Aging induced endoplasmic reticulum stress alters sleep and sleep homeostasis

Abstract

Alterations in the quality, quantity, and architecture of baseline and recovery sleep have been shown to occur during aging. Sleep deprivation induces endoplasmic reticular (ER) stress and upregulates a protective signaling pathway termed the unfolded protein response. The effectiveness of the adaptive unfolded protein response is diminished by age. Previously, we showed that endogenous chaperone levels altered recovery sleep in Drosophila melanogaster. We now report that acute administration of the chemical chaperone sodium 4-phenylbutyrate (PBA) reduces ER stress and ameliorates age-associated sleep changes in Drosophila. PBA consolidates both baseline and recovery sleep in aging flies. The behavioral modifications of PBA are linked to its suppression of ER stress. PBA decreased splicing of X-box binding protein 1 and upregulation of phosphorylated elongation initiation factor 2 α, in flies that were subjected to sleep deprivation. We also demonstrate that directly activating ER stress in young flies fragments baseline sleep and alters recovery sleep. Alleviating prolonged or sustained ER stress during aging contributes to sleep consolidation and improves recovery sleep or sleep debt discharge.

Keywords: 4-phenylbutyrate; Aging; Chaperone; Sleep; Sleep loss/ deprivation; Unfolded protein response.

Fig. 1

Total sleep time decreases with age. (A) Total sleep in minutes in young (n=143) and aged (n=106) wCS10 flies. Total sleep significantly decreases in the aged flies (p<0.001) in comparison to the young flies. Treatment with 5mM of PBA has little effect on young flies, but it significantly increases total daytime sleep in the aged flies (p<0.01). (B) Total sleep in minutes in young (n=54) and aged (n=40) w1118^ex flies. PBA significantly increased total sleep in aged flies during both the day and night (p<0.05). (C, D) The number of sleep bouts significantly decreases during the daytime (p<0.01) in the aged flies in comparison to the young flies. Treatment with PBA significantly increased the number of sleep bouts (p<0.01) in the aged flies during the day and the young flies during the night. PBA significantly increased the sleep bout numbers in the young (night) and aged flies (day and night, p<0.05). (E, F) Sleep bout duration in the aged wCS10 during the day and in the aged w1118^ex in both the day and night (p<0.05) was increased with PBA. (G-J) Mean wake bout duration is increased with PBA during the daytime in the young wCS10 (p<0.001) and the w1118^ex flies (p<0.05). In the aged wCS10 and w1118^ex flies, treatment with PBA significantly decreased wake bout duration during the daytime in comparison to their age-matched controls (wCS10, p<0.001; w1118^ex p<-0.05). In the aged w1118^ex flies, wake bout duration was also significantly decreased during the night (p<0.05). Mean + S.E.M. shown *P<0.05, **P<0.01, ***P<0.001

Fig. 1

Total sleep time decreases with age. (A) Total sleep in minutes in young (n=143) and aged (n=106) wCS10 flies. Total sleep significantly decreases in the aged flies (p<0.001) in comparison to the young flies. Treatment with 5mM of PBA has little effect on young flies, but it significantly increases total daytime sleep in the aged flies (p<0.01). (B) Total sleep in minutes in young (n=54) and aged (n=40) w1118^ex flies. PBA significantly increased total sleep in aged flies during both the day and night (p<0.05). (C, D) The number of sleep bouts significantly decreases during the daytime (p<0.01) in the aged flies in comparison to the young flies. Treatment with PBA significantly increased the number of sleep bouts (p<0.01) in the aged flies during the day and the young flies during the night. PBA significantly increased the sleep bout numbers in the young (night) and aged flies (day and night, p<0.05). (E, F) Sleep bout duration in the aged wCS10 during the day and in the aged w1118^ex in both the day and night (p<0.05) was increased with PBA. (G-J) Mean wake bout duration is increased with PBA during the daytime in the young wCS10 (p<0.001) and the w1118^ex flies (p<0.05). In the aged wCS10 and w1118^ex flies, treatment with PBA significantly decreased wake bout duration during the daytime in comparison to their age-matched controls (wCS10, p<0.001; w1118^ex p<-0.05). In the aged w1118^ex flies, wake bout duration was also significantly decreased during the night (p<0.05). Mean + S.E.M. shown *P<0.05, **P<0.01, ***P<0.001

