness as well as
evening activity in anticipation of the lights-off transition
(Figure 8A–8C; Figure S11). The cyc01(elav-Pdf).cyc phenotype
is clearly different from that of flies with ablated PDF-expressing
LNvs or defective expression of the PDF neuropeptide [32], which
also lack consolidated rhythms in constant darkness but show the
opposite effect on anticipation of the lights-on and lights-off
transitions. The persistence of morning anticipation, which is
thought to be attributable to PDF signaling from the s-LNvs [33]
suggests that residual PDF expression and function persisted in
s-LNvs with a cyc01circadian cycle arrest. Moreover, it is insightful
to compare the behavior of cyc01(elav-Pdf).cyc flies to previously
published observations for flies, in which rescue of the per01
mutation in clock-bearing cells was blocked in the PDF-expressing
clock neurons (per01(elav-Pdf).per) [31]. The behavior reported for
per01(elav-Pdf).per flies resembles that of our cyc01(elav-Pdf).cyc
flies with respect to the persistence of morning anticipation as well
as reduced rhythmicity in constant darkness [31]. However, loss of
evening anticipation appears to be unique to the cyc01-based as
opposed to the per01-based arrest of the LNvs. Thus, cyc-depleted
LNvs seemed to delay the generation of an evening activity signal
by the neural clock circuits. It is possible that a slow rhythmic
component in the disorganized clock circuit of cyc01(elav-Pdf).cyc
flies is responsible for this delay in evening activity. Consistent with
this notion, the few (weakly) rhythmic cyc01(elav-Pdf).cyc flies
represented in Figure 8 and Figure S11 exhibited residual long
period rhythms (t for females: 25.161.24 h, n=5; t for males
24.460.83 h, n=17).
Next, we compared molecular and neuro-anatomical pheno-
types of the PDF-expressing LNvs in cyc01(elav-Pdf).cyc flies
with those of controls lacking either the Pdf-Gal80 (cyc01elav.cyc) or
the UAS-cyc transgenes (cyc01(elav-Pdf).-). Flies of these three
genotypes were raised at ambient temperature and entrained as
adults to LD cycles at 25uC. Brains were harvested 2 h prior to
lights-on (ZT22) and stained using antibodies against PDF and the
PDP1. The Pdp1 gene is a direct target gene for CLK/CYC and
mutations in Clk or cyc strongly affect PDP1 protein expression
in larvae and adults [26]. Indeed, cyc01(elav-Pdf).- flies, which
completely lack CLK/CYC function, exhibited greatly reduced
The Adult Clock Requires CLK/CYC Developmentally
PLoS Genetics | www.plosgenetics.org5July 2011 | Volume 7 | Issue 7 | e1002167