Download View Full PDF - Biochemical Society Transactions

Mechanistic and Functional Studies of Proteins
Organization of endogenous clocks in insects
C. Helfrich-Förster1
Institut für Zoologie, Universität Regensburg, Universitätsstrasse 31, 93040 Regensburg, Germany
Abstract
Insect and mammalian circadian clocks show striking similarities. They utilize homologous clock genes,
generating self-sustained circadian oscillations in distinct master clocks of the brain, which then control
rhythmic behaviour. The molecular mechanisms of rhythm generation were first uncovered in the fruit fly
Drosophila melanogaster, whereas cockroaches were among the first animals where the brain master clock
was localized. Despite many similarities, there exist obvious differences in the organization and functioning
of insect master clocks. These similarities and differences are reviewed on a molecular and anatomical level.
Introduction
Endogenous circadian clocks help organisms to adapt to the
daily cycling in environmental conditions. Animals possess
circadian master clocks in the brain, and insects were among
the first animals where these clocks have been localized to
specific brain areas (for a review see [1]). The brain master
clocks of insects and mammals show astonishing similarities.
They are anatomically and functionally connected to the optic
system, allowing synchronization with the environmental
light–dark cycles, and possess multiple output pathways to
control diverse endocrine, autonomic and behavioural functions (for a review see [2]). Furthermore, insects and mammals
utilize homologous clock genes generating self-sustained
circadian oscillations via interconnected negative and positive
feedback loops on the transcriptional and translational level
(for a review see [3]). The core players in the negative feedback
loop are the transcription factors CLOCK (CLK), CYCLE
(CYC, also called BMAL) and PERIOD (PER). CLK
and CYC heterodimerize and bind to E-box enhancer elements present in the upstream-regulating sequences of period
(per) thus activating per transcription. Transcription of per
is inhibited by accumulation of PER in the nucleus and its
consecutive binding to the CLK:CYC dimer. The fourth main
player in this negative feedback loop is TIMELESS (TIM)
for insects and CRYPTOCHROME (CRY1 and CRY2) for
mammals. Transcription of tim is also activated by the
CLK:CYC dimer. TIM dimerizes with PER regulating its
stability and its nuclear transport. Furthermore, TIM plays
a role in light input to the clock. In mammals, orthologues
of TIM providing a comparable function similar to insect
TIM, have not been detected; there, CRY1 and CRY2 appear
to have overtaken TIM’s functional role in the negative feedback loop.
Key words: circadian rhythm, clock gene, clock neuron, Drosophila melanogaster, pigmentdispersing factor.
Abbreviations used: CCID, CLK:CYC inhibition domain; CLD, cytoplasmic localization domain;
NES, nuclear export sequence; NLS, nuclear localization sequence; PAS, PER/ARNT/SIM; PDF,
pigment-dispersing factor.
1
email [email protected]
Homologous clock factors in different
insects
Core clock proteins
The best-characterized clock protein is PER. So far, PER
was isolated from several Dipterean [4,5] and Lepidopterean
species [6–9], two Hymenopterean species [10,11] and two
Hemipterean species [7,12] (Figure 1). PER possesses a PAS
(PER/ARNT/SIM) domain, split into two structural motifs (PAS-A and PAS-B) which facilitate protein–protein
interactions. Furthermore, it includes a CLD (cytoplasmic
localization domain), an NES (nuclear export sequence), a
C-terminal CCID (CLK:CYC inhibition domain) and two
NLSs (nuclear localization sequences), one at the N-terminus,
the other within the CCID [13] (Figure 1). These domains
are well conserved between different species, with amino acid
identities ranging from 46 to 57%. A high degree of functional
conservation of PER was shown by the ability of the silk moth
PER or housefly PER to re-establish rhythmic behaviour
in per0 mutants of Drosophila melanogaster [5,14].
TIM orthologues have been isolated from several
Diptereans [15–17] and one Lepidopterean species [18,19].
