The clock-gene-expressing lateral neurons are essential for the locomotor activity rhythm of Drosophila melanogaster. Traditionally, these neurons are divided into three groups: the dorsal lateral neurons (LN(d)), the large ventral lateral neurons (l-LN(v)), and the small ventral lateral neurons (s-LN(v)), whereby the latter group consists of four neurons that express the neuropeptide pigment-dispersing factor (PDF) and a fifth PDF-negative neuron. So far, only the l-LN(v) and the PDF-positive s-LN(v) have been shown to project into the accessory medulla, a small neuropil that contains the circadian pacemaker center in several insects. We show here that the other lateral neurons also arborize in the accessory medulla, predominantly forming postsynaptic sites. Both the l-LN(v) and LN(d) are anatomically well suited to connect the accessory medullae. Whereas the l-LN(v) may receive ipsilateral photic input from the Hofbauer-Buchner eyelet, the LN(d) invade mainly the contralateral accessory medulla and thus may receive photic input from the contralateral side. Both the LN(d) and the l-LN(v) differentiate during midmetamorphosis. They do so in close proximity to one another and the fifth PDF-negative s-LN(v), suggesting that these cell groups may derive from common precursors.

This chapter summarizes our present knowledge about the master clock of the fruit fly at the neuronal level. The clock is organized in distinct groups of interconnected pacemaker neurons with different functions. All of these neurons appear to communicate with one another in order to produce the species-specific activity rhythm, which is organized in morning (M) and evening (E) activity bouts. These two activity components are differentially influenced by distinct groups of pacemaker neurons reminiscent of the Pittendrigh-Daan dual oscillator model. In the original work (Grima et al. 2004;Stoleru et al. 2004), the ventrolateral (LN v ) and dorsolateral (LN d ) plus some dorsal groups (DN) of clock neurons have been defined as M and E cells, respectively. We further specify that the clock neurons belong to the M and E oscillators and define a more complex picture of the Drosophila brain clock.

No abstract

The neuropeptide pigment-dispersing factor (PDF) is a key transmitter in the circadian clock of Drosophila melanogaster. Here we studied the rhythmic behavior of neural mutants with modified arborizations of the large PDF neurons. In sine oculis(1) (so(1)) mutants we found a higher density of PDF fibers in the fly's pacemaker center, the accessory medulla. These flies exhibited a significantly longer period (24.6 h) than control flies. When PDF levels were elevated to very high levels in the dorsal brain as true for so(mda) mutants and small optic lobes;so(1) double mutants (sol(1);so( 1)), a short-period component split off the long period in behavioral rhythmicity. The short period became shorter the higher the amount of PDF in this brain region and reached a value of approximately 21 h. The period alterations were clearly dependent on PDF, because so(1);Pdf 01 and so(mda);Pdf 01 double mutants showed a single free-running component with a period similar to Pdf 01 mutants (approximately 22.5 h) and significantly longer than the short period of so(mda) mutants. These observations indicate that PDF feeds back on the clock neurons and changes their period. Obviously, PDF lengthens the period of some clock neurons and shortens that of others.

The neuropeptide pigment-dispersing factor (PDF) plays an essential role in the circadian clock of the fruit fly Drosophila melanogaster, but many details of PDF signaling in the clock network are still unknown. We tried to interfere with PDF signaling by blocking the GTPase Shibire in PDF neurons. Shibire is an ortholog of the mammalian Dynamins and is essential for endocytosis of clathrin-coated vesicles at the plasma membrane. Such endocytosis is used for neurotransmitter reuptake by presynaptic neurons, which is a prerequisite of synaptic vesicle recycling, and receptor-mediated endocytosis in the postsynaptic neuron, which leads to signal termination. By blocking Shibire function via overexpression of a dominant negative mutant form of Shibire in PDF neurons, we slowed down the behavioral rhythm by 3 h. This effect was absent in PDF receptor null mutants, indicating that we interfered with PDF receptor-mediated endocytosis. Because we obtained similar behavioral phenotypes by increasing the PDF level in regions close to PDF neurons, we conclude that blocking Shibire did prolong PDF signaling in the neurons that respond to PDF. Obviously, terminating the PDF signaling via receptor-mediated endocytosis is a crucial step in determining the period of behavioral rhythms.