Fig. 1

Total sleep time decreases with age. (A) Total sleep in minutes in young (n=143) and aged (n=106) wCS10 flies. Total sleep significantly decreases in the aged flies (p<0.001) in comparison to the young flies. Treatment with 5mM of PBA has little effect on young flies, but it significantly increases total daytime sleep in the aged flies (p<0.01). (B) Total sleep in minutes in young (n=54) and aged (n=40) w1118^ex flies. PBA significantly increased total sleep in aged flies during both the day and night (p<0.05). (C, D) The number of sleep bouts significantly decreases during the daytime (p<0.01) in the aged flies in comparison to the young flies. Treatment with PBA significantly increased the number of sleep bouts (p<0.01) in the aged flies during the day and the young flies during the night. PBA significantly increased the sleep bout numbers in the young (night) and aged flies (day and night, p<0.05). (E, F) Sleep bout duration in the aged wCS10 during the day and in the aged w1118^ex in both the day and night (p<0.05) was increased with PBA. (G-J) Mean wake bout duration is increased with PBA during the daytime in the young wCS10 (p<0.001) and the w1118^ex flies (p<0.05). In the aged wCS10 and w1118^ex flies, treatment with PBA significantly decreased wake bout duration during the daytime in comparison to their age-matched controls (wCS10, p<0.001; w1118^ex p<-0.05). In the aged w1118^ex flies, wake bout duration was also significantly decreased during the night (p<0.05). Mean + S.E.M. shown *P<0.05, **P<0.01, ***P<0.001

Fig. 2

Sleep homeostasis is impaired during aging. Aging alters the homeostatic response after 6 h of sleep deprivation. Flies were sleep deprived (SD) for 6 h from (Zeitgeber times, ZT) ZT18 to ZT24 and allowed to recover sleep for 24 h after deprivation. Young flies efficiently recover lost sleep whereas the aged flies do not. Recovery sleep per hour 24 hours post sleep deprivation in (A) young (n=47) and aged untreated control (n=43) flies. Data is shown as the difference in sleep per hour post deprivation to that over same time interval on preceding day for each fly averaged across all flies with standard error of the mean. PBA has minimal effects in the (B) young (n=47) flies, however, it increased the amount and efficiency of recovery sleep in the (C) aged wCS10 flies (n=37). Data is displayed as the average amount of total minutes of sleep after baseline subtraction and S.E.M.

Fig. 2

Sleep homeostasis is impaired during aging. Aging alters the homeostatic response after 6 h of sleep deprivation. Flies were sleep deprived (SD) for 6 h from (Zeitgeber times, ZT) ZT18 to ZT24 and allowed to recover sleep for 24 h after deprivation. Young flies efficiently recover lost sleep whereas the aged flies do not. Recovery sleep per hour 24 hours post sleep deprivation in (A) young (n=47) and aged untreated control (n=43) flies. Data is shown as the difference in sleep per hour post deprivation to that over same time interval on preceding day for each fly averaged across all flies with standard error of the mean. PBA has minimal effects in the (B) young (n=47) flies, however, it increased the amount and efficiency of recovery sleep in the (C) aged wCS10 flies (n=37). Data is displayed as the average amount of total minutes of sleep after baseline subtraction and S.E.M.

