Measuring metabolic rate in single flies during sleep and waking states via indirect calorimetry

Abstract

Background: Drosophila melanogaster is a leading genetic model for studying the neural regulation of sleep. Sleep is associated with changes in behavior and physiological state that are largely conserved across species. The investigation of sleep in flies has predominantly focused on behavioral readouts of sleep because physiological measurements, including changes in brain activity and metabolic rate, are less accessible. We have previously used stop-flow indirect calorimetry to measure whole body metabolic rate in single flies and have shown that in flies, like mammals, metabolic rate is reduced during sleep.

New method: Here, we describe a modified version of this system that allows for efficient and highly sensitive acquisition of CO₂ output from single flies.

Results: In this modified system, we show that sleep-dependent changes in metabolic rate are diminished in aging flies, supporting the notion that sleep quality is reduced as flies age. We also describe a modification that allows for simultaneous acquisition of CO₂ and O₂ levels, providing a respiratory quotient that quantifies how metabolic stores are utilized. We find that the respiratory quotient identified in flies on an all-sugar diet is suggestive of lipogenesis, where the dietary sugar provided to the flies is being converted to fat.

Comparison with existing methods and conclusions: Taken together, the measurement of metabolic rate via indirect calorimetry not only provides a physiological readout of sleep depth, but also provides insight the metabolic regulation of nutrient utilization, with broad applications to genetic studies in flies.

Keywords: Aging; CO(2) output; Drosophila; Indirect calorimetry; Metabolism.

Figure 1 –. Single fly system overview

The Sleep, Activity, and Metabolic Monitor (SAMM) system can be used to measure sleep and CO₂ output in single flies. Individual flies were fed 5% sucrose, during which sleep and CO₂ output was measured over the course of 24 hours. Zero-grade air is first passed through a scrubbing column (1). The resulting dehumidified, CO₂-free air is then re-humidified and pumped to two mass-flow control valves to maintain constant air flow into the system (2). One mass flow control valve regulates clean, reference air flow directly to the CO₂ analyzer. The other mass flow control valve regulates air flow to the multiplexer, which directs the sampling of CO₂ that accumulates in each behavioral chamber over a specified time-period (4). The behavioral chambers are housed inside a Drosophila Activity Monitor (DAM), allowing for the simultaneous measurement of CO₂ output and activity/sleep (3). The accumulated air from each behavior chamber is then sent to the CO₂ analyzer, where the reference air is subtracted from the air collected from each behavioral chamber to determine CO₂ output for each individual fly (5). Both the reference and animal air were then passed individually through scrubbing columns before entering the O₂ analyzer, where the reference air is again subtracted from the air collected from each behavioral chamber to determine O₂ consumed for each group of flies. A detailed description of each component can be found in the ‘Measurement of CO₂ output’ section of the Methods.

Figure 2 –. Single fly system data males/females

Nighttime sleep and CO₂ output are reduced in male flies. (A) Schematic of experimental design. Sleep and CO₂ output were assessed in male and female flies for 24 hrs on 5% sucrose. ZT=0 indicates time of lights on, while ZT=12 indicates time of lights off. (B) Sleep profile of female and male flies. (C) There is a significant effect of sex (two-way ANOVA: F_1,156 = 21.32, P<0.0001, N = 40 per sex) and time of day (two-way ANOVA: F_1,156 = 39.00, P<0.0001, N = 40 per sex) on sleep duration. Females sleep significantly more at night (P<0.0001), while no significant differences in sleep duration between day and night was observed in males (P=0.8362). Females also sleep significantly more than males during the night (P<0.0001), but no significant differences in sleep duration was observed between the sexes during the day (P=0.1798). (D) Representative trace from an individual female and male, indicating the unadjusted amount of CO₂ produced over a 5 min time period (ppm: parts per million). (E) Profile of CO₂ output of female and male flies. (F) There is a significant effect of sex (two-way ANOVA: F_1,156 = 38.55, P<0.0001, N = 40 per sex) and time of day (two-way ANOVA: F_1,156 = 12.47, P=0.0005, N = 40 per sex) on CO₂ output. CO₂ output is significantly higher in females during the day, relative to the night (P=0.0003), while no significant differences in CO₂ output between day and night was observed in males (P=0.4760).CO₂ output is also significantly higher in females both during the day (P<0.0001) and night (P=0.0065), compared to males. (G) Linear regression of daytime CO₂ output as a function of time asleep for both males and females. The slope of each regression is significantly different from zero (females: F_1,399 = 74.92, P<0.0001; males: F_1,390 = 137.3, P<0.0001) as well as significantly different from each other (ANCOVA with time asleep as the covariate: F_1,789 = 4.682, P=0.0308). (H) Linear regression of nighttime CO₂ output as a function of time asleep for both males and females. The slope of each regression is significantly different from zero (females: F_1,432 = 43.32, P<0.0001; males: F_1,376 = 80.53, P<0.0001) as well as significantly different from each other (ANCOVA with time asleep as the covariate: F_1,808 = 7.925, P=0.0050). For profiles and linear regressions, white background indicates daytime, while gray background indicates nighttime. ZT indicates zeitgeber time. Error bars represent +/− standard error from the mean. Grey dots represent measurements of individual flies. ** = P<0.01; **** = P<0.0001.

