Pan-neuronal knockdown of calcineurin reduces sleep in the fruit fly, Drosophila melanogaster

Abstract

Sleep is a unique physiological state, which is behaviorally defined, and is broadly conserved across species from mammals to invertebrates such as insects. Because of the experimental accessibility provided by various novel animal models including the fruit fly, Drosophila melanogaster, there have been significant advances in the understanding of sleep. Although the physiological functions of sleep have not been fully elucidated, accumulating evidence indicates that sleep is necessary to maintain the plasticity of neuronal circuits and, hence, is essential in learning and memory. Calcineurin (Cn) is a heterodimeric phosphatase composed of CnA and CnB subunits and known to function in memory consolidation in the mammalian brain, but its neurological functions in the fruit fly are largely unknown. Here, we show that Cn is an important regulator of sleep in Drosophila. A pan-neuronal RNA interference-mediated knockdown of Cn expression resulted in sleep loss, whereas misexpression of the constitutively active form of a CnA protein led to increased sleep. Furthermore, CnA knockdown also impaired the retention of aversive olfactory memory. These results indicate a role for Cn and calcium-dependent signal transduction in sleep and memory regulation and may bring insight into the relationship between them.

Figure 1.

Sleep loss due to CnA or CnB knockdown. A, B, Locomotor activity profiles of three representative control (left panel), CnA#1 designed against CanA-14F (middle), and CnB#2 designed against CanB2 (right) RNAi-expressing flies by elav-Gal4;UAS-Dicer-2 driver for 3 d in LD (A) and DD conditions (B). White, gray, and black bars under the horizontal axis indicate day, subjective day, and night, respectively. Total daily activity (C) and total sleep (D) for control, three CnA (#2, #3, each designed against Pp2B-14D and CanA1) RNAi, and two CnB (#1 designed against CanB) RNAi-expressing fly lines in DD conditions. Asterisks indicate statistically significant differences compared with control as given by Tukey–Kramer HSD test for normally distributed data (p < 0.05). n = 32, 30, 30, 8, 32, and 40 (all males) for control, CnA#1, CnA#2, CnA#3, CnB#1 and CnB#2 RNAi flies, respectively. Data are presented as mean ± SEM. E, Efficiency and specificity of Cn gene knockdown. The expression levels of CanA-14F, Pp2B-14D, CanA1, CanB, CanB2, and RpL32 genes in the head of male flies expressing various Cn RNAi transgenes in all neurons are expressed as the relative values to the control flies. As described in Materials and Methods, each mRNA level was quantified by qPCR and first normalized to GAPDH2. Then, the values were normalized to the average of independent control samples, which were set at 100%. Statistical significance between control and RNAi flies: *p < 0.05, Student's t test, n = 3 for each group, except for CnA#3 RNAi (n = 1). Data are presented as mean ± SEM.

Figure 2.

CnA and CnB knockdown flies exhibit a short sleep-bout length. Total sleep (A), sleep-bout length (B), and sleep-bout number (C) for control (white bars), CnA#1 RNAi (gray bars), and CnB#2 RNAi (black bars) during subjective day and subjective night in DD. The numbers of male flies tested are control = 57, CnA#1 RNAi = 57, CnB#2 RNAi = 27. Flies with no sleep during 12 h were excluded from the calculations of sleep-bout length in B. Asterisks indicate statistically significant differences from control determined by Tukey–Kramer HSD test for normally distributed data (p < 0.05). Data are presented as mean ± SEM.

Figure 3.

Misexpression of a constitutively active form of CanA-14F protein induces sleep. A, B, Sleep profiles in 30 min intervals for control (elav-Gal4;tub-Gal80^ts × w¹¹¹⁸, black circles, n = 17), male flies misexpressing CanA-14F-myc (elav-Gal4;tub-Gal80^ts × UAS-CanA-14F-myc, blue circles, n = 23) or CanA-14F^act-myc (elav-Gal4;tub-Gal80^ts × UAS-CanA-14F^act-myc, red circles, n = 24) at 30°C in 12 h light:12 h dark (LD) (A) and constant dark (DD) (B) conditions. Flies from all three groups were raised at 19°C until adulthood to suppress UAS transgene expression. Behavior was monitored over 2 d at 19°C (only the second day is shown) in LD and DD and shifted to 30°C for 2 d to allow UAS transgene expression, and then placed back to 19°C. C, Total daily sleep for controls (black), flies misexpressing CanA-14F-myc (blue), and CanA-14F^act-myc (red) at 19°C and 30°C in LD. Asterisks indicate statistically significant differences as determined by one-way ANOVA with Tukey–Kramer HSD post hoc test for normally distributed data (p < 0.05). Data are presented as mean ± SEM. D, Verification of UAS transgene expression in each misexpressed line at 30°C. Control male flies or those expressing CanA-14F-myc or CanA-14F^act-myc were collected at 12 h after the temperature shift from 19°C to 30°C at ZT 0. cDNA was prepared from total head RNA. The transgene fragments containing the Myc-encoding sequence were amplified by RT-PCR. GAPDH2 served as an internal control.

