A PERIOD3 variant causes a circadian phenotype and is associated with a seasonal mood trait

Abstract

In humans, the connection between sleep and mood has long been recognized, although direct molecular evidence is lacking. We identified two rare variants in the circadian clock gene PERIOD3 (PER3-P415A/H417R) in humans with familial advanced sleep phase accompanied by higher Beck Depression Inventory and seasonality scores. hPER3-P415A/H417R transgenic mice showed an altered circadian period under constant light and exhibited phase shifts of the sleep-wake cycle in a short light period (photoperiod) paradigm. Molecular characterization revealed that the rare variants destabilized PER3 and failed to stabilize PERIOD1/2 proteins, which play critical roles in circadian timing. Although hPER3-P415A/H417R-Tg mice showed a mild depression-like phenotype, Per3 knockout mice demonstrated consistent depression-like behavior, particularly when studied under a short photoperiod, supporting a possible role for PER3 in mood regulation. These findings suggest that PER3 may be a nexus for sleep and mood regulation while fine-tuning these processes to adapt to seasonal changes.

Keywords: PER3; circadian clock; circadian rhythms; familial advanced sleep phase; seasonal affective disorder.

Fig. 1.

FASP pedigree and the amino acid alignment around the rare variants. (A) FASP kindred 7107. Circles represent women; squares, men. Filled symbols denote affected individuals; open symbols, unaffected individuals. Symbols marked with a cross are unknown (i.e., do not meet criteria for affected or unaffected). Diagonal lines through symbols indicate that the individual is deceased. Subject 42637 is the proband. (B) Alignments for human (h) and mouse (m) PER proteins in the region harboring the variants. The P415A/H417R variants are highlighted.

Fig. 2.

Expression of PER3-P415A/H417R in mice lengthens the period and delays the phase of activity rhythm. (A) Wheel-running activity profiles of hPER3-WT-Tg and hPER3-P415A/H417R-Tg mice during LD 12:12 (n = 7–13). (B) The period of wheel-running activity under LL and DD conditions, respectively (n = 7–29). *P < 0.05, Student’s t test. (C) Double-plotted actograms of wheel-running activity (black bars) for representative hPER3-WT-Tg and hPER3-P415A/H417R-Tg mice. Gray areas indicate periods of darkness before release into LL. (D) Double-plotted actograms of wheel-running activity for representative WT, Per3^−/−, hPER3-WT-Tg, and hPER3-P415A/H417R-Tg mice during LD 4:20. (E) Wheel-running activity profiles of WT, hPER3-P415A/H417R-Tg, and Per3^−/− mice (Upper) and WT and hPER3-WT-Tg mice (Lower) during LD 4:20 (n = 6–20). (F) Activity onset and offset time under LD 4:20 (n = 6–20). *P < 0.05, ***P < 0.005, Student’s t test. The white bars indicate light periods; black bars, dark periods. Error bars represent SEM.

Fig. S1.

Expression of PER3-P415A/H417R in flies advances behavioral phase compared with PER3-WT–expressing flies. (A) Normalized locomotor activity profiles during LD for 1 d, followed by 4 d of DD. White bars indicate light periods; black bars, dark periods; gray bars, subjective light periods (n = 28–40). (B) Relative mRNA abundance for hPER3 from whole head extracts of timGAL4/UAShPER3-WT (2), timGAL4/+;UAShPER3-WT/+ (3), timGAL4/+;UAShPER3-P415A/H417R/+ (1), and UASPER3-P415A/H417R;timGAL4/+ (7) flies was determined by qRT-PCR (n = 2–6). For each experiment, the value of timGAL4/UAShPER3-WT (2) was set to 1. Error bars represent SEM.

Fig. 3.

