Altogether, these results establish that the baseline activity of dorsal FB neurons and its homeostatic modulation are disrupted in cv-c mutants, resulting

in deficient sleep. A homeostat (or, in engineering terms, a controller) senses the operation of a system, compares the measured behavior to a set point, computes the necessary corrective action, and actuates the system in order to effect the desired change (Åström and Murray, 2008). The experiments reported here identify sleep-promoting neurons of the dorsal FB as the effector arm of the sleep homeostat. Molecular machinery within these neurons transduces sleep pressure PD98059 in vitro into increased sleep. The output of this machinery, which includes the Rho-GAP encoded by the cv-c gene ( Figures 3 and 4), is the modulation of membrane excitability ( Figures 6 and 7). This modulation appears to be coordinated across the small population of sleep-promoting neurons with

axonal projections to the dorsal FB ( Figure 6). Spiking of these neurons is, in itself, sufficient for the induction of sleep ( Donlea et al., 2011). Given that spike generation is a threshold process, even modest modulation of excitability could result in an effective on/off switch. Sleep has long been associated with widespread changes in neuronal activity (Brown et al., 2012, Steriade et al., 1993, Steriade et al., 2001, Vyazovskiy and Harris, 2013 and Vyazovskiy et al., 2009). Z-VAD-FMK ic50 In sleeping mammals, these changes are reflected in characteristic patterns of extracellular field potentials (Brown et al., 2012, Steriade et al., 1993 and Vyazovskiy and Harris, 2013). Several features distinguish these changes from those that are key to the operation of the fly’s sleep homeostat. First, homeostatic sleep control involves a localized increase in the excitability of a circumscribed

cluster of sleep-promoting neurons (Figures 6 and 7). This localized gain in electrical responsiveness is in sharp contrast to the diffuse “down” states or “off” periods of reduced activity that commonly accompany non-rapid-eye-movement sleep (Steriade et al., 2001 and Vyazovskiy et al., 2009). Second, the homeostatic gain in excitability is a cause and not a consequence of sleep, given that molecular Dichloromethane dehalogenase lesions that prevent it lead to insomnia (Figures 1 and 3). No such causal link has been established for any of the other excitability changes that coincide with sleep. The regulatory logic upstream of the action potential output of the sleep-control neurons is currently unknown. A priori, the remaining elements of the homeostatic feedback loop—sensors, set point, and comparator—could also be housed within the dorsal FB neurons. It is conceivable that these neurons monitor byproducts of prolonged wakefulness, such as changes in the strength of afferent synapses (Tononi and Cirelli, 2003), the release of adenosine (Brown et al., 2012 and Porkka-Heiskanen et al.