Like PER, TIM contains an NLS, NESs [20] and a C-terminal CLD; furthermore, several α-helices of still unknown
function are present. TIM and PER physically associate as
indicated in Figure 1. By directly binding to PER’s CLD
domain TIM masks this domain and enables nuclear entry
of the PER–TIM dimer via the two NLSs of PER [13,21].
The activity of TIM’s CLD is also suppressed by the physical
interaction between PER and TIM, although this CLD lies
outside the direct-binding sites of both proteins. Inhibition
of CLK:CYC occurs via the CCID of PER [13]; TIM has
no intrinsic inhibitory capacity, but appears to contribute
CLK:CYC inhibition via PER.
The CLK and CYC proteins have been isolated from few
Dipterean and Lepidopterean species (summarized in [18]).
These also possess a PAS domain with PAS-A and PAS-B
motifs and a region homologous to the CLD domain of PER
but with unknown function. A C-terminal glutamine-rich
transcriptional activation domain for activating per and tim
C 2005
Biochemical Society
957
958
Biochemical Society Transactions (2005) Volume 33, part 5
Figure 1 Schematic comparison of PER, TIM, CLK and CYC
between different species
Grey bars represent primary amino acid sequences of each protein to
scale. White boxes represent specific domains: ‘bH’, basic helix–loop–
helix (bHLH) domain; A, B, PAS-A and PAS-B motifs; C, CLD domains;
Q, poly-glutamine stretches; T, transcriptional activating domains. Black
boxes represent NLS sequences and light-grey boxes NESs. Modified
with permission from Current Biology 13: Chang, D.C. and Reppert,
S.M. ‘A novel C-terminal domain of drosophila PERIOD inhibits dCLOCK:
c 2003 Elsevier; Neuron
CYCLE-mediated transcription.’, pp. 758–762 17: Saez, L. and Young, M.W. ‘Regulation of nuclear entry of the
c 1996
Drosophila clock proteins period and timeless.’, pp. 911–920 c
Elsevier; and [18] 2003 American Society for Biochemistry and
Molecular Biology, with additional data from [9,20].
transcription is either present in CYC or CLK, depending on
the species [18].
PDF (pigment-dispersing factor) as a factor
downstream of the clock
PDF is a putative output factor of the circadian clock controlling rhythmic activity [22,23]. PDF is a neuropeptide with
18 amino acids that is highly conserved in different insects
(reviewed in [24]). In D. melanogaster its secretion appears to
be under clock control [25] and its absence due to a mutation
C 2005
Biochemical Society
results in a disturbed activity rhythm [22]. Furthermore, PDF
appears to synchronize the oscillations between different
clock neurons [26,27]. A recent study in cockroaches shows
that PDF also synchronizes ultradian oscillations in neuronal
spiking of different neurons [28].
Distribution of clock factors in the brain
Immunocytochemical studies revealed distinct neurons in the
insect brain that express the different clock proteins and
the output factor PDF. In D. melanogaster PER-, TIM-,
CLK- and CYC-positive neurons are located in the lateral
protocerebrum just at the border to the optic lobes (the socalled lateral neurons) as well as in the dorsal protocerebrum
(the so-called dorsal neurons) [2]. Two ventral groups of the
lateral neurons co-express PDF [29]. The PDF-neurons show
a wide fibre network on the surface of the optic lobe, connect
both optic lobes and run into the dorsal protocerebrum. This
arborization pattern makes them suited to transfer rhythmic
signals to other brain areas, as well as to synchronize the cycling of all clock neurons.
