Cucaracha Circadian - UNAM
Transcript of Cucaracha Circadian - UNAM
REVIEW
Neural Organization of the Circadian System of theCockroach Leucophaea maderae
Uwe Homberg,* Thomas Reischig, and
Monika Stengl
Fachbereich Biologie, Tierphysiologie, Universitat Marburg,
Marburg, Germany
ABSTRACT
The cockroach Leucophaea maderae was the first animal in which lesion experiments
localized an endogenous circadian clock to a particular brain area, the optic lobe. The
neural organization of the circadian system, however, including entrainment pathways,
coupling elements of the bilaterally distributed internal clock, and output pathways
controlling circadian locomotor rhythms are only recently beginning to be elucidated.
As in flies and other insect species, pigment-dispersing hormone (PDH)-immunoreac-
tive neurons of the accessory medulla of the cockroach are crucial elements of the
circadian system. Lesions and transplantation experiments showed that the endoge-
neous circadian clock of the brain resides in neurons associated with the accessory
medulla. The accessory medulla is organized into a nodular core receiving photic input,
and into internodular and peripheral neuropil involved in efferent output and coupling
input. Photic entrainment of the clock through compound eye photoreceptors appears to
occur via parallel, indirect pathways through the medulla. Light-like phase shifts in
circadian locomotor activity after injections of g-aminobutyric acid (GABA)- or Mas-
allatotropin into the vicinity of the accessory medulla suggest that both substances are
involved in photic entrainment. Extraocular, cryptochrome-based photoreceptors
appear to be present in the optic lobe, but their role in photic entrainment has not
*Correspondence: Uwe Homberg, Fachbereich Biologie=Tierphysiologie, Philipps-Universitat
Marburg, D-35032 Marburg, Germany; Fax: þ49-6421-2828941; E-mail: [email protected]
marburg.de.
CHRONOBIOLOGY INTERNATIONAL
Vol. 20, No. 4, pp. 577–591, 2003
DOI: 10.1081=CBI-120022412 0742-0528 (Print); 1525-6073 (Online)
Copyright # 2003 by Marcel Dekker, Inc. www.dekker.com
577
been examined. Pigment-dispersing hormone-immunoreactive neurons provide efferent
output from the accessory medulla to several brain areas and to the peripheral visual
system. Pigment-dispersing hormone-immunoreactive neurons, and additional hetero-
lateral neurons are, furthermore, involved in bilateral coupling of the two pacemakers.
The neuronal organization, as well as the prominent involvement of GABA and
neuropeptides, shows striking similarities to the organization of the suprachiasmatic
nucleus, the circadian clock of the mammalian brain.
Key Words: Insect brain; Circadian clock; Pigment-dispersing hormone; Light
entrainment; Accessory medulla; Cockroach; Leucophaea maderae.
INTRODUCTION
Internal circadian clocks are universal adaptations of living organisms to the 24h
periodicity of the environment. Their molecular organization and localization, the
analysis of clock-controlled effectors, and mechanisms for entrainment of these clocks
have been the focus of intensive research in organisms of all main phyla. Work in insects
has focused heavily on the fruit fly Drosophila melanogaster, owing largely to the
multitude of available molecular genetic tools, and has greatly advanced our under-
standing of the molecular machinery underlying the circadian system (Allada et al., 2001;
Stanewsky, 2002; Young and Kay, 2001). These studies also revealed: (i) in addition to a
‘‘master clock’’ in the brain for the control of behavioral output, additional clocks exist in
the nervous system and in peripheral organs (Giebultowicz, 1999); (ii) parallel mechan-
isms and pathways operate for light entrainment of the clock (Helfrich-Forster et al.
2001); and (iii) distinct neural and molecular pathways are required for the control of
specific outputs of the clock (Helfrich-Forster, 1998; McNeil et al., 1998; Zhang et al.,
2000). Comparative studies, however, showed that other insects may considerably differ
from D. melanogaster with respect to clock localization, entrainment mechanisms, and
molecular basis of the clock (Helfrich-Forster et al., 1998). For a general understanding
of the circadian system in insects, studies on a variety of insect species are, therefore,
essential. Insect species with a larger brain size than that of D. melanogaster, such as the
cricket Gryllus bimaculatus, the cockroach Leucophaea maderae, or the moth Antheraea
polyphemus, furthermore, offer the advantage of a cellular and physiological approach to
the circadian system.
Research on the circadian system of the cockroach L. maderae is favored by the
prominent circadian rhythm of locomotor activity of this insect. Owing to its large size and
robustness, L. maderae was the first animal in which lesion experiments have localized the
circadian clock to a particular brain region (Nishiitsutsuji-Uwo and Pittendrigh, 1968b).
More recently, the organization of the putative circadian timekeeping center in the brain,
the accessory medulla, has been analyzed at the cellular level. The present paper reviews
the neuronal organization of the circadian system of L. maderae with special emphasis on
its internal organization, distribution of putative neurotransmitters and neuropeptides
operating in the circadian system, light entrainment pathways, clock outputs, and bilateral
coupling mechanisms.