Fig. 3

Tunicamycin fragments baseline sleep. (A) Total sleep at baseline during both the day and the night (12µM [p<0.05]) is decreased. (B) Sleep bout numbers are increased (p<0.05) and (C) sleep bout duration is decreased (p<0.05) during the night in young wCS10 flies treated with tunicamycin (n=36). (D) Recovery sleep is altered in young flies treated with 12µM tunicamycin after 6h of sleep deprivation in comparison to the young untreated control flies (n=20) subjected to SD. Mean +S.E.M. shown, *P<0.05, **P<0.01

Fig. 4

PBA consolidates sleep in the short sleeping Sleepless (sss^p1) mutant. (A) Total sleep in sss^p1 (n=20) and its background control strain iso³¹ (n=32). PBA increased total sleep (p<0.01) during the day in the sss^p1 mutant flies (n=28). (B) PBA also significantly increased both sleep bout number (p<0.01) and (C) sleep bout duration during the day (p<0.001) in the sss^p1 mutants. Total sleep (NS) and sleep bout duration, (p<0.05) are decreased during the night in the control strain (iso³¹), treated with PBA (n=32). Mean + S.E.M. shown,*P<0.05, **P<0.01, ***P<0.001

Fig. 5

Upregulation of ER stress markers with sleep deprivation (SD). (A) Densitometric quantification of BiP expression in young undisturbed (YUD), young sleep deprived (YSD), young PBA (YPBA) and young PBA sleep deprived (YPBASD) wCS10 fly heads (n=4 trials, 10 heads/pool). 6 h of SD increased BiP levels in young flies in comparison to their non-SD counterparts (p=0.007). Young flies treated with PBA responded similarly to young untreated flies when sleep deprived (p<0.05). (B) Representative western blots. 15µg of protein were loaded per well from undisturbed and sleep deprived flies. (C) Quantification of BiP expression in aged undisturbed (AUD); aged sleep deprived (ASD), aged PBA (APBA) and aged PBA sleep deprived (APBASD) fly heads. (D) Representative western blot shows 15µg of protein loaded per well from undisturbed and SD flies. BiP expression increased in aged sleep deprived flies (ASD) versus aged undisturbed flies (AUD) (p<0.05) (n=4 trials, 10 heads/pool). Aged flies treated with PBA showed no significant induction of BiP after sleep deprivation. All gels were stripped and reprobed with β-actin as the loading control. (E-F) ER stress triggers Drosophila XBP1 mRNA splicing, which is characterized by the elimination of a 23 base pair intron from Drosophila XBP1 mRNA. PBA reduced both the unspliced and spliced variant of XBP1 in the YPBASD and APBA flies. In the APBASD flies XBP1 splicing is decreased. Mean +S.E.M. shown for XBP1 *P<0.05, **P<0.01, #P<0.001 (G) Densitometric quantification of phosphoeIF2α levels in fly heads (n=4 trials, 10 heads/pool). 6 h of sleep deprivation (SD) increased peIF2α expression in YSD flies in comparison to their undisturbed counterparts (YUD) (p=0.032). Aged flies (AUD) have increased basal levels of p-eIF2α that are significantly higher than the YUD flies (p=0.006). There was no significant difference between the AUD and aged ASD flies; however, p-eIF2α levels are markedly reduced in aged SD flies with acute application of PBA (ASDPBA) (p<0.05). Mean +S.E.M. shown, *P<0.05, **P<0.01. (H) Representative Western blots of p-eIF2α shows 15µg of protein loaded per well from undisturbed and SD flies. All gels were stripped and reprobed with β-actin as the loading control.