Figure 3. Single fly system details with sleep/metabolism breakdown with aging flies

Aging reduces sleep duration and metabolic rate during waking in female flies. (A) Schematic of experimental design. Sleep and CO₂ output were assessed in young, 10 day-old flies and old 40-day old flies for 24 hrs on 5% sucrose. ZT=0 indicates time of lights on, while ZT=12 indicates time of lights off. (B) Sleep profile of 10 day-old flies and 40 day-old flies. (C) There is a significant effect of age (two-way ANOVA: F_1,152 = 213.4, P<0.0001, N = 37–41) and time of day (two-way ANOVA: F₁₁₅₂₀ = 235.6, P<0.0001, N = 37–41) on sleep duration. For both ages, sleep duration is significantly higher during the night, relative to the day (10d: P<0.0001; 40d: P<0.0001). Sleep duration is significantly reduced in 40 day-old flies both during the day (P<0.0001) and night (P<0.0001), compared to 10 day-old flies. (D) Bout length is significantly reduced in 40 day-old flies, compared to 10 day-old flies (t-test: t₇₃=9.318, P<0.0001). (E) Bout number is significantly increased in 40 day-old flies, compared to 10 day-old flies (t-test: t₇₃=04.659, P<0.0001). (F) Waking activity is significantly increased in 40 day-old flies, compared to 10 day-old flies (t-test: t₇₃=2.558, P=0.0126). (G) Representative CO₂ trace from an individual 10 day-old and 40 day-old fly. (H) Profile of CO₂ output of young, 10 day-old flies and old, 40 day-old flies. (I) There is no effect of age on CO₂ output (two-way ANOVA: F_1,152 = 3.155, P=0.0777, N = 37–41), but there is an effect of time of day (two-way ANOVA: F_1,152 = 108.3, P<0.0001, N = 37–41) on CO₂ output. At both ages, CO₂ output is significantly higher during the day, relative to the night (10d: P<0.0001; 40d: P<0.0001). (J) When awake, there is a significant effect of age (two-way ANOVA: F_1,154 = 15.90, P<0.0001, N = 37–41) and time of day (two-way ANOVA: F_1,154 = 17.03, P<0.0001, N = 37–41) on CO₂ output. For both ages, CO₂ output is significantly higher during the day, relative to the night (10d: P<0.0457; 40d: P<0.0007). CO2 output is also significantly reduced in 40 day-old flies both during the day (P<0.0487) and night (P<0.0015), compared to 10 day-old flies. (K) When sleeping, there is no effect of age (two-way ANOVA: F_1,154 = 1.794, P<0.1827, N = 37–41), but there is a significant effect of time of day (two-way ANOVA: F_1,154 = 69.18, P<0.0001, N = 37–41) on CO₂ output. For both ages, CO₂ output is significantly higher during the day, relative to the night (10d: P<0.0001; 40d: P<0.0001). (L) Linear regression of daytime CO₂ output as a function of time asleep for 10 day-old and 40 day-old flies. The slope of 10 day-old flies is significantly different from zero (F_1,420 = 104.6, P<0.0001), while the slope of 40 day-old flies is not (F_1,304 = 3.191, P<0.0751). They are, however, significantly different from each other (ANCOVA: F_1,724 = 18.55, P<0.0001). (M) Linear regression of nighttime CO₂ output as a function of time asleep for 10 day-old and 40 day-old flies. The slope of each regression is significantly different from zero (10d: F_1,427 = 47.37, P<0.0001; 40d: F_1,475 = 17.85, P<0.0001) as well as significantly different from each other (ANCOVA: F_1,902 = 5.841, P=0.0159). For profiles and liner regressions, white background indicates daytime, while gray background indicates nighttime. ZT indicates zeitgeber time. Error bars represent +/− standard error from the mean. Grey dots represent measurements of individual flies. * = P<0.05; ** = P<0.01; *** = P<0.001; **** = P<0.0001.

Figure 4 –. O2 measurements and group data with males/females.

The Sleep, Activity, and Metabolic Monitor (SAMM) system can be used to measure CO₂ output and O₂ consumption in groups of flies. Groups of 25 flies were fed 5% sucrose, during which both CO₂ and O₂ were measured, as described in Figure 1, over the course of 24 hours. (A) Profile of CO₂ output in female and male flies. (B) There is a significant effect of sex (two-way ANOVA: F_1,80 = 85.25, P<0.0001, N = 21 per sex) and time of day (two-way ANOVA: F_1,80 = 29.24, P<0.0001, N = 21 per sex) on CO₂ output. For both sexes, CO₂ output is significantly higher during the day, relative to the night (females: P<0.0001; males: P=0.0301). CO₂ output is also significantly higher in males both during the day (P<0.0001) and night (P<0.0001), compared to females. (C) Profile of O₂ consumption in female and male flies. (D) There is a significant effect of sex (two-way ANOVA: F_1,80 = 95.36, P<0.0001, N = 21 per sex) and time of day (two-way ANOVA: F_1,80 = 18.72, P<0.0001, N = 21 per sex) on O₂ consumption. For both sexes, O₂ consumption is significantly higher during the day, relative to the night (females: P<0.0001; males: P=0.0125). O₂ consumption is also significantly higher in females both during the day (P<0.0001) and night (P<0.0001), compared to males. (E) Profile of the respiratory quotient (CO₂ output / O₂ consumption) in female and male flies. (F) There is no significant difference in the respiratory quotient between male and females (two-way ANOVA: F_1,80 = 0.1353, P=0.7140, N = 21 per sex). For profiles, white background indicates daytime, while gray background indicates nighttime. ZT indicates zeitgeber time. Error bars represent +/− standard error from the mean. Grey dots represent measurements of individual flies. **** = P<0.0001.