578 Homberg, Reischig, and Stengl
LOCALIZING THE INTERNAL CLOCK THROUGH
LESIONS AND TRANSPLANTATIONS
Cockroaches show robust circadian rhythms in wheel running activity (Roberts, 1960)
and in electroretinogram recordings (Wills et al., 1985). Both assays have been used to
monitor circadian clock function after various lesions to the cockroach brain. Transections
of the optic lobe at various levels showed that the circadian pacemaker controlling
locomotor activity (Nishiitsutsuji-Uwo and Pittendrigh, 1968b; Roberts, 1974), neural
activity in the ventral nerve cord (Colwell and Page, 1990), and electroretinogram rhythms
(Wills et al., 1985) resides in the optic lobe, and is associated with the medulla=lobula
but not with the lamina. A sustained electroretinogram rhythm could be recorded from the
isolated optic lobe for several days (Colwell and Page, 1990), and transplantation of
the optic lobe into an animal, whose own optic lobes had been removed, restored circadian
rhythmicity of locomotor activity after several weeks (Page, 1982). These experiments
showed that the optic lobe contains a self-sustained circadian clock controlling locomotor
behavior and visual sensitivity.
Microlesions using tungsten microelectrodes further specified the site of the clock
(Page, 1978; Sokolove, 1975) and showed that lesions in or near the lobula, particularly in
the ventral cell body cortex between the medulla and lobula, produced a high percentage of
arrhythmic cockroaches. Immunocytochemical studies revealed that this area contains the
cell bodies of neurons with arborizations in the accessory medulla, a small neuropil at the
anterior ventromedian edge of the medulla (Homberg et al., 1991; Petri et al., 1995).
Distinct and partly overlapping subpopulations of these neurons are immunoreactive with
antisera against serotonin, g-aminobutyric acid (GABA), and a variety of neuropeptides,
including Mas-allatotropin, Dip-allatostatin, FMRFamide, pigment-dispersing hormone
(PDH), and leucokinin (Petri et al., 1995; 2002). These neurons send processes into
the accessory medulla, and most of them have additional projections into other areas of the
brain. These studies suggest that neuropeptides play a key role in the circadian system of
the cockroach and draw attention to the accessory medulla as the possible synaptic
integration center of the clock.
Immunostaining with an antiserum against the crustacean peptide PDH (Rao, 2001) is
particularly selective for neurons of the accessory medulla, and throughout the brain of the
cockroach, only three groups of neurons are immunoreactive. Peptides closely related to
crustacean PDH, termed pigment-dispersing factors (PDFs), have been identified in several
insect species, including the cockroach Periplaneta americana (Rao, 2001). Two groups of
PDH-immunoreactive neurons with cell bodies at the posterior dorsal and posterior ventral
edge of the lamina were termed PDFLa cells and, a third group of 16 neurons near the
accessory medulla, PDFMe neurons [Homberg et al., 1991; Reischig and Stengl, 1996;
Fig. 1(A)]. The PDFLa neurons have arborizations throughout the lamina and probably
contribute processes to a fan of immunostained fibers along the distal face of the medulla
[Fig. 1(A)]. Their possible involvement in the circadian system is unknown. The role of the
accessory medulla and the contribution of PDFMe neurons to the circadian system was
addressed in lesion and transplantation experiments (Reischig and Stengl, 2002; Stengl
and Homberg, 1994). Bilateral transections of the optic stalk in the cockroach abolished
circadian wheel-running activity. In many individuals, however, the circadian rhythm
reappears after several weeks corresponding with mass regeneration of fibers from the
optic lobe to the median protocerebrum (Page, 1983a).
Neural Organization of Leucophaea maderae 579
Immunocytochemistry using PDH antiserum shows a strong correlation between
ingrowth and reinnervation of original targets in the lateral and superior protocerebrum by
fibers from PDFMe neurons in animals that regained circadian activity after lesions (Fig. 2;
Stengl and Homberg, 1994). This suggests that PDF acts as an output signal of the clock to
particular brain areas involved in motor control. Finally, ectopic transplantations of
Figure 1. Arborizations of PDH-immunoreactive neurons in the brain of L. maderae (A) and
anatomical organization of the accessory medulla (B). (A) Camera lucida reconstruction of PDH-
immunoreactive neurons embedded in a computer-aided three-dimensional reconstruction of the
cockroach brain (modified after Reischig and Stengl, 2002). PDFMe neurons have cell bodies near
the accessory medulla (AMe). The neurons have dense ramifications in the AMe and send processes
through the distalmost layer of the medulla (Me) to the lamina (La) and a second set of axonal fibers
to various areas in the median protocerebrum. Immunoreactive fibers in the anterior and posterior
optic commissures (AOC, POC) provide heterolateral connections between the two AMea. Two
additional groups of PDH-immunoreactive neurons with somata near the posterior dorsal (dPDFLa)
and posterior ventral edge of the medulla contribute to immunostained arborizations in the lamina
and first optic chiasma. (B) Computer-aided three-dimensional reconstruction corresponding to the
boxed area in (A). Somata of the AMe-neurons are organized in six groups: median neurons (MNe),
ventro-median neurons (VMNe), medial frontoventral neurons (MFVNe), ventral neurons (VNe),
distal frontoventral neurons (DFVNe), and ventroposterior neurons (VPNe). The core of the AMe,
composed of dense nodular neuropil (NN) and coarse internodular neuropil (IN) is surrounded by a
shell of coarse neuropil (SN). Two of the six cell groups (VNe, VMNe) are shown separately in the
right panel. Lo, lobula. Scale bars: 200 mm (A), 20mm (B).
