|Home | About | Journals | Submit | Contact Us | Français|
The mechanistic basis of arousal is controversial. A new study in Drosophila, where dopamine has been shown to be involved in several types of attentional processes, demonstrates that it independently regulates distinct types of arousal. These data provide evidence for molecularly convergent but anatomically divergent, task-specific, arousal circuits.
The idea that arousal state affects the ability of an animal to carry out complex tasks was articulated over 50 years ago. Since that time our theories about the nature of attention have ranged from thinking about it as a unitary phenomenon that can influence multiple subsystems (Pfaff et al., 2005), to the idea that there are multiple independent types of attention (Hebb, 1955; Parasuraman, 1998). This debate is particularly salient when one considers the role of neuromodulators like dopamine in attention. This catecholamine (along with several other compounds: norepinephrine, serotonin and acetylcholine) has been shown to be involved in diverse motivational and arousal processes in many organisms. In mammals, neocortical dopaminergic pathways influence the activity of prefrontal cortex which is critical for focus and selection (Boulougouris and Tsaltas, 2008). The diffuse nature of the projections of this pathway makes the anatomy relatively uninformative on the question of whether arousal is a single generalized process or a set of specialized processes. Pharmacological studies can to some extent subdivide the roles of particular receptor subtypes, but again the broad expression of each receptor subtype makes it difficult to resolve this issue.
In Drosophila, the situation is similarly complicated. Dopaminergic neurons and receptors are found in many brain areas. Mutations and drugs that affect dopamine signaling have been shown to have an impact on a broad range of behaviors that have attentional components: locomotion, learning, sleep, sexual performance, mechanosensory startle and visual tracking (Andretic et al., 2005; Bainton et al., 2000; Kim et al., 2007; Kume et al., 2005; Ye et al., 2004). The evidence that dopamine is involved in the regulation of basal locomotor activity and its distribution over the sleep/wake cycle was clear. In contrast, the evidence in Drosophila for the involvement of dopamine in arousal produced by environmental stimuli was confusing. While it was apparent (using many methodologies) that altering dopamine levels or the activity of dopamine-producing neurons could affect the response of animals to startle, the results were not totally consistent over different paradigms. The overall picture that emerged suggested that dopamine might be inhibitory for startle-induced locomotion, but the stimulatory effects of dopamine on basal activity made unambiguous interpretation of experiments difficult. How to resolve this? One proposition that has been around for a long time is that the generalized pro-arousal effects of dopamine have an inverted U-shaped dose-response curve: both too little and too much dopamine suppress arousal. The opposite sign effects of dopamine on different processes were interpreted as a function of this curve. An alternative explanation is that there are multiple arousal circuits that can be affected in opposite ways by dopamine and the level of engagement of these individual circuits determines the attentional state(s) of the fly.
In this issue of Neuron, Lebestky et al. (Lebestky et al., 2009) take a broad view of this problem and provide some resolution. Based on the observation that repetitive delivery of a strong startle (in this case a puff of air that blows the fly against the wall of a tube) can cause a sustained period of hyperlocomotion, they develop a new automated assay that can capture changes in what they term Repetitive Startle-induced Hyperactivity (ReSH). By a number of criteria, including cross-over of sensitization of the response to other sensory modalities, ReSH appears to be a stress-induced enhancement of arousal state. Using this assay as a platform for gene discovery they isolated a mutant in the Drosophila gene encoding the D1 receptor homolog, DopR. Complete loss of this gene product did not affect viability, allowing the authors to assess the behavior of adult animals that lack the D1 receptor and to do selective rescue to dissect the circuit(s).
DopR mutants were abnormal in several ways in the ReSH assay. First, they had increased basal locomotion, i.e. the velocity of DopR flies before they received an air puff was significantly greater than wild type. More interesting, however, was what happened after the series of air puffs. DopR flies showed a prolongation of hyperactivity that was not due to the alteration in baseline activity. Scaling the mutant and wild type responses showed that the rate of decay in the DopR mutants was clearly much slower. This was a dopamine-specific effect. When wild type animals were fed cocaine, a drug whose actions are believed to be the result of enhancing dopaminergic transmission by blocking its reuptake from the synaptic cleft, the rate of decay of ReSH was faster, and this cocaine effect was blocked by mutation of DopR. These data indicated that dopamine, via the DopR D1 receptor, normally functions to suppress stress-induced arousal.
