Dietary modulation of Drosophila sleep-wake behaviour

Abstract

Background: A complex relationship exists between diet and sleep but despite its impact on human health, this relationship remains uncharacterized and poorly understood. Drosophila melanogaster is an important model for the study of metabolism and behaviour, however the effect of diet upon Drosophila sleep remains largely unaddressed.

Methodology/principal findings: Using automated behavioural monitoring, a capillary feeding assay and pharmacological treatments, we examined the effect of dietary yeast and sucrose upon Drosophila sleep-wake behaviour for three consecutive days. We found that dietary yeast deconsolidated the sleep-wake behaviour of flies by promoting arousal from sleep in males and shortening periods of locomotor activity in females. We also demonstrate that arousal from nocturnal sleep exhibits a significant ultradian rhythmicity with a periodicity of 85 minutes. Increasing the dietary sucrose concentration from 5% to 35% had no effect on total sucrose ingestion per day nor any affect on arousal, however it did lengthen the time that males and females remained active. Higher dietary sucrose led to reduced total sleep by male but not female flies. Locomotor activity was reduced by feeding flies Metformin, a drug that inhibits oxidative phosphorylation, however Metformin did not affect any aspects of sleep.

Conclusions: We conclude that arousal from sleep is under ultradian control and regulated in a sex-dependent manner by dietary yeast and that dietary sucrose regulates the length of time that flies sustain periods of wakefulness. These findings highlight Drosophila as an important model with which to understand how diet impacts upon sleep and wakefulness in mammals and humans.

Figure 1. Effects of dietary yeast extract on sleep-wake behaviour of male Drosophila.

Male w¹¹¹⁸ flies were housed in polycarbonate tubes and provided with agar containing 5% sucrose (AS) or agar, 5% sucrose and 2% yeast extract (ASY). Locomotor activity was recorded as the number of times a fly broke the path of an infra-red beam at the midpoint of the tube. Periods of 5 minutes without beam crossing were regarded as a single period of sleep. (A) Actograms showing averaged beam crossing data for three consecutive days (d2, d3 and d4). The light and shaded areas denote the 12-hour periods of light and darkness of the 24-hour cycle. A single day's data is re-plotted on the following line so that the relationship between the light-dark cycle and the rhythm of locomotor activity can be seen better. (B) Averaged beam crossing data binned to every half-hour of the 24-hr cycle for flies fed the different diets; dark bars represent night and day activity, respectively. The grey bar denotes an average of 2 beam crosses per half hour and is presented to aid comparisons of the data. (C) Minutes of sleep per 30 minutes. The white and black bar represents the 12 hour light and dark phases of the 24 hour cycle. (D) Total amount of time spent asleep for flies fed the agar-sucrose (AS) and agar-sucrose-yeast (ASY) diets. (E) The average length of each sleep bout by flies fed the different diets. (F) The mean number of sleep bouts. (G) The amount of locomotor activity undertaken by flies on the different diets. (H) Average length of activity bouts. *P<0.01; #P<0.05; n = 32 flies for each diet in A, B and C; rest of data is n = 100 flies for each diet, data was pooled from five independent trials.

Figure 2. Effects of dietary yeast extract on sleep-wake behaviour of female Drosophila.

(Legend as for figure 1). *P<0.01; #P<0.05; n = 48 flies for each diet, data pooled from two independent trials.

Figure 3. Dietary yeast affects arousal.

The behaviour of male flies was assessed as described in Figure 1 and the Methods section. (A) The graph shows the fold change in sleep promoted by 5 mg/ml 3IY, an inhibitor of dopamine synthesis, when supplemented in either an agar-sucrose (AS) or yeast-containing AS diet (ASY). (B) CAFE data showing that amount of food ingested after five days on the AS and ASY diets, supplemented with the dopamine synthesis inhibitor, 3IY. (C) Average percentage of flies fed the AS diet or the ASY diet (black and grey line, respectively) aroused in response to a mechanical provocation (arrow) performed at ZT16. (D) Quantified data for the provocation test. #P<0.01, *P<0.01; CAFE assays used 9–10 flies per treatment; for the arousal provocation tests (C and D) two independent trials were performed involving 16 flies for each treatment.