580 Homberg, Reischig, and Stengl
accessory medullae and adjacent cell bodies into animals whose own optic lobes had been
removed restored circadian activity (Reischig and Stengl, 2003). In these experiments, a
small piece of tissue (about 200 mm diameter) including the accessory medulla and its
neurons was transplanted from a donor cockroach to the optic or antennal lobe of a host
cockroach whose accessory medullae or optic lobes had been removed. In nearly 30% of
Figure 2. Reappearance of circadian wheel-running activity after optic stalk severance correlates
with mass regeneration of PDH-immunoreactive fibers from the AMe to the midbrain (modified after
Stengl and Homberg, 1994). (A) Activity record of a cockroach kept in DD, which received
transection of the right optic stalk and crushing of the left optic stalk at day 22 (arrowhead). After the
operation, activity is reduced and the animal becomes arrhythmic. Five days after reappearance of
rhythmic activity at day 40 the cockroach was sacrificed and the brain was examined for PDH
immunostaining. (B) Pigment-dispersing hormone immunostaining in the brain of the cockroach in
(A) at day 44. Mass regeneration of PDH-immunoreactive fibers has occurred predominantly from
the right optic lobe to the median protocerebrum. Many fibers have reinvaded their original targets in
the superior median, superior lateral, inferior lateral, and ventro-lateral protocerebrum (SMP, SLP,
ILP, VLP) or have entered the posterior optic commissure (POC). Scale bar: 200 mm.
Neural Organization of Leucophaea maderae 581
these animals, a circadian rhythm of locomotor activity reappeared and in all animals that
regained rhythmicity, transplanted PDFMe neurons had regenerated fibers to their original
targets in the protocerebrum (Reischig and Stengl, 2003). These data further confirm that
the accessory medulla is the site of the circadian pacemaker controlling locomotor activity
rhythms in the cockroach and suggest that neural connections by PDH-immunoreactive
fibers from the accessory medulla to their original targets in the brain may be essential for
clock control of locomotor activity.
INTERNAL ORGANIZATION AND CHEMOARCHITECTURE OF
THE ACCESSORY MEDULLA
The accessory medulla has a pear-shaped form with a maximum longitudinal size
of about 90 mm (Reischig and Stengl, 1996). It is composed of a core consisting of
dense knots of neuropil termed noduli. The nodular core is embedded in coarse
internodular neuropil and is surrounded by a shell of coarse neuropil, which is
anteriorly continuous with the adjacent neuropil of the medulla [Fig. 1(B)]. Somata
arranged in six groups in the cell cortex antero-ventrally to the accessory medulla
contribute processes to the accessory-medulla neuropil [Fig. 1(B)]. Eight of the 16
PDFMe neurons have somata in the VNe group, four small somata are in the DFVNe
group (see Fig. 1 for nomenclature), and the four remaining somata are in the cell
cortex posterior to the medulla=lobula neuropil not shown in Fig. 1. The other cell
groups contain accessory-medulla neurons immunostained with antisera against other
neuropeptides and transmitter candidates, e.g., GABA-immunstained neurons in group
MNe and Mas-allatotropin-immunostained neurons concentrated in the DFVNe and
MFVNe cell groups (Petri et al., 1995; see Fig. 1 for nomenclature). The three
compartments of the accessory medulla, noduli, internodular neuropil, and peripheral
shell, are differentially supplied by immunocytochemically characterized populations of
AMe neurons. Neurons immunostained with antisera against GABA, Mas-allatotropin,
and leucokinin have processes concentrated in the nodular core of the accessory
medulla (Petri et al., 1995; 2002). These neurons are either local interneurons of the
accessory medulla (Mas-allatotropin) or connect the distalmost layer of the medulla
(leucokinin) or a median layer of the medulla (GABA) to the nodular core of the
accessory medulla. In contrast, neurons immunostained with antisera against serotonin,
PDH, FMRFamide, Dip-allatostatin, and corazonin have processes concentrated in the
internodular neuropil and peripheral shell of the accessory medulla (Petri et al., 1995;
2002). These neurons connect the accessory medulla through a fan of fibers with the
distalmost layer of the medulla and with the lamina (PDH, FMRFamide, serotonin) and
have axonal projections to several areas in the median protocerebrum and to the
contralateral optic lobe (PDH, FMRFamide). The concentration of neuropeptides in the
accessory medulla correlates with an abundance of dense core vesicles in fiber profiles
in the accessory medulla revealed by ultrastructural examination. Based on differences
in size and electron density, at least four different types of dense core vesicles have
been distinguished, which are differentially distributed in the nodular core, in the
internodular and shell neuropil. Granular vesicles are almost completely confined to
the nodular core, while subpopulations of profiles with PDH-immunoreactive
large dense-core vesicles are concentrated in the internodular neuropil, and
582 Homberg, Reischig, and Stengl
PDH-immunoreactive medium size dense-core vesicles are most common in the shell
neuropil of the accessory medulla (Reischig and Stengl, 1996).
