Targets of Activity

DCO, the catalytic subunit of Protein kinase A, is preferentially expressed in the mushroom bodies. PKA is the target of cAMP, degraded through enzymatic activity of the Dunce protein. PKA is a heterotetramer, consisting of two subunits of DC0 and two subunits of a regulatory subunit. Binding of cAMP to the regulatory subunits causes them to dissociate from the tetramer, activating PKA enzymatic activity. Mutants for DCO produce homozygous lethality and a 40% decrease in PKA activity in heterozygotes. This decrease has mild effects on learning but no effect on memory. However, the 80% reduction in activity obtained by constructing double mutant heteroallelic viable animals results in a dramatic learning and memory deficit. These results suggest that PKA plays a crucial role in the cAMP cascade in mushroom bodies to mediate learning and memory processes (Skoulakis, 1993).

There are several genes expressed at abnormal levels in the memory mutant, dunce. These mutants have an elevated cyclic AMP (cAMP) content due to a mutation in the structural gene for cAMP phosphodiesterase, so the isolated genes are potentially those regulated by cAMP. There are two uninterrupted and complete open reading frames (SER1 and SER2) and part of a third (SER3), all found to be homologous to the serine protease family of enzymes. Only the SER1-related genes were differentially expressed in dnc mutants. The putative serine protease genes are abundantly expressed in the larval gut, suggesting a major function in digestion. Feeding normal flies cAMP, isobutylmethylxanthine, or forskolin results in a decreased RNA level of the SER1-related genes. Thus, RNA levels of this serine protease gene family are negatively regulated through cAMP (Yun, 1989).

Hangover links nuclear RNA signaling to cAMP regulation via the phosphodiesterase 4d ortholog dunce

The hangover gene defines a cellular stress pathway that is required for rapid ethanol tolerance in Drosophila melanogaster. To understand how cellular stress changes neuronal function, Hangover function was analyzed on a cellular and neuronal level. Evidence is provided that Hangover acts as a nuclear RNA binding protein and the phosphodiesterase 4d ortholog dunce as a target RNA. A transcript-specific dunce mutant was generated that is impaired not only in ethanol tolerance but also in the cellular stress response. At the neuronal level, Dunce and Hangover are required in the same neuron pair to regulate experience-dependent motor output. Within these neurons, two cyclic AMP (cAMP)-dependent mechanisms balance the degree of tolerance. The balance is achieved by feedback regulation of Hangover and dunce transcript levels. This study provides insight into how nuclear Hangover/RNA signaling is linked to the cytoplasmic regulation of cAMP levels and results in neuronal adaptation and behavioral changes (Ruppert, 2017).

In mammals and insects alike, repetitive ethanol exposure leads to the development of tolerance. Tolerance is defined as an increased resistance to the behavioral effects of ethanol upon previous exposure. Tolerance is also used as a criterion to diagnose alcohol use disorders. Previous work has established Drosophila melanogaster as a suitable model system to analyze the molecular and neural bases of rapid ethanol tolerance. At least two different mechanisms contribute to the rapid development of tolerance. One requires the monoamine octopamine, which shares functional similarities to noradrenaline in vertebrates, whereas the other relies on a cellular stress response uncovered by the phenotypic characterization of hangover (hang) mutants. The hang gene encodes a large, approximately 210-kDa nuclear zinc finger protein that is expressed in most, if not all, neurons in the adult brain. A Hang-related protein, ZNF699, is associated with alcoholism in humans, supporting the idea that a cellular stress mechanism underlying ethanol tolerance is evolutionarily conserved between humans and D. melanogaster. hang mutants develop reduced ethanol tolerance and show defects in their response to oxidative stress and heat shock-induced ethanol resistance. However, the cellular signaling process upon which hang acts remains unclear, and how the broadly expressed hang protein mediates specific behavioral changes in response to global increases of cellular stressors such as ethanol remains an open question (Ruppert, 2017).

