Increased Volatile Anesthetic Requirement in Short-sleeping Drosophila  Mutants

Drosophila melanogaster  were bred in the laboratory at 21°C, 68% humidity, on yeast, dark corn syrup, and agar food. For determination of sleep and wakefulness, male and female fruit flies were used in equal numbers. To exclude age-associated effects, only young flies (≤ 2 weeks) were tested for all experiments.5,Drosophila  stocks used were Sh^mns , Sh¹⁰², Sh¹²⁰  Na[har³⁸] , and wild-type Canton-S. To remove modifiers, stocks were consequently outcrossed for at least five rounds to Canton-S background as described before.4Canton-S is not known to be resistant to volatile anesthetics. Sh^mns , Sh¹⁰² ), and Sh¹²⁰ ) are different mutant alleles of the Shaker  locus, encoding the alpha subunit of a tetrameric voltage-dependent potassium channel.6The Na  gene encodes a sodium leak channel,7which exerts opposite effects on excitability to the Shaker  gene; e.g., Na[har³⁸]  is known to be hypersensitive to volatile anesthetics.8 

Sleep and wakefulness were determined from individual fruit flies placed in a Drosophila  activity monitoring system (Trikinetics, Waltham, MA) at constant environmental conditions. Activity measurement was recorded for consecutive 1-min periods for 1week after 1day of adaptation, and analyzed with custom-designed software developed in our laboratory. As described before, sleep was defined as any period of uninterrupted behavioral immobility (0 counts per minute) lasting > 5 min.4,9,10The total duration of sleep episodes was then calculated exactly to the minute.

Anesthetic sensitivity was tested in a custom-made Drosophila  anesthesia chamber (V = 200 ml) connected to isoflurane or sevoflurane vaporizers, respectively, with a constant flow of 1.6 l/min. For each experiment, at least 10 young (≤ 2 weeks) wild-type or mutant strain fruit flies were placed inside the chamber and exposed to distinct anesthetic concentrations. After a 10-min exposure the chamber was rotated and shaken for 2 s under the control of a motor, which caused the flies to fall from their current position to the bottom of the chamber. With this accepted method of sleep deprivation,10we were able to distinguish between sleep and anesthesia. The numbers of mobile and immobile flies were counted by a blinded observer, but a convulsion was not considered movement. The results were recorded for subsequent statistical analysis. All experiments were carried out at constant environmental temperature of 21°C, and concentrations of the volatile anesthetics were continuously monitored at the chamber outflow with a Datex-Ohmeda Capnomac Ultima (GE Healthcare, Chalfont St. Giles, England).

A Student t  test was used to assess statistically significant differences for periods of sleep and wakefulness between Drosophila  strains. Based on the response of the flies at different concentrations of isoflurane and sevoflurane, concentration-response curves were generated according to the method of Waud for quantal biologic responses.11The half-maximum effective concentration (EC⁵⁰) values and 95% CIs were calculated and compared for statistically significant differences using GraphPad Prism version 4.03 for Windows (GraphPad Software, La Jolla, CA).

For determination of locomotor activity, male and female Drosophila  were used in equal numbers. As described before, the duration of sleep and wakefulness was different in wild-type Drosophila  and Shaker  mutants.4The average amount of daily sleep in wild-type Drosophila  (n = 64) was 965 ± 15 min (mean ± SEM), as compared with 584 ± 13 min for Sh^mns  flies (n = 64, P < 0.01), and as compared with wild-type Sh¹²⁰  and Sh¹⁰² ; 412 ± 22 min for Sh¹⁰² ) flies (n = 32, all P < 0.01) and 782 ± 25 min for Sh¹²⁰ ) (n = 32, all P < 0.01). Thus, the short-sleeping phenotype was most pronounced in Sh¹⁰² ), moderately less expressed in Sh^mns  and weakest in Sh¹²⁰ ). Na[har³⁸]  showed a sleeping phenotype comparable to wild-type (1,022 ± 29 min, n = 32, P > 0.05).

Response of different Drosophila  strains to the volatile anesthetics isoflurane and sevoflurane measured at various concentrations ranging from 0.13 to 5% for isoflurane and from 0.21 to 4% sevoflurane, respectively, yielded specific concentration-response curves. The EC⁵⁰values for both volatile anesthetics, isoflurane and sevoflurane, were significantly increased statistically in fruit flies expressing the short-sleeping phenotype, and decreased in Na[har³⁸] , as compared to wild-type Drosophila . Moreover, EC⁵⁰values for isoflurane and sevoflurane were associated with the severity of the short-sleeping phenotype. The differences in the anesthetic requirement of Sh^mns , Sh¹⁰² ), Sh¹²⁰ ), and Na[har³⁸]  were also found to be statistically significant. The results for isoflurane and sevoflurane are summarized in tables 1 and 2, respectively. Typical concentration-response curves are shown in figure 1.