PARALLEL PATHWAYS FOR LIGHT ENTRAINMENT
Light entrainment of the circadian clock of L. maderae occurs through photoreceptors
in or near the compound eye. Bilateral transections of the optic nerves connecting the
compound eyes to the optic lobes result in free-running circadian locomotor activity, while
removal of the ocelli had no effect on light entrainment (Nishiitsutsuji-Uwo and
Pittendrigh, 1968a; Roberts, 1965; 1974). These data indicate that ocellar photoreceptors
are not essential for photic entrainment and provide no evidence for the involvement of
other nonretinal photoreceptors. A series of bilateral lesion experiments, moreover,
demonstrated that light entrainment of each of the two bilaterally distributed circadian
pacemakers occurs via photoreceptors in the ipsi- and contralateral compound eye (Page,
1978; Page et al., 1977).
Recent studies from our laboratories showed that neural connections from photo-
receptors to the accessory medulla involve intercalated interneurons and may be organized
in several parallel pathways. Histamine immunostaining used to label all compound-eye
photoreceptor cells showed that photoreceptors terminate in the lamina and in distal layers
of the medulla, but not in the accessory medulla (Loesel and Homberg, 1999). Intracellular
recordings showed that only certain cell types of the accessory medulla respond to photic
stimuli (Loesel and Homberg, 2001). Excitatory responses to light were found in neurons
with tangential dendritic processes in median layers of the medulla, projections to the
nodular core of the accessory medulla, and fan-shaped, apparently axonal, processes to the
lamina [Fig. 3(A); Loesel and Homberg, 2001]. These neurons are, therefore, likely
candidates for photic input elements to the accessory medulla, but whether they receive
direct input from compound eye photoreceptors remains to be demonstrated. In the same
study light-sensitive commissural neurons were identified, which interconnect median
layers of the medulla and nodular core of the accessory medulla of both hemispheres.
These neurons are candidates for providing light entrainment from the contralateral eye
[Fig. 3(C)]. Another line of evidence supporting the role of the nodular core of the
accessory medulla as the recipient for light entrainment signals comes from phase shifting
experiments. Injections of GABA or Mas-allatotropin into the medulla result in phase-
dependent phase shifts of circadian locomotor activity (Petri et al., 2002). The phase
response curves have striking resemblence to phase response curves obtained for 6h light
pulses [Fig. 4(A)], and suggest that both substances are involved in light entrainment. At
least one GABA-immunoreactive neuron is a member of the light-responsive tangential
inputs from the medulla, many others connect the medulla via the distal tract, a fiber
bundle along the distal face of the medulla (Reischig and Stengl, 1996), to the nodular core
of the accessory medulla. Mas-allatotropin-immunostained local neurons of the nodular
core likewise appear to be involved in the photic input pathway.
In addition to light signalling pathways from compound-eye photoreceptors via
medulla interneurons to the accessory medulla, two additional pathways might contribute
to light sensitivity in the accessory medulla. A class of apparently centrifugal neurons with
ramifications in the median protocerebrum and axonal projections to the accessory
medulla is strongly inhibited by light (Loesel and Homberg, 2001). One of the recorded
Neural Organization of Leucophaea maderae 583
neurons was an ocellar interneuron. A prominent role of this pathway in photic entrain-
ment is unlikely, given the evidence from lesion studies cited above, but a modulatory role
in adjusting light sensitivity, as shown for an ocellar contribution in a cricket (Rence et al.,
1988), or a role in entrainment under particular lighting schedules as shown for the
geniculo-hypothalamic pathway in rats (Edelstein and Amir, 1999), appears possible.
