Research ArticleCOGNITIVE NEUROSCIENCE

Most sleep does not serve a vital function: Evidence from Drosophila melanogaster

See allHide authors and affiliations

Science Advances  20 Feb 2019:
Vol. 5, no. 2, eaau9253
DOI: 10.1126/sciadv.aau9253
  • Fig. 1 Great variability in sleep amounts in a nonmutant population of D. melanogaster.

    Descending sorted distribution of sleep amount (A) in a group of 881 female CantonS flies and (B) in a group of 485 male CantonS flies. In both panels, the left graph shows sleep amount for each individual fly over a period of 5 days in bouts of 30 min [legend in (A)]. The right graph indicates the average sleep amount in 24 hours for female [pink in (A) and the rest of the figures] and male [cyan in (B) and the rest of the figures] flies. The 19 female animals whose sleep is highlighted in red are the ones for which raw video sample is available at the ZENODO repository (37). (C) Average sleep amount measured in a tube predicts sleep amount measured in a different tube. Average of 6 days for both, with 1 day in between (nmale = 242 and nfemale = 242).

  • Fig. 2 Micromovements account for the newly described short-sleeping phenotype.

    (A) Average occurrence of behavior over the 24-hour period in male (top) and female (bottom) CantonS flies. (B) Sleep amount for each individual male (cyan) and female (pink) fly plotted as computed with ethoscopes (y axis) and with virtual Drosophila Activity Monitor (vDAM) analysis (x axis) (31). The size of each dot represents the average amount of micromovements observed over the 24-hour period. (C) Average sleep amount over the 24-h period in male and female flies, plotted as computed with ethoscopes (continuous lines) or virtual DAM analysis (dashed lines). (D) Average positional distribution of behaviors for male (left) and female (right) flies over the 24-hour period, broken into the three behavioral states identified by ethoscopes. (E) Four-dimensional representation of behavioral transitions over the 24-hour period. Gray shades indicate the dark period (ZT12 to ZT24), while red shades indicate the light period (ZT0 to ZT12). Same dataset shown in Fig. 1 (A and B).

  • Fig. 3 Mating reduces sleep amount.

    (A) Sleep profile of all the female flies used in the mating experiment: green, flies that underwent successful mating event (n = 86); gray, flies that met a male but did not engage in copulation (n = 152). The light blue vertical shade indicates the timing of the mating event. (B) Average position along the tube of the same flies shown in (A) in 30-min bins. (C) Four-dimensional representation of behavioral transitions over the 24-hour period for nonmated flies (gray background), mated flies (green background), and naturally short-sleeping unmated flies (pink background; same dataset highlighted in gray in Fig. 1A). (D) Hierarchical clustering based on pairwise distance, in the time-behavior domain, of the same three cohorts shown in (C).

  • Fig. 4 Chronic mechanical sleep deprivation is largely not lethal in D. melanogaster.

    (A) Lifelong sleep restriction in male (top) or female (bottom) CantonS flies subjected to mechanical sleep deprivation triggered by a 20-s inactivity bout. (B) Survival curve for male (cyan) or (C) female (pink) sleep-deprived flies and their sex-matched undisturbed mock control [gray in both (B) and (C)]. Sleep measurements become noisier as the number of flies decreases. n = 38 to 40 for all four groups. (D) Linear regression analysis in search of a correlation between sleep amount (x axis) and life span (y axis) in individual undisturbed female (pink) and male (cyan) flies. Same dataset as the gray flies in (A) and (B).

  • Fig. 5 Sleep rebound is not linearly proportional to sleep loss.

    (A and E) Sleep profile for the entire dataset: 818 male (A to D) and 912 female (E to H) CantonS flies. (B and F) Sleep (cyan and pink dots and black markers) or immobility (gray markers) for the entire dataset spanning 10 different immobility interval triggers (20 to 1000 s). Control flies were never actively stimulated but laid adjacently to the experimental flies. (C and G) Number of tube rotations triggered by immobility bouts. (D and H) Amount of rebound sleep in the ZT0 to ZT3 interval following the sleep deprivation for the entire dataset.

  • Fig. 6 Sleep pressure is largely under the control of the circadian rhythm.

    (A and B) Sleep profile for male (A, cyan) and female (B, pink) CantonS flies during the length of the experiment compared with their sex-matched undisturbed mock controls (gray in both). Day 0 marks the beginning of the chronic sleep deprivation procedure, lasting 228 hours (indicated by a purple shade on top). The green shade indicates the rebound day blown up in (C) and (D). The three yellow shades mark the timings chosen for the CaLexA quantifications shown in (G). (C and D) Magnification of sleep deprivation to rebound transition. (C′ and D′) Quantification of sleep amount during ZT0 to ZT3 of rebound day. (E) Moving activity of flies (continuous lines) and number of rotations over the average 24-hour period (dashed lines). Moving activity combines both walking and micromoving. (F) Average number of tube rotations over the length of the sleep deprivation experiment (dashed lines) or seasonal trend (continuous lines; see Materials and Methods for details). n = 93 to 95 for all four groups. (G) Quantification of CaLexA-dependent green fluorescent protein (GFP) levels in the subregion of the ellipsoid bodies labeled by the expression of the R30G03 driver after 0.5, 5.5, and 9.5 days. Ns are indicated in the panel. a.u., arbitrary units.

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/5/2/eaau9253/DC1

    Fig. S1. Representative tracings of the behavioral activity over the course of 48 hours as recorded in real time by ethoscopes for all 881 female flies shown in Fig. 1A.

    Fig. S2. Sorted hierarchical cluster analysis based on pairwise distance, as supplement to Fig. 3.

    Fig. S3. Decrease in locomotion activity in sleep-deprived flies over time, a possible sign of physical fatigue.

    Fig. S4. Circadian rhythm, and not homeostatic drive, is the major contributor to sleep pressure during long-term sleep deprivation.

    Movie S1. Visual representation of the distribution of behavioral features across 24 hours in the dataset shown in Figs. 1 (A and B) and 2.

  • Supplementary Materials

    The PDF file includes:

    • Legend for fig. S1
    • Fig. S2. Sorted hierarchical cluster analysis based on pairwise distance, as supplement to Fig. 3.
    • Fig. S3. Decrease in locomotion activity in sleep-deprived flies over time, a possible sign of physical fatigue.
    • Fig. S4. Circadian rhythm, and not homeostatic drive, is the major contributor to sleep pressure during long-term sleep deprivation.
    • Legend for movie S1

    Download PDF

    Other Supplementary Material for this manuscript includes the following:

    • Fig. S1 (.pdf format). Representative tracings of the behavioral activity over the course of 48 hours as recorded in real time by ethoscopes for all 881 female flies shown in Fig. 1A.
    • Movie S1 (.mov format). Visual representation of the distribution of behavioral features across 24 hours in the dataset shown in Figs. 1 (A and B) and 2.

    Files in this Data Supplement:

Navigate This Article