Various other insects including Archeognatha (bristletails),
Zygentoma (firebrats), Ephemeroptera (mayflies), Odonata
(damselflies), Plecoptera (stoneflies), Caelifera (locusts),
Blattaria (cockroaches), Heteroptera (bugs), Coleoptera
(beetles), Hymenoptera (bees), Lepidoptera (butterflies),
Trichoptera (caddisflies) and other Diptera (flies) have been
analysed for localization of their clock neurons with antibodies against PER and PDF [30–33]. Other insects were
examined exclusively for PER-positive neurons [34,35] or
PDF-positive neurons [36–38]. Similar to Drosophila, most
insects had (a) PER- and PDF-positive neurons in the lateral
protocerebrum, whereby the PDF-positive neurons showed
wide-field arborizations into the central brain as well as into
the optic lobe (Figure 2) and (b) PER-positive neurons in the
dorsal protocerebrum. In these insects the ‘lateral neurons’
were located in the proximal optic lobe (close to the
medulla) rather than in the lateral protocerebrum. This is
due to morphological differences between the very compact
fly brain and the brains of most other insects that have a pronounced optic tract. In the following, I will generally refer to
these neurons as optic lobe neurons. Most importantly, the
PDF-positive optic lobe neurons were not identical with
the PER-positive optic lobe neurons in a great majority
of insects [32]. Co-localization of PDF and PER was only
found in D. melanogaster. Another important difference
between D. melanogaster and most other insects was found
in the subcellular localization of PER: as predicted from the
above-described negative feedback loop, Drosophila PER was
entirely nuclear at certain times and cytoplasmic at other
times. In a great majority of all other insects, PER was,
however, only detectable in the cytoplasm. Exceptions were
a few small neurons in the dorsal protocerebrum of Manduca
sexta [34], in the optic lobes of Pachymorpha sexguttata [30]
and in the optic lobes of Leucophaea maderae [28] that
expressed PER in the nucleus. These differences indicate
that the clocks’ functional mechanisms differ in the various
Mechanistic and Functional Studies of Proteins
Figure 2 PER- and PDF-immunoreactive neurons in the brain
(protocerebrum) of different insects
PER is shown in grey and PDF in black. The PDF-arborization pattern
investigated for some of these insects is shown in the right-hand
panel. For cockroaches (here Pe. americana), PER-immunocytochemistry
has not yet been performed. Adapated from Závodská, R., Sauman,
I. and Sehnal, F. ‘Distribution of PER, protein, pigment-dispersing
hormon, prothoracicotropic hormone, and eclosion hormone in the
cephalic nervous system of insects.’, Journal of Biological Rhythms 18,
c 2003 Sage Publications Inc., modified with permission by
pp. 106–120, Sage Publications Inc.; and Cell Tissue Research 312 (2003) pp. 113–125,
‘Immunocytochemical distribution of pigment-dispersing hormone in the
cephalic ganglia of polyneopteran insects.’ by Sehadova, H., Sauman, I.
c 2003 Springer, with kind permission of
and Sehnal, F., Figure 4, Springer Science and Business Media.
marginata had both types of neurons only in the optic
lobe [32] (Figure 2). Interestingly, such differences were also
observed between closely related species. Another beetle, the
ground beetle Pachy. sexguttata, had both types of neurons in
the dorsal protocerebrum and the optic lobe [30]. The blow
fly Phormia regina had PER-positive cells only in the dorsal
protocerebrum, but PDF-positive neurons in the optic lobe
and the dorsal protocerebrum [32] (Figure 2), whereas the
fruit fly D. melanogaster showed PER-positive neurons in
both brain areas and PDF-positive neurons only in the optic
lobe (Figure 2).
Finally, it has to be mentioned that additional PER-positive
neurons have also been found in other brain regions, e.g. in
the suboesophageal ganglion and the pars intercerebralis of
a few insects [32,35]. Additional PDF-positive neurons have
been localized in the distal optic lobe of some apterygote and
exopterygote insects (and as an exception the ground beetle
Pachy. sexguttata) [32,36,38] (Figure 2: Siphlonurus armatus,
Locusta migratoria, Periplaneta americana).
Localization of the master clocks
insects in spite of the high structural conservation of clock
proteins (see the Conclusions section).
The so far described distribution of the clock neurons
suggests that insects may possess two anatomically connected
clock centres: (i) an optic lobe clock and (ii) a central
brain clock. However, in some insects only one of these clocks
appears to be present. The sphinx moth M. sexta showed
PER- and PDF-positive neurons only in the dorsal protocerebrum [34,39], and the goldsmith beetle Pachnoda
The distribution pattern of PER already suggests that the
master clock driving rhythmic behaviour of insects is either
located in the optic lobe, in the dorsal protocerebrum or in
both brain areas. This has been shown by sophisticated lesion
and transplantation studies as well as by genetic manipulations, whereby many of these studies had been already
performed before the distribution of PER was known [1].