Figure 3. Camera lucida reconstructions of single intracellularly recorded and stained neurons of the
accessory medulla (AMe) (modified after Loesel and Homberg, 2001). (A) Candidate neuron involved
in photic entrainment of the clock. The neuron has its cell body in cell group MNe (see Fig. 1). It has
tangential arborizations in a median layer of the medulla (Me), innervates the nodular core of the AMe
and projects via a fan of fibers to the lamina (La) and to small accessory neuropils of the lamina
(arrowhead). The neuron was strongly excited by light during daytime. (B) Candidate output neuron of
the AMe with cell body in cell group VNe. The neuron has sparse ramifications in the shell and
internodular neuropil of the AMe and sends axonal processes to the La and to the median
protocerebrum near the pedunculus (P) of the mushroom body. The neuron was unresponsive to
light stimuli at daytime. (C) Candidate neuron involved in light entrainment via the contralateral eye. Its
cell body lies in group VMNe (see Fig. 1). The neuron has bilateral arborizations in the Me and AMe.
The neuron was strongly excited by light and showed pronounced polarization sensitivity. aL, Ca,
a-lobe, and calyces of the mushroom body; Lo, lobula. Scale bars: 100mm in (A) and (B); 200mm in (C).
584 Homberg, Reischig, and Stengl
Finally, ultrastructural studies and immunoreactivity for the blue light photopigment
cryptochrome provided evidence for the presence of two extraocular photoreceptor organs
in the optic lobe of the cockroach (Fleissner et al., 2001). The lamina organ is an elongated
structure which runs parallel to the lamina in front of the first optic chiasm, and the second
organ, termed lobula organ, lies near the accessory medulla in front of the second optic
chiasm. So far, however, a possible involvement of these organs in photic entrainment of
the clock has not been investigated at any level.
OUTPUT PATHWAYS
Reappearance of circadian wheel-running activity after bilateral optic stalk severance
or transplantations of the accessory medulla correlates with regenerated PDH-immuno-
reactive fibers from PDFMe neurons to the median protocerebrum (Fig. 2; Reischig and
Stengl, 2003; Stengl and Homberg, 1994). Therefore, PDFMe neurons might provide
the connection between the pacemaker and motor control areas in the protocerebrum.
While in animals with optic stalk severance, equal reinnervation of all original targets in
the protocerebrum was observed (Stengl and Homberg, 1994), regeneration in animals
with transplanted accessory medulla occurred predominantly to the superior median and
lateral protocerebrum, suggesting that these areas might be more important for clock
Figure 4. Phase response curves demonstrate different temporal sensitivities of the circadian clock
to injections of Mas-allatotropin, GABA (A modified after Petri et al., 2002) and PDF (B, after Petri
and Stengl, 1997). (A) Injections of Mas-allatotropin and GABA into the medulla lead to phase-
dependent phase shifts in circadian locomotor activity which show striking resemblance to the phase-
shifting effects of light. This suggests that GABA and a peptide related to Mas-allatotropin are
involved in light entrainment pathways. (B) Injections of PDF into the medulla lead to an all-delay
type phase-response curve with maximum phase shifting effects in the late subjective day. This
suggests that PDF is involved in non-photic inputs to the circadian clock. Data are merged into 2h
bins. Solid stars indicate phase delays that are significantly different from phase shifts following
control injections. Phase delays at CT 8:00–10:00h (double star) and at CT 10:00–12:00h (single
star) are significantly different ( p< 0.05) from PDF injections at other circadian times (open stars).
Neural Organization of Leucophaea maderae 585
control of locomotion than others (Reischig and Stengl, 2003). PDFMe neurons also have
fan-like processes through distalmost layer 1 of the medulla toward the lamina [Fig.
1(A)], suggesting that they are also involved in circadian control of visual sensitivity at
the level of the lamina. In single-cell recordings, neurons of the accessory medulla that
were unresponsive to light stimuli during the day were regarded as candidate output
neurons of the clock, since light does not cause phase shifts during the day. Like PDFMe
neurons, these neurons had ramifications confined to the internodular and peripheral shell
neuropil of the accessory medulla. In all cases, their axonal fibers had two targets, fan-like
distal projections to the lamina or first optic chiasm, and proximal projections to certain
areas in the median protocerebrum [Loesel and Homberg, 2001; Fig. 3(B)].
COUPLING OF THE BILATERALLY SYMMETRIC CLOCKS
The two pacemakers in the right and left optic lobe of the cockroach are coupled
through neural pathways (Page, 1978; 1983b; Page et al., 1977). Lesions of one of the two
pacemakers invariably leads to an increase in the free-running period of the rhythm,
indicating that coupling of both pacemakers accelerates the system, possibly through
mutual inhibition (Page, 1983b). Several lines of evidence suggest that PDF may be
involved in the coupling pathway. Dextran tracing studies and double-labeling of
commissural neurons and PDH immunostaining showed that three of the 16 PDFMe
neurons are commissural neurons that interconnect the two accessory medullae (Petri,
1998; Reischig and Stengl, 2002). The neurons project via the anterior and posterior optic
commissures, innervate the internodular and shell neuropil of the accessory medulla, and
send a fan of fibers along the distal surface of the medulla into the lamina [Fig. 5(B)].