Figure 4. Arousal from sleep is ultradian and disrupted by dietary yeast.

The time of day that a male fly aroused from a sleep bout was recorded and the length of that sleep bout was calculated and plotted as a 1-hour running average over 2.5 consecutive days. (A) Data from 250 different male flies from five independent experiments is presented. During the daytime there is no obvious ultradian rhythm of arousal, whereas at night there are several discrete epochs during which flies are more or less likely to arouse from sleep. (B) The effect of the AS (black line) and ASY (gray line) diets upon the ultradian pattern of nocturnal arousal. Data for the maximum length of sleep bouts is presented. Flies fed the AS diet show an ultradian rhythm of arousal, whereas, flies on the ASY diet have similar periodicity ultradian rhythm of arousal but lower peaks of sleep length due to shorter periods of sleep on the ASY diet (arrows). (C) Periodogram analysis of the data combined from all experiments (n = 360 male w¹¹¹⁸ flies) established that arousal occurs according to a significant ultradian rhythm with a periodicity of 90 minutes. (D) Periodogram showing a significant ultradian rhythm of nocturnal arousal for male flies fed sucrose only. (E) The periodogram for male flies fed sucrose and yeast extract shows a significant ultradian rhythm of 85 minutes and a trend towards a longer frequency rhythm, of approximately 130 minutes, that does not reach significance. The black and white bar denotes the 12-hour light and dark periods. n = 189 and 180 flies from independent experiments for the AS and ASY diets, respectively.

Figure 5. Effect of dietary sucrose on male Drosophila sleep-wake behaviour.

Male w¹¹¹⁸ flies were fed diets containing different sucrose concentrations and their sleep-wake behaviour monitored for three consecutive days. (A) Activity plots binned to every half-hour of the 24-hr cycle for flies fed the different diets; dark bars represent night and day activity, respectively. The grey bar denotes an average of 2 beam crosses per half hour and is presented to aid comparisons of the data. (B) Sleep (periods of 5 minutes without a fly crossing the beam) of flies on the two diets. (C) Length of sleep bouts. (D) The number of sleep bouts. (E) The amount of locomotor activity undertaken by flies on the different diets. (F) Average length of activity bouts. (G) Intensity of activity bouts. Grey boxes refer to 24 hr data, open boxes refer to the 12 hour light period and filled boxes relate to data collected during the 12-hour dark period. *P<0.01; #P<0.05; n = 32 flies for each diet and is representative of four independent trials.

Figure 6. Effect of dietary sucrose on female Drosophila sleep-wake behaviour.

Female w¹¹¹⁸ flies were fed diets containing different sucrose concentrations and their sleep-wake behaviour monitored for three consecutive days. (A)–(G) Legend as for figure 5. *P<0.01; #P<0.05; n = 15–16 flies for each diet and is representative of two independent trials.

Figure 7. The effect of Metformin on sleep-wake behaviour, basal ATP and food intake.

(A) Flies were provided with agar based diet containing 30% sucrose, supplemented with Metformin at the stated dose. The three charts show mean (±SEM) beam crosses from three days of behavioural monitoring. The arrows are identically placed within each chart and point to the decrease in morning activity. (B) Locomotor activity (beam crosses) on diets containing difference concentrations of Metformin. (C) Length of each activity bout. (D) Intensity of locomotor activity. (E) Total amount of sleep. (F) Length of each sleep bout. (G) Basal ATP in flies. (H) Food intake by flies on control (filled boxes) or 10 mM Metformin (open boxes), as measured by CAFE assay. In B–F open and filled boxes relate to data collected during the day and night of the 12 hr∶12 hr light-dark cycle, respectively. Data are expressed as the mean (±SEM) and is pooled from two impendent trials using a total of 32 flies for each treatment. *P<0.01, # = P<0.05, ns  =  not significant.