The existence of a pacemaker in the dorsal protocerebrum
is especially likely for species like the giant sphinx moth
M. sexta that have no PER- and PDF-expressing cells in the
lateral protocerebrum, whereas a pacemaker in the proximal
optic lobe is predictable for beetles like Pachn. marginata
that lack PER- and PDF-expressing neurons in the dorsal
protocerebrum. Indeed, the existence of a dorsal brain clock
controlling rhythmic eclosion and flight activity was shown
for the silk moths Antherea pernyi and Hyalophora cecropia
[40], and the existence of an optic lobe pacemaker controlling
rhythmic activity and the electroretinogram of the eye was
shown for the beetles Pachy. sexguttata and Blaps gigas [41]. A
comparable optic lobe master clock exists in the cockroaches
L. maderae and Pe. americana [42–44]. In crickets, the reports
about the position of the master clock are more divergent.
The cricket Teleogryllus commodus has a master clock in the
proximal optic lobe, for Gryllus bimaculatus the master clock
may reside in the distal optic lobe and in Achaeta domesticus
it may even reside in the central brain (reviewed by [45]).
Unfortunately, not all insects in which the master clocks
have been allocated to specific brain areas (as in cockroaches
and crickets) were studied simultaneously in respect to the
distribution of PER-neurons, and several insects where PERimmunohistochemistry was performed (as bristletails, firebrats, mayflies, damselflies, stoneflies and bugs) were not analysed in respect of the localization of their master clocks.
Future studies are necessary to obtain a more coherent
picture.
C 2005
Biochemical Society
959
960
Biochemical Society Transactions (2005) Volume 33, part 5
Conclusions
The results presented indicate that the basic components
of the brain master clock are conserved throughout the
class of Insecta. The clock proteins show high homology
between different insects, suggesting that they play similar
roles in the circadian system. The conservation of NLS sequences in insect PER proteins suggests that PER nuclear
entry may be widely true, although PER has only been
detected in the cytoplasm in most insects. PER might enter the
nucleus only transiently and at low levels. Experiments in S2
cells of D. melanogaster show that a transcriptional feedback
loop can be constructed from A. pernyi CLK, BMAL, PER
and TIM proteins and the A. pernyi per gene [18], indicating
that a Drosophila-like feedback loop may form the core of
the circadian clockwork in insects. On the other hand, there
are several observations that do not speak for an obligatory
transcriptional feedback. Whereas oscillations in PER appear
to be required for oscillations in per mRNA, the converse
does not seem to be true [46]. per01 flies carrying a transgene
that constitutively expresses per mRNA in photoreceptors
nevertheless show a cycling in PER abundance in these cells.
This demonstrates that circadian cycling of per mRNA is
not necessary for PER cycling, and that the oscillator can
run solely through post-transcriptional mechanisms. Protein instability and/or rhythmic degradation of PER and
TIM are probably important features for the dynamics of
both PER and TIM cycling [47,48]. Recent observations in
Cyanobacteria show indeed that a cycling in clock protein
levels occurs when all necessary clock proteins are put
together in vitro [49]. This implies that the negative feedback
of clock proteins on their own transcription is optional. Some
insects may use it in all clock neurons, others only in certain
clock neurons, and the majority of insects may not use it at all.
Thus the transcriptional negative feedback loop in all clock
neurons of the fruit fly might be exceptional, and it might
just coincidentally show striking similarities to the circadian
system of nocturnal rodents. Similarly, the mouse-, the
hamster- and the rat-circadian clocks might represent special
cases. So far, very few other mammals have been investigated
systematically for the molecular function of their clock. At
present it is unknown whether a transcriptional feedback loop
is evolutionarily ancient or whether it was acquired later as
an additional feature fine-tuning clock protein cycling.