Neurons of this morphological type were intracellularly stained [Fig. 5(A)] and were found
to be unresponsive to light stimuli during the day (Loesel and Homberg, 2001). Injections
of PDF into the medulla caused phase shifts in circadian wheel-running activity which
provides additional support for a role of PDF as a coupling signal in the circadian system
(Petri and Stengl, 1997). The resulting phase-response curve is monophasic with phase
delays during the late subjective day and, therefore, differs considerably from the phase-
response curve obtained with light pulses (Fig. 4). Pigment-dispersing factor is, therefore,
not part of the light entrainment pathway but participates in non-photic input to the clock.
Taken together with the immunocytochemical data, PDH-immunoreactive commissural
neurons are therefore, strong candidates for the bilateral coupling pathway of the two optic
lobe pacemakers. Evidence supporting a dual role of PDH-immunostained neurons as
output and coupling pathways is provided by ultrastructural studies, which revealed input
as well as output synapses from PDH-immunoreactive neurons in the accessory medulla
(Reischig and Stengl, 2003). Computer simulations, finally, demonstrated that in addition
to all-delay-type interactions as found with PDF injections, all-advance-type interactions
between the two pacemakers are required to explain all experimental findings on mutual
pacemaker coupling in L. maderae (Petri and Stengl, 2001). It is, therefore, likely, that
additional non-PDH-immunostained commissural neurons participate in bilateral coupling.
Candidates are commissural FMRFamide-immunostained connections between the two
accessory medullae, but the effects of FMRFamide-related peptides on locomotor activity
have not been tested so far.
586 Homberg, Reischig, and Stengl
COMPARISON BETWEEN FLIES, COCKROACH, AND MAMMALS
The circadian system of L. maderae, particularly with respect to the involvement
of PDF-containing neurons shows striking similarity to the well-studied circadian system
of the fruitfly D. melanogaster. In D. melanogaster, the so-called ‘‘lateral neurons’’ (LNs)
near the medulla are circadian pacemaker cells (reviewed by Helfrich-Forster et al.,
1998; Kaneko, 1998; Stanewsky, 2002). The neurons express the clock genes period,
timeless, and doubletime and the presence of these neurons as well as their
cyclical expression of period is essential for behavioral circadian rhythmicity (Helfrich-
Forster, 1998; Kaneko et al., 1997). A subset of the LNs contain PDF and mutations of the
pdf-gene or ectopic expression of pdf leads to severe deficits in circadian rhythms
Figure 5. Coupling pathways between the bilaterally distributed pacemakers in the AMe. (A)
Hypothetical neuron involved in coupling of both pacemakers. The neuron interconnects the AMe,
the distalmost layer of the medulla and the lamina of both hemispheres. The original Lucifer
injection (modified from Loesel and Homberg, 2001) showed incomplete staining of the neuron with
ramifications in the right brain hemisphere and a commissural fiber in the posterior optic commissure
(POC) toward the left optic lobe. Mirror-symmetric arborizations in the left optic lobe were added
using Adobe Photoshop 5.0 software because evidence from dextran backfills suggests that this cell
type has bilaterally symmetric arborizations in both optic lobes. The neuron was unresponsive to
light stimuli. (B) Schematic diagram showing two presumptive PDH-immunoreactive circadian
coupling pathways through the anterior (AOC; dark gray) and posterior optic commissure (POC;
light gray) as revealed by tracer injections combined with PDH immunostaining (modified after
Reischig and Stengl, 2002). ILP, inferior lateral protocerebrum; La, lamina; Me, medulla; POTu,
posterior optic tubercle; SLP, SMP, superior lateral, resp. median protocerebrum. Filled circles
indicate possible outputs and=or interactions between the two systems. Scale bar: 200mm in (A).
Neural Organization of Leucophaea maderae 587
(Helfrich-Forster et al., 2000; Renn et al., 1999). As in the cockroach, distinct subpopula-
tions of pdf-expressing neurons with different branching patterns can be distinguished in the
fly, consistent with a double role of the peptide as an output and bilateral coupling signal of
the clock. Differences between D. melanogaster and L. maderae appear to exist with respect
to light entrainment pathways; in the fly the pdf-expressing pacemaker cells possess their
own extraocular photopigment, cryptochrome, which is apparently not present in the
PDFMe neurons of the cockroach (Emery et al., 2000; Fleissner et al., 2001). Furthermore,
a direct synaptic input from an extraocular photoreceptor organ, termed the Hofbauer-
Buchner eyelet in the fly (Helfrich-Forster et al., 2001), has also not been found in the
cockroach. While in D. melanogaster strong evidence suggests that the pdf-expressing cells
are pacemaker cells, the identity of clock cells in the circadian system of L. maderae has not
been determined. Current experiments aimed at identifying the clock molecules PER and
TIM in the cockroach brain might help to solve this question.