In neuronal cell lines, ethanol exposure reduces cyclic AMP (cAMP) levels, which results in cellular tolerance. cAMP-dependent phosphodiesterases (PDEs) are essential components of the cAMP signaling cascade and fine-tune cyclic nucleotide signaling. In Drosophila, three different cAMP-dependent PDEs have been identified; two of them play a dual role in cyclic guanosine monophosphate (cGMP) and cAMP signaling, whereas only Dunce (Dnc) is cAMP specific. The dnc gene encodes for at least eight isoforms with high homology to the human phosphodiesterase 4 class of PDEs. In general, PDE4s can be divided into two classes based on the presence or absence of the two highly conserved N-terminal domains. The long forms contain the upstream conserved region 1 and 2 (UCR1 and UCR2, respectively) domains, whereas the short forms lack the UCR1 domain and have either a partial or complete UCR2 domain. The dnc isoforms share a common PDE domain but differ in their N termini, suggesting that they may exhibit different regulatory and cellular functions. This hypothesis has been further supported by the phenotypic characterization of various dnc mutants. For example, dnc1 and dncM11 share similar initial learning and memory defects but differ at the level of the neuroanatomical structure of the larval neuromuscular junction. Nerve terminal growth is increased in dnc1 mutants, whereas the nerve terminals grow normally in dncM11 mutants. These phenotypic differences correlate with differences in PDE activity. dnc1 mutants show reduced PDE activity, and dncM11 mutants display no detectable PDE activity; as a result, the cAMP level is elevated in both mutants. While the function of PDE4s has been analyzed in terms of the cellular stress response to oxidative stress and in non-neuronal tissues such as pulmonary endothelial cells, the function of PDE4s in ethanol-induced cellular stress in neurons has not been addressed (Ruppert, 2017 and references therein).

This study provides evidence that Hang functions as a nuclear RNA binding protein and dnc is identified as a target gene. To confirm the role of the PDE4 ortholog Dnc in ethanol tolerance, the consequence of altered cAMP signaling was investigated by analyzing the dnc1 mutant for defects in ethanol-induced behaviors. Because multiple Dnc isoforms are altered in this dnc mutant, a new dncRA transcript-specific mutant, dnc&Delta143, was generated, enabling isoform-specific analysis of ethanol tolerance. To identify the neurons required for Dnc-dependent ethanol tolerance, a dncRA-specific promoter-GAL4 driver (dncRA-GAL4). Phenotypic characterization revealed that both the dnc&Delta143 and hang mutants exhibited similar impairment in the cellular stress response to ethanol. Both genes were required in the same neurons, enabling them to participate in the same cellular signaling process. In addition, DncPA regulates Hang function during the development of ethanol tolerance. Furthermore, cytoplasmic Dnc, rather than nuclear Dnc, is required for tolerance development. The identification of the PDE4 ortholog Dnc as an interaction partner of Hang links nuclear Hang function to the regulation of cytoplasmic cAMP levels. These results substantiate the conserved nature of the signaling process underlying the neuronal stress response to ethanol (Ruppert, 2017).

This study shows how the nuclear hang protein, which is broadly expressed in nearly all neurons of the CNS, mediates specific behavioral changes in response to a global increase in the stressor ethanol by identifying the neurons required for Hang-dependent ethanol tolerance. In these neurons, hang can bind RNA and modify the transcript levels of the phosphodiesterase 4d ortholog dnc, linking nuclear signaling to the regulation of cAMP levels. Consistently, hang mutants show increased cAMP levels. In addition, this study shows that tolerance development requires the presence of the specific dnc isoform DncRA in the cytoplasm of the same neurons. The functional relationship between dnc and hang depends on feedback regulation that balances the degree of tolerance formed (Ruppert, 2017).

Although emerging evidence in mice suggests that pharmacological inhibition of PDE4s reduces alcoholism-associated behaviors such as ethanol intake and preference, what type of cellular process this inhibition of PDE4 activity is embedded in and whether this intervention acts on a neuronal level remains completely unclear. Every cell in organisms uses cAMP signaling as a second messenger system, and during ethanol exposure, the concentration of ethanol increases more or less everywhere in the brain. Nevertheless, the specificity of cAMP signaling for Hang- and DncPA-dependent tolerance development can be ascribed to a pair of F1 neurons. In the neurons, other participants in the cAMP signaling cascade, such as the adenylyl cyclase Rutabaga, are present and functional, as shown by the fact that Rutabaga expression in the F1 neurons ameliorates visual learning defects in those mutants. The specific requirement of cAMP signaling in these neurons for tolerance might be linked to the more general function of the F1 neurons in motor control. At first glance, visual memory learning, particularly contour learning, and ethanol tolerance may not seem to have much in common because their sensory input differs. However, learning is measured as a change in the direction of flight of a tethered fly, linking visual information processing to motor output. In addition, there is evidence that the activity of the F1 neurons is required for the maintenance of high walking motivation, again linking the function of these neurons to motor output regulation. Behavioral ethanol tolerance is also measured by analyzing changes in motor output, such as the ability to maintain or regain posture after a previous ethanol experience. The dnc&Delta143 mutants showed no difference in ethanol sensitivity, and therefore, the basic requirements of the behavior are normal. Considering the other observed functions of F1 neurons in locomotion, a more general function for F1 neurons is proposed that involves linking previous motor experience to new information and changes in behavioral output. All cells within the brain are exposed to an increasing ethanol concentration and might respond with a cellular stress response, but only defects in neurons in the specific neuronal networks uncovered here impact performance on the assayed task (Ruppert, 2017).