Fig. 5

Upregulation of ER stress markers with sleep deprivation (SD). (A) Densitometric quantification of BiP expression in young undisturbed (YUD), young sleep deprived (YSD), young PBA (YPBA) and young PBA sleep deprived (YPBASD) wCS10 fly heads (n=4 trials, 10 heads/pool). 6 h of SD increased BiP levels in young flies in comparison to their non-SD counterparts (p=0.007). Young flies treated with PBA responded similarly to young untreated flies when sleep deprived (p<0.05). (B) Representative western blots. 15µg of protein were loaded per well from undisturbed and sleep deprived flies. (C) Quantification of BiP expression in aged undisturbed (AUD); aged sleep deprived (ASD), aged PBA (APBA) and aged PBA sleep deprived (APBASD) fly heads. (D) Representative western blot shows 15µg of protein loaded per well from undisturbed and SD flies. BiP expression increased in aged sleep deprived flies (ASD) versus aged undisturbed flies (AUD) (p<0.05) (n=4 trials, 10 heads/pool). Aged flies treated with PBA showed no significant induction of BiP after sleep deprivation. All gels were stripped and reprobed with β-actin as the loading control. (E-F) ER stress triggers Drosophila XBP1 mRNA splicing, which is characterized by the elimination of a 23 base pair intron from Drosophila XBP1 mRNA. PBA reduced both the unspliced and spliced variant of XBP1 in the YPBASD and APBA flies. In the APBASD flies XBP1 splicing is decreased. Mean +S.E.M. shown for XBP1 *P<0.05, **P<0.01, #P<0.001 (G) Densitometric quantification of phosphoeIF2α levels in fly heads (n=4 trials, 10 heads/pool). 6 h of sleep deprivation (SD) increased peIF2α expression in YSD flies in comparison to their undisturbed counterparts (YUD) (p=0.032). Aged flies (AUD) have increased basal levels of p-eIF2α that are significantly higher than the YUD flies (p=0.006). There was no significant difference between the AUD and aged ASD flies; however, p-eIF2α levels are markedly reduced in aged SD flies with acute application of PBA (ASDPBA) (p<0.05). Mean +S.E.M. shown, *P<0.05, **P<0.01. (H) Representative Western blots of p-eIF2α shows 15µg of protein loaded per well from undisturbed and SD flies. All gels were stripped and reprobed with β-actin as the loading control.

Fig. 5

Upregulation of ER stress markers with sleep deprivation (SD). (A) Densitometric quantification of BiP expression in young undisturbed (YUD), young sleep deprived (YSD), young PBA (YPBA) and young PBA sleep deprived (YPBASD) wCS10 fly heads (n=4 trials, 10 heads/pool). 6 h of SD increased BiP levels in young flies in comparison to their non-SD counterparts (p=0.007). Young flies treated with PBA responded similarly to young untreated flies when sleep deprived (p<0.05). (B) Representative western blots. 15µg of protein were loaded per well from undisturbed and sleep deprived flies. (C) Quantification of BiP expression in aged undisturbed (AUD); aged sleep deprived (ASD), aged PBA (APBA) and aged PBA sleep deprived (APBASD) fly heads. (D) Representative western blot shows 15µg of protein loaded per well from undisturbed and SD flies. BiP expression increased in aged sleep deprived flies (ASD) versus aged undisturbed flies (AUD) (p<0.05) (n=4 trials, 10 heads/pool). Aged flies treated with PBA showed no significant induction of BiP after sleep deprivation. All gels were stripped and reprobed with β-actin as the loading control. (E-F) ER stress triggers Drosophila XBP1 mRNA splicing, which is characterized by the elimination of a 23 base pair intron from Drosophila XBP1 mRNA. PBA reduced both the unspliced and spliced variant of XBP1 in the YPBASD and APBA flies. In the APBASD flies XBP1 splicing is decreased. Mean +S.E.M. shown for XBP1 *P<0.05, **P<0.01, #P<0.001 (G) Densitometric quantification of phosphoeIF2α levels in fly heads (n=4 trials, 10 heads/pool). 6 h of sleep deprivation (SD) increased peIF2α expression in YSD flies in comparison to their undisturbed counterparts (YUD) (p=0.032). Aged flies (AUD) have increased basal levels of p-eIF2α that are significantly higher than the YUD flies (p=0.006). There was no significant difference between the AUD and aged ASD flies; however, p-eIF2α levels are markedly reduced in aged SD flies with acute application of PBA (ASDPBA) (p<0.05). Mean +S.E.M. shown, *P<0.05, **P<0.01. (H) Representative Western blots of p-eIF2α shows 15µg of protein loaded per well from undisturbed and SD flies. All gels were stripped and reprobed with β-actin as the loading control.

Fig. 6

Model demonstrating the relationship between aging, UPR activation and sleep. Sleep fragmentation resulting from age-related impairments in the UPR can be ameliorated by administration of a chemical chaperone. Treatment with the chemical chaperone results in sleep consolidation and improved sleep debt discharge, recapitulating a young healthy phenotype. Tunicamycin treatment induces ER stress, which leads to sleep fragmentation and confers an aged phenotype.