While a distinct accessory medulla is hardly visible in D. melanogaster owing to fusion
with the adjacent medulla, the anatomical and neurochemical organization of the accessory
medulla of the cockroach bears considerable similarities to the organization of the
mammalian circadian pacemaker, the suprachiasmatic nucleus (reviewed by Moore et al.,
2002; Reghunandanan et al., 1993; van Esseveldt et al., 2000). Both areas are densely packed
with neuropeptides and have a clear internal substructure into a core receiving retinal input,
and a shell (internodular and shell neuropil in the cockroach) receiving nonphotic input. In
mammals outputs to effector systems exit the core as well as the shell, while in the cockroach,
outputs appear to exit primarily from the internodular and peripheral neuropil. These
similarities may be accidental, but more likely represent similarly optimized solutions for the
neural organizations of brain circadian pacemakers.
ACKNOWLEDGMENTS
Research in the authors’ laboratories on the insect circadian system has been
supported by DFG grant HO 950=9 to U.H. and by DFG grant STE 531=7 and a grant
from Human Science Frontier to M.S.
REFERENCES
Allada, R., Emery, P., Takahashi, J. S., Rosbash, M. (2001). Stopping time: the genetics of
fly and mouse circadian clocks. Annu. Rev. Neurosci. 24:1091–1119.
Colwell, C. S., Page, T. L. (1990). A circadian rhythm in neural activity can be recorded
from the central nervous system of the cockroach. J. Comp. Physiol. A 166:643–649.
Edelstein, K., Amir, S. (1999). The role of the intergeniculate leaflet in entrainment of
circadian rhythms to a skeleton photoperiod. J. Neurosci. 19:372–380.
Emery, P., Stanewsky, R., Helfrich-Forster, C., Emery-Le, M., Hall, J. C., Rosbash, M. (2000).
Drosophila CRY is a deep brain circadian photoreceptor. Neuron 26:493–504.
Fleissner, Ge., Loesel, R., Fleissner, Gu., Waterkamp, M., Kleiner, O., Batschauer, A.,
Homberg, U. (2001). Candidates for extraocular photoreceptors in the cockroach
suggest homology to the lamina and lobula organs in beetles. J. Comp. Neurol.
433:401–414.
588 Homberg, Reischig, and Stengl
Giebultowicz, J. M. (1999). Insect circadian clocks: is it all in their heads? J. Insect
Physiol. 45:791–800.
Helfrich-Forster, 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.
Helfrich-Forster, C., Stengl, M., Homberg, U. (1998). Organization of the circadian system
in insects. Chronobiol. Int. 15(6):567–594.
Helfrich-Forster, C., Tauber, M., Park, J. H., Muhlig-Versen, M., Schneuwly, S., Hofbauer, A.
(2000). Ectopic expression of the neuropeptide pigment-dispersing factor alters
behavioral rhythms in Drosophila melanogaster. J. Neurosci. 20:3339–3353.
Helfrich-Forster, C., Winter, C., Hofbauer, A., Hall, J. C., Stanewsky, R. (2001). The
circadian clock of fruit flies is blind after elimination of all known photoreceptors.
Neuron 30:249–261.
Homberg, U., Wurden, S., Dircksen, H., Rao, K. R. (1991). Comparative anatomy of
pigment-dispersing hormone-immunoreactive neurons in the brain of orthopteroid
insects. Cell Tissue Res. 266:343–357.
Kaneko, M. (1998). Neural substrates of Drosophila rhythms revealed by mutants and
molecular manipulations. Curr. Opin. Neurobiol. 8:652–658.
Kaneko, M., Helfrich-Forster, C., 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.
Loesel, R., Homberg, U. (1999). Histamine-immunoreactive neurons in the brain of the
cockroach Leucophaea maderae. Brain Res. 842:408–418.
Loesel, R., Homberg, U. (2001). Anatomy and physiology of neurons with processes in the
accessory medulla of the cockroach Leucophaea maderae. J. Comp. Neurol.
439:193–207.
McNeil, G. P., Zhang, X., Genova, G., Jackson, F. R. (1998). A molecular rhythm
mediating circadian clock output in Drosophila. Neuron 20:297–303.
Moore, R. Y., Speh, J. C., Leak, R. K. (2002). Suprachiasmatic nucleus organization. Cell
Tissue Res. 309:89–98.
Nishiitsutsuji-Uwo, J., Pittendrigh, C. S. (1968a). Central nervous system control of
circadian rhythmicity in the cockroach. II. The pathway of light signals that entrain
the rhythm. Z. Vgl. Physiol. 58:1–13.
Nishiitsutsuji-Uwo, J., Pittendrigh, C. S. (1968b). Central nervous system control of
circadian rhythmicity in the cockroach. III. The optic lobes, locus of the driving
oscillation? Z. Vgl. Physiol. 58:14–46.
Page, T. L. (1978). Interactions between bilaterally paired components of the cockroach
circadian system. J. Comp. Physiol. 124:225–236.
Page, T. L. (1982). Transplantation of the cockroach circadian pacemaker. Science
216:73–75.