The main findings of our study are that mutations that cause a short-sleeping phenotype in Drosophila  are also responsible for a significant difference in anesthetic requirement, and that the quantity of volatile anesthetic required to anesthetize Drosophila  is associated with the severity of the short-sleeping phenotype.

Although sleep and general anesthesia are not identical, there has been increasing consensus that both states are neurophysiologically related. It has been shown that anesthetics may act partly by duplicating activities of brain regions important in initiating or maintaining sleep, and the effects on regional neuronal activity suggest activation of endogenous sleep-promoting pathways.12–14Sleep deprivation potentiates anesthetic-induced loss of the righting reflex, and anesthetic agents increase sleep when administered into brain regions known to regulate sleep.15,16In addition, the neurotransmitter adenosine increases the sleep requirement, enhances anesthetic potency, and delays recovery from halothane anesthesia.17–19Animal experiments suggest that anesthetic agents induce loss of consciousness, at least in part, via  activation of endogenous, nonrapid eye movement, sleep-promoting hypothalamic pathways.20,21Differences in the anesthetic sensitivity to volatile anesthetics have been reported for several mutations in genes affecting ion channels, and neurotransmitters and their receptors in Caenorhabditis elegans  and Drosophila .22–25However, until now there have been no reports of common mechanisms of naturally occurring sleep and anesthesia on a molecular level.

Drosophila  is an ideal model for investigating mechanisms involved in anesthesia in humans, as these flies have a complex nervous system and possess many of the same ion channels, neurotransmitters, and neurotransmitter receptors as vertebrates. Recently, some of us identified Sh^mns , a Drosophila  strain exhibiting an extreme reduction in sleep requirement, as compared with wild-type flies. We also found that other severe loss-of-function mutations of Shaker , including Sh¹⁰² , were short sleepers, while weak hypomorph alleles such as Sh¹²⁰  show only little variance.4Previous electrophysiological and molecular studies found that the Shaker  current and a normal-sized protein product were completely absent in short-sleeping mutants such as Sh¹⁰² , whereas in Sh¹²⁰  mutants the Shaker  current is present, although reduced.26,27With the present study, we demonstrate that a single-gene mutation affecting sleep regulation in Drosophila  is also associated with an increased anesthetic requirement in these fruit flies. Moreover, the severity of the short-sleeping phenotype among different alleles was consistent, with an increased anesthetic requirement in Drosophila . The fact that the hypersensitive strain Na[har³⁸]  does not show a significant long-sleeping phenotype underscores the relationship between the Shaker  gene, sleep, and anesthetic requirement. Moreover, it should be mentioned that other authors have identified Na[har³⁸]  as a long- sleeper.28This might be as a result of differences in the presence of genetic modifiers.

In contrast to intravenous anesthetics and opioids that have been shown to exert their anesthetic and analgesic properties mainly because of specific receptor-ligand interactions, the mechanism of action of volatile anesthetics remains largely elusive. In this study we showed that a mutation in a voltage-gated potassium channel powerfully affects the anesthetic requirement of Drosophila . Shaker  controls membrane repolarization after action potentials and presynaptic transmitter release.6Neurotransmitters and their receptors have been well conserved during evolution, and homologous ion channels in vertebrates have similar properties.29Also, our study and previous quantitative comparisons of the EC⁵⁰values of volatile anesthetics reveal an impressive correlation between Drosophila  and humans,3although the EC⁵⁰calculated after a 10-min exposure may reflect a complex mixture of pharmacokinetic and pharmacodynamic effects of the mutation. Furthermore, it is important to know that looking at different anesthetic endpoints in Drosophila  may lead to completely different results.2,30 

Our findings may have implications for at least two reasons: They demonstrate a link between sleep and anesthesia on a molecular level, and they show that a single-gene mutation can have a drastic effect on the susceptibility to volatile anesthetics.

The authors thank Olaf Wendt (Precision Mechanic, Christian-Albrechts- University of Kiel, Kiel, Germany) for constructing and producing the Drosophila  anesthesia chamber.