Concerning the location of the master clock, it is likely
that originally two clocks have been used by insects, one in
the dorsal protocerebrum of the central brain (close to the
neuroendocrine system), and the other near the optic lobe.
During evolution some insects may have lost the central brain
oscillator, others the optic lobe pacemaker, and the majority
of insects may still utilize both clocks, but to different degrees.
This reminds us of the situation in vertebrates. In mammals,
the suprachiasmatic nucleus is the pacemaker. It is associated
with the visual system and might therefore be analogous to
the optic lobe clock of insects. The pineal is associated with
neuroendocrine functions and has a pacemaking function in
lower vertebrates such as fishes, amphibians, reptiles and
C 2005
Biochemical Society
birds. The SCN and pineal gland are interconnected in a
complex manner, as is the optic lobe pacemaker of insects
with the central oscillator.
I thank C. Wülbeck for discussions and for a critical reading of this
paper.
References
1 Helfrich-Förster, C., Stengl, M. and Homberg, U. (1998) Chronobiol. Int.
15, 567–594
2 Helfrich-Förster, C. (2004) J. Comp. Physiol. 190, 601–613
3 Stanewsky, R. (2003) J. Neurobiol. 54, 111–147
4 Colot, H.V., Hall, J.C. and Rosbash, M. (1988) EMBO J. 7, 3929–3937
5 Piccin, A., Couchman, M., Clayton, J.D., Chalmers, D., Costa, R. and
Kyriacou, C.P. (2000) Genetics 154, 747–758
6 Regier, J.C., Fang, Q.Q., Mitter, C., Peigler, R.S., Friedlander, T.P. and Solis,
M.A. (1998) Mol. Biol. Evol. 15, 1172–1182
7 Reppert, S.M., Tsai, T., Roca, A.L. and Sauman, I. (1994) Neuron 13,
1167–1176
8 Froy, O., Chang, D.C. and Reppert, S.M. (2002) Curr. Biol. 12, 147–152
9 Takeda, Y., Chuman, Y., Shirasu, N., Sato, S., Matsushima, A., Kaneki, A.,
Tominaga, Y., Shimogashi, Y. and Shimohigashi, M. (2004) Zool. Sci. 21,
903–915
10 Toma, D.P., Bloch, G., Moore, D. and Robinson, G.E. (2000) Proc. Natl.
Acad. Sci. U.S.A. 97, 6914–6919
11 Shimizu, I., Kawai, Y., Taniguchi, M. and Aoki, S. (2001) Zool. Sci. 18,
779–789
12 Lin, G.G., Liou, R. and Lee, H. (2002) Chronobiol. Int. 19, 1023–1040
13 Chang, D.C. and Reppert, S.M. (2003) Curr. Biol. 13, 758–762
14 Levine, J.D., Sauman, I., Imbalzano, M., Reppert, S.M. and Jackson, F.R.
(1995) Neuron 15, 147–157
15 Myers, M.P., Wagner-Smith, K., Wesley, C.S., Young, M.W. and Sehgal, A.
(1995) Science 270, 805–808
16 Ousley, A., Zafarullah, K., Chen, Y., Emerson, M., Hickman, L. and
Sehgal, A. (1998) Genetics 148, 815–825
17 Goto, S.G. and Denlinger, D.L. (2003) J. Insect Physiol. 48, 803–816
18 Chang, D.C., McWatters, H.G., Williams, J.A., Gotter, A.L., Levine, J.D. and
Reppert, S.M. (2003) J. Biol. Chem. 278, 38149–38158
19 Markova, E.P., Ueda, H., Sakamoto, K., Oishi, K., Shimada, T. and
Takeda, M. (2003) Comp. Biochem. Physiol. B 134, 535–542