Figure 4.

Sleep is unaffected by misexpression of a constitutively active form of Pp2B–14D protein. Sleep profiles were recorded under the same condition described in Figure 3 for control flies (elav-Gal4;tub-Gal80^ts × w¹¹¹⁸, black circles, n = 17) or male flies misexpressing Pp2B–14D^act (elav-Gal4;tub-Gal80^ts × UAS-Pp2B–14D^act, orange circles, n = 14). Behavior was assayed for 2 d at 19°C (only the second day is shown) in DD and shifted to 30°C to allow UAS transgene expression for 2 d, and then placed back at 19°C. Data are presented as mean ± SEM.

Figure 5.

Hyporesponsiveness to mechanical stimuli of sleeping CnA#1 RNAi flies. Mechanical stimuli were applied to the male flies and response rate (the ratio of the number of aroused flies to the number of the sleeping flies expressed as a percentage) was calculated. Compared with control flies (elav-Gal4;UAS-Dicer-2 × w¹¹¹⁸, open bars), the CnA#1 RNAi flies (elav-Gal4;UAS-Dicer-2 × CnA#1, solid bars) showed a smaller response rate to both weak and strong stimuli (n = 19 or 20 for strongly stimulated CnA#1 RNAi flies and other stimulated groups, respectively). In contrast, there was no significant difference in the occurrence of spontaneous arousals (arousals without mechanical stimulations; see Materials and Methods for details) between control and CnA#1 RNAi flies (n = 40 for both genotypes) as calculated from behavioral data. Asterisks indicate statistically significant differences between control and CnA#1 RNAi flies (Student's t test; p < 0.05). Data are presented as mean ± SEM.

Figure 6.

Homeostatic response to sleep deprivation in CnA#1 RNAi flies. A, B, Sleep profiles plotted in 2 h bins over 2 d under DD conditions for control flies (elav-Gal4;UAS-Dicer-2 × w¹¹¹⁸) (A) or CnA#1 RNAi (elav-Gal4;UAS-Dicer-2 × CnA#1) male flies (B). Gray and black bars under the horizontal axis indicate subjective day and night, respectively. Twelve hours of sleep deprivation (SD) was started at CT 12. Only flies that were active >80% of the time during SD were analyzed. The values for baseline (open circles) and sleep-deprived siblings (closed circles) are the group average for 2 h (n = 11–16). C, The amount of sleep during the first 4 h of baseline (open bars) and recovery (solid bars). Significant sleep rebound was observed in both strains (Student's t test; *p < 0.01). Data are presented as mean ± SEM.

Figure 7.

Homeostatic response to sleep deprivation in CnB#2 RNAi flies. A, B, Sleep profiles plotted in 2 h bins over 2 d under DD conditions for control flies (elav-Gal4;UAS-Dicer-2 × w¹¹¹⁸) (A) or CnB#2 RNAi (elav-Gal4;UAS-Dicer-2 × CnB#2) male flies (B). The conditions are same as Figure 6 (n = 11–12). C, The amount of sleep during the first 4 h of baseline (open bars) and recovery (solid bars). Significant sleep rebound was observed in both strains (Student's t test; *p < 0.01). Data are presented as mean ± SEM.

Figure 8.

CnA#1 RNAi flies show defective memory retention. A, The 3 min memory (immediate memory) and 2 h memory of control (elav-Gal4;UAS-Dicer-2 × w¹¹¹⁸, white bars) and CnA#1 RNAi (elav-Gal4;UAS-Dicer-2 × CnA#1, gray bars) male flies were determined in an aversive olfactory conditioning test after training. B, Avoidance responses of control (white bars), CnA#1 RNAi (gray bars), and CnB#2 RNAi (elav-Gal4;UAS-Dicer-2 × CnB#2, black bars) male flies to 3-octanol (3-OCT), 4-methylcyclohexanol (4-MCH), and electric shock (Shock). Asterisks indicate statistically significant differences from control determined by Tukey–Kramer HSD test for normally distributed data (p < 0.05). Data are presented as mean ± SEM (n = 8–10 for each group).

Figure 9.

CanA-14F, CanB, and CanB2 genes are expressed ubiquitously in the adult brain. In situ hybridization and immunohistochemistry were performed as described in Materials and Methods. A, C, E, Representative frontal sections of the male fly head showing the brain, optic lobes, and retina labeled with DIG-labeled antisense RNA probes for CanA-14F (A), CanB (C), and CanB2 (E). B, D, F, Sense RNA probes were used as control. All tissue sections were counterstained with Nuclear Fast Red following in situ hybridization. The box areas are shown in higher magnification in A′–F′. Whole-mount adult male brains of control (elav-Gal4;UAS-Dicer-2 × w¹¹¹⁸) (G) and CnB#2 RNAi (elav-Gal4;UAS-Dicer-2 × CnB#2) (H) were stained with an anti-Calcineurin B antibody. Representative maximum projection images of 1.5 μm confocal sections of the entire brain are shown. Scale bars, 100 μm.