At the behavioral level, the cAMP second messenger system has been implicated in the regulation of ethanol sensitivity in both mammals and invertebrates. The Drosophila cAMP mutant rutabaga ethanol sensitivity increase (Moore, 1998) correlates with reduced cAMP levels (Cheung, 1999). Therefore, the increase in cAMP levels caused by the loss of PDE function should result in increased resistance to ethanol; however, neither the dnc1 nor the dncΔ143 mutants, which have increased cAMP levels, showed a defect in ethanol sensitivity. In addition, other dnc mutants, such as dncM11, show normal ethanol sensitivity (Moore, 1998). To date, a role for dnc in sensitivity seems likely, given the increased sensitivity of rutabaga mutants; however, this role might not depend on the analyzed isoforms, as the tested dnc mutants showed normal ethanol sensitivity. In addition to its potential role in the regulation of ethanol sensitivity, Dnc-dependent cAMP function is required for the regulation of rapid ethanol tolerance. The dncΔ143-like dnc1 mutants showed reduced ethanol tolerance but normal ethanol sensitivity, supporting the dissociation between cAMP function in ethanol sensitivity and tolerance. On a cellular level, differences in cAMP regulation might result from subcellular differences in PDE activity due to differences in PDE localization. The mouse PDE4D gene, which exhibits the highest homology with dnc, encodes at least four isoforms that are expressed in different subcellular domains. In addition, in the mollusk Aplysia, three different PDE4 isoforms localize to different subcellular membranes. DncPA function in a specific subcellular compartment during tolerance development is consistent with the finding that nuclear DncPG did not improve the reduced tolerance of the dncΔ143 mutants. DncPA appears functionally similar to Dncall because both isoforms restored or improved the ethanol tolerance of the dncΔ143 mutants (Ruppert, 2017).

At first glance, the prolongation of cAMP signaling due to the elevation of cAMP levels seems to reduce ethanol tolerance. Consistently, the dncΔ143 and hangAE10 mutants, with reduced tolerance, had elevated cAMP levels. These observations indicate that under normal circumstances, the termination of cAMP signaling plays a critical role in determining the level of tolerance formed. However, the true situation is more complex because the level of tolerance in the hangAE10, dncΔ143 double mutants with increased cAMP levels was normal. Therefore, the dynamics of the cAMP changes are more crucial than the absolute cAMP level. The importance of dynamic changes of cAMP levels was already implicated by the observed similar impairments of learning and memory in rutabaga mutants with reduced cAMP levels and dnc mutants with increased cAMP levels. The dynamical change in ethanol tolerance might be achieved through the regulation of each gene's transcript level. In the hangAE10 mutants, dnc was reduced, whereas temporally restricted expression of hang in the adult brain in a dncRA-GAL4-dependent manner increased tolerance and dnc levels, consistent with the idea that hang is a positive regulator of dnc transcript levels. In contrast, DncRA is a negative regulator of hang expression because in the dncΔ143 mutants, hang was upregulated. The existence of a feedback regulation between dunce and hang is further supported by the observation that overexpression of hang in a dncRA-GAL4 dependent manner produces similar changes in dncRA- transcript levels and tolerance as found in the dncΔ143; hangAE10 double mutants. However, the interaction is even more complicated. Overexpressing hang throughout development was lethal, and temporally restricted hang expression in the adult increased tolerance. That hang function is extremely dose sensitive was also demonstrated by previous results showing that the loss or gain of hang function at the NMJ results in similar morphological defects. Taken together, these results indicate that there is complex feedback regulation between hang and dnc (Ruppert, 2017).

In summary, at the cellular/systems level, a reduction in cAMP level is required for behavioral tolerance. Crucial regulators at the cellular level include two conserved proteins, the PDE4d ortholog dnc and the cellular stress regulator Hang. The subcellular function of dnc and spatially and temporally controlled cAMP levels is important. cAMP regulation is linked via a feedback mechanism to the nuclear RNA binding protein hang under tight temporal control. Furthermore, the identification of the requirement of hang and the PDE4d ortholog dnc in the same neurons for regulation of experience-dependent motor output in Drosophila provides insight into how specific behavioral changes are achieved in response to global increases in cellular stressors such as ethanol (Ruppert, 2017).

dunce: Biological Overview | Evolutionary Homologs | Developmental Biology | Effects of Mutation | References

Home page: The Interactive Fly © 1997 Thomas B. Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.