20 Ashmore, L., Sathyanarayanan, S., Silvestre, D.W., Emerson, M.M.,
Schotland, P. and Sehgal, A. (2003) J. Neurosci. 23, 7810–7819
21 Saez, L. and Young, M.W. (1996) Neuron 17, 911–920
22 Renn, S.C.P., Park, J.H., Rosbash, M., Hall, J.C. and Taghert, P.H. (1999)
Cell (Cambridge, Mass) 99, 791–802
23 Helfrich-Förster, C., Täuber, M., Park, J.H., Mühlig-Versen, M.,
Schneuwly, S. and Hofbauer, A. (2000) J. Neurosci. 20, 3339–3353
24 Sato, S., Chuman, Y., Matsushima, A., Tominaga, Y., Shimohigashi, Y. and
Shimohigashi, M. (2002) Zool. Sci. 19, 821–828
25 Park, J.H., Helfrich-Förster, C., Lee, G., Liu, L., Rosbash, M. and Hall, J.C.
(2000) Proc. Natl. Acad. Sci. U.S.A. 97, 3608–3613
26 Peng, Y., Stoleru, D., Levine, J.D., Hall, J.C. and Rosbash, M. (2003)
PLoS Biol. 1, 1–9
27 Lin, Y., Stormo, G.D. and Taghert, P.H. (2004) J. Neurosci. 24, 7951–7957
28 Schneider, N.-L. and Stengl, M. (2005) J. Neurosci. 25, 5138–5147
29 Helfrich-Förster, C. (1995) Proc. Natl. Acad. Sci. U.S.A. 92, 612–616
30 Frisch, B., Fleissner, G., Fleissner, G., Brandes, C. and Hall, J.C. (1996)
Cell Tissue Res. 286, 411–429
31 Sauman, I. and Reppert, S.M. (1996) Neuron 17, 889–900
32 Závodská, R., Sauman, I. and Sehnal, F. (2003) J. Biol. Rhythms 18,
106–120
33 Bloch, G., Solomon, S.M., Robinson, G.E. and Fahrbach, S.E. (2003)
J. Comp. Neurol. 464, 269–284
34 Wise, S., Davis, N.T., Tyndale, E., Noveral, J., Folwell, M.G., Bedian, V.,
Emery, I.F. and Siwicki, K.K. (2002) J. Comp. Neurol. 447, 366–380
35 Závodská, R., Sauman, I. and Sehnal, F. (2003) Insect Biochem. Mol. Biol.
33, 1227–1238
36 Homberg, U., Würden, S., Dircksen, H. and Rao, K.R. (1991)
Cell Tissue Res. 266, 343–357
Mechanistic and Functional Studies of Proteins
37 Nässel, D.R., Shiga, S., Wikstrand, E.M. and Rao, K.R. (1991)
Cell Tissue Res. 266, 511–523
38 Sehadova, H., Sauman, I. and Sehnal, F. (2003) Cell Tissue Res. 312,
113–125
39 Homberg, U., Davis, N.T. and Hildebrand, J.G. (1991) J. Comp. Neurol.
303, 35–52
40 Truman, J.W. (1972) J. Comp. Physiol. 81, 99–114
41 Fleissner, G. (1982) J. Comp. Physiol. 149, 311–316
42 Nishiitsutsuji-Uwo, J. and Pittendrigh, C.S. (1968) Z. Vergl. Physiol. 88,
21–30
43 Page, T.L. (1982) Science 216, 73–75
44
45
46
47
Stengl, M. and Homberg, U. (1994) J. Comp. Physiol. 175, 203–213
Tomioka, K. and Abdelsalam, S. (2004) Zool. Sci. 21, 1153–1162
Cheng, Y. and Hardin, P.E. (1998) J. Neurosci. 18, 741–750
Dembinska, M., Stanewsky, R., Hall, J.C. and Rosbash, M. (1997)
J. Biol. Rhythms 12, 157–172
48 Stanewsky, R., Frisch, B., Brandes, C., Hamblen-Coyle, M.J., Rosbash, M.
and Hall, J.C. (1997) J. Neurosci. 17, 676–696
49 Nakajima, M., Imai, K., Ito, H., Nishiwaki, T., Murayama, Y., Iwasaki, H.,
Oyama, T. and Kondo, T. (2005) Science 308, 414–415
Received 20 June 2005
C 2005
Biochemical Society
961