There were several things that were still confusing, however. The increase in basal locomotor activity in flies with no D1 receptors was puzzling in light of previous work indicating that dopamine had a stimulatory effect on locomotion- a receptor mutant would be expected to have lower basal locomotion. The authors investigated this apparent contradiction in several ways. First, they looked at spontaneous activity over many days using a beam-break assay. Like humans, flies are diurnal. They have a circadian pattern of locomotor activity with highest levels of activity at dawn and dusk and lowest levels of activity during the night. This nighttime decrease in locomotor activity level is often used as a proxy for sleep, which is defined as periods of inactivity lasting more than 5 minutes associated with increased arousal threshold and altered gross brain activity (Shaw, 2003). The authors found that animals that were either heterozygous (DopR/+) or homozygous (DopR/DopR) for the D1 mutation had lower levels of locomotor activity at night, suggesting that DopR-positive neurons were involved in stimulation of spontaneous nighttime activity. The specificity of the DopR mutation’s effects was demonstrated in a dramatic manner when the authors fed the animals cocaine. Wild type animals were significantly more active when fed cocaine. Remarkably, cocaine exposure produced a level of locomotor activity at night was even higher than daytime activity. This effect was totally blocked by mutation of DopR, suggesting that 100% of cocaine’s profound basal locomotor effects were mediated by interaction of dopamine with this receptor.
These data were consistent with the literature in indicating dopamine (here shown to act via DopR) provides a pro-arousal signal, but they apparently contradicted their own ReSH data that showed elevated spontaneous activity in the mutant. Why did the two assays give different results with regard to spontaneous activity? The answer the authors posit is interesting both from the point of view of dopamine function and from the point of view of how one interprets behavior experiments. The salient difference between the two assays turned out to be that spontaneous activity was measured at different times relative to when the flies were handled and put into the apparatus. In the circadian assays, flies are loaded and typically entrained to a light/dark cycle before data are collected. In the ReSH assay, flies are loaded into tubes and allowed to settle for only 10 min. Given that DopR mutants have a prolonged hyperlocomotor response to stress, the authors hypothesized that this higher initial velocity was due to long-lasting residual effects of handing, an unintended stressor. Consistent with this, allowing the flies to acclimate for 30 instead of 10 min significantly decreased the basal locomotor velocity but did not affect the ReSH response. So caveat spectator: your assay can measure things other than what you expect. Luckily, those things are sometimes interesting.
Having successfully sorted out the different phenotypes of this mutant for exogenously and endogenously produced locomotion, the authors turned their attention to the neural circuits that produced the behaviors. Drosophila offers an ideal system for this type of question. Using the rich collection of genetic tools that have been characterized in this system, investigators can easily and selectively “add back” a gene product to a mutant and ask exactly which cells in the brain require that gene product for the behavior they are interested in (Figure 1). Locomotor activity is known to be influenced by several brain areas in the fly, notably the mushroom bodies, a region previously shown to require DopR for learning (Kim et al., 2007). Interestingly, DopR expression in mushroom bodies was unable to rescue ReSH. Only expression of DopR in the ellipsoid bodies, a subregion of the central complex, was sufficient to restore normal dopaminergic suppression of hyperactivity. Surprisingly, spontaneous nighttime activity was not restored by expression of DopR in this brain region. The only brain area where expression had a partially restorative effect corresponded to a small number of cells that are part of the circadian clock, the ventral lateral neurons. These cells have been shown previously to be involved in arousal (Shang et al., 2008; Sheeba et al., 2008), but it seems likely that there are other DopR-positive cells that participate in regulation of spontaneous activity that remain to be identified.
Lebestky et al. make a good case that one of the important chemical mediators of attention, dopamine, is not a one-size-fits-all neuromodulator: it can act to enhance some types of arousal while suppressing others. Attention and the molecules that mediate it are quite clearly ancient- this is not surprising since selecting relevant stimuli to respond to in complex situations has survival value. Independent arousal circuits may serve to allow the animal more flexibly in managing responses to the varied challenges of navigating the natural world.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.