Page, T. L. (1983a). Regeneration of the optic tracts and circadian pacemaker activity in
the cockroach Leucophaea maderae. J. Comp. Physiol. 152:231–240.
Page, T. L. (1983b). Effects of optic-tract regeneration on internal coupling in the circadian
system of the cockroach. J. Comp. Physiol. 153:353–363.
Page, T. L., Caldarola, P. C., Pittendrigh, C. S. (1977). Mutual entrainment of bilaterally
distributed circadian pacemakers. Proc. Natl. Acad. Sci. USA 74(3):1277–1281.
Neural Organization of Leucophaea maderae 589
Petri, B. (1998). Neuronal Organization of a Circadian Clock. Ph.D. thesis, University of
Regensburg, Germany.
Petri, B., Stengl, M. (1997). Pigment-dispersing hormone shifts the phase of the
circadian pacemaker of the cockroach Leucophaea maderae. J. Neurosci.
17:4087–4093.
Petri, B., Stengl, M. (2001). Phase response curves of a molecular model
oscillator: implications for mutual coupling of paired oscillators. J. Biol. Rhythms
16:125–141.
Petri, B., Stengl, M., Wurden, S., Homberg, U. (1995). Immunocytochemical character-
ization of the accessory medulla in the cockroach Leucophaea maderae. Cell Tissue
Res. 282:3–19.
Petri, B., Homberg, U., Loesel, R., Stengl, M. (2002). Evidence for a role of GABA and
Mas-allatotropin in photic entrainment of the circadian clock of the cockroach
Leucophaea maderae. J. Exp. Biol. 205:1459–1469.
Rao, K. R. (2001). Crustacean pigmentary-effector hormones: chemistry and functions of
RPCH, PDH, and related peptides. Amer. Zool. 41:364–379.
Reghunandanan, V., Reghunandanan, R., Singh, P. I. (1993). Neurotransmitters of the
suprachiasmatic nucleus: role in the regulation of circadian rhythms. Prog. Neuro-
biol. 41:647–655.
Reischig, T., Stengl, M. (1996). Morphology and pigment-dispersing hormone immuno-
cytochemistry of the accessory medulla, the presumptive circadian pacemaker of the
cockroach Leucophaea maderae: a light and electron-microscopic study. Cell Tissue
Res. 285:305–319.
Reischig, T., Stengl, M. (2002). Optic lobe commissures in a three-dimensional brain
model of the cockroach Leucophaea maderae: a search for the circadian coupling
pathways. J. Comp. Neurol. 443:388–400.
Reischig, T., Stengl, M. (2003). Ectopic transplantation of the accessory medulla restores
circadian locomotor rhythms in arrhythmic cockroaches (Leucophaea maderae).
J. Exp. Biol. in press.
Rence, B. G., Lisy, M. T., Garves, B. R., Qinlan, B. J. (1988). The role of ocelli in
circadian singing rhythms of crickets. Physiol. Entomol. 13:201–212.
Renn, S. C. P., Park, J. H., Rosbash, M., Hall, J. C., 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.
Roberts, S. K. (1960). Circadian activity in cockroaches. I. The free-running activity in
steady-state. J. Cell. Comp. Physiol. 59:99–110.
Roberts, S. K. (1965). Photoreception and entrainment of cockroach activity rhythms.
Science 148:958–959.
Roberts, S. K. (1974). Circadian rhythms in cockroaches. Effects of optic lobe lesions.
J. Comp. Physiol. 88:21–30.
Sokolove, P. G. (1975). Localization of the cockroach optic lobe circadian pacemaker with
microlesions. Brain Res. 87:13–21.
Stanewsky, R. (2002). Clock mechanisms in Drosophila. Cell Tissue Res. 309:11–26.
Stengl, M., Homberg, U. (1994). Pigment-dispersing hormone-immunoreactive neurons in
the cockroach Leucophaea maderae share properties with circadian pacemaker
neurons. J. Comp. Physiol. A 175:203–213.
590 Homberg, Reischig, and Stengl
van Esseveldt, L. E., Lehman, M. N., Boer, G. J. (2000). The suprachiasmatic nucleus and
the circadian time-keeping system revisited. Brain Res. Rev. 33:34–77.
Wills, S. A., Page, T. L., Colwell, C. S. (1985). Circadian rhythms in the electroretinogram
of the cockroach. J. Biol. Rhythms 1(1):25–37.
Young, M., Kay, S. A. (2001). Time zones: a comparative genetics of circadian clocks.
Nature Rev. Genetics 2:702–715.
Zhang, X., McNeil, G. P., Hilderbrand-Chae, M. J., Franklin, T. M., Schroeder, A. J.,
Jackson, F. R. (2000). Circadian regulation of the lark RNA-binding protein within
identifiable neurosecretory cells. J. Neurobiol. 45:14–29.
Neural Organization of Leucophaea maderae 591