P415A/H417R variants destabilize PER3. (A) Repressor activity of WT or P415A/H417R PER3 protein on CLOCK-BMAL1–mediated expression of Per2 promoter-driven luciferase was examined by a reporter assay performed in HEK293T cells (n = 3). Luciferase activity was normalized to activity levels in the absence of CLOCK and BMAL1 (white column). Addition of CLOCK and BMAL1 resulted in a nearly 10-fold increase in luciferase activity (black column). PER3-P415A/H417R had reduced repressor activity across different titers compared with PER3-WT. PER2 (red column) and CRY2 (yellow column) were positive controls for repressor activity. ***P < 0.001, two-way ANOVA. (B) PER3 protein levels in the nuclear and the cytoplasmic fractions. HEK293T cells were transfected with indicated plasmid vectors. At 48 h after transfection, the cells were harvested and then fractionated into nuclear and cytoplasmic fractions. Results are expressed as mean ± SEM (n = 3). **P < 0.01, ***P < 0.005, Student′s t test. (C) HEK293T cells were transfected with indicated plasmid vectors. At 24 h after transfection, the culture medium was replaced by measuring medium containing CHX (100 µg/mL). Bioluminescence was measured at 10-min intervals. Bioluminescence at time point 0 was set to 1. Half-lives of PER3-LUCs were calculated by fitting the bioluminescence signals to exponential functions. Error bars show mean ± SEM (n = 3). **P < 0.01, Student’s t test. (D) HEK293T cells were transfected with indicated plasmid vectors. At 40 h after transfection, the cells were treated with CHX (100 µg/mL) for 4 or 8 h. PER3 protein levels at time point 0 were set to 1, and relative PER3 protein levels were plotted. Error bars show mean ± SEM (n = 3). ***P < 0.005, Student’s t test. (E) Western blots of nuclear and cytoplasmic fractions of liver extracts from hPER3-WT-Tg and hPER3-P415A/H417R-Tg mice at ZT12. LMNB1 and β-actin served as loading controls in the nuclear and cytoplasmic fractions, respectively. Nuclear PER3 protein levels were normalized to LMNB1 levels and are expressed as mean ± SEM (n = 3 per genotype). *P < 0.05, Student′s t test. (F) Expression levels of hPER3 mRNA in liver extracts from hPER3-WT-Tg and hPER3-P415A/H417R-Tg animals (n = 5 per genotype), as assayed by three independent TaqMan qPCR assays across different exon boundaries. Expression levels in hPER3-P415A/H417R-Tg mice were normalized to those in hPER3-WT-Tg mice. Results are expressed as mean ± SEM. **P < 0.01, ***P < 0.001, Student’s t test.

Fig. S2.

hPER3-P415A/H417R is lower than hPER3-WT in transgenic flies. (A) Representative Western blots of whole head extracts of timGAL4/UAShPER3-WT (2) and UASPER3-P415A/H417R;timGAL4/+ (7) flies during DD day 2. (B and C) Quantification of Western blots in A (n = 18). For C, hPER3-Myc protein levels were normalized to the mRNA levels of hPER3 shown in Fig. S1B. Significant differences were calculated using Student’s t test. *P < 0.05; ***P < 0.001.

Fig. 4.

PER3-P415A/H417R has reduced ability to stabilize PER1/PER2. (A and B) (Left) representative Western blots of PER2 in nuclear and cytoplasmic fractions of hypothalamus (A) and liver (B) of hPER3-WT-Tg and hPER3-P415A/H417R-Tg mice at Zeitgeber Time 6 (ZT6). β-actin served as a loading control, and LMNB1 and caspase3 served as nuclear and cytoplasmic markers, respectively. (Right) Quantification of PER2 protein levels in nuclear and cytoplasmic fractions of hypothalamus (A) and liver (B) from hPER3-WT-Tg and hPER3-P415A/H417R-Tg mice (n = 10–11). Expression levels were normalized to average levels in hPER3-WT-Tg mice. Results are expressed as mean ± SEM. *P < 0.05, ANOVA with Tukey's post hoc multiple-comparison test for A and Student′s t test for B. (C) HEK293T cells were transfected with the indicated plasmid vectors. At 24 h after transfection, the culture medium was replaced with measuring medium containing CHX (100 µg/mL). Bioluminescence was measured at 10-min intervals. Bioluminescence at the time point 0 was set to 1. Half-lives of PER1-LUC and PER2-LUC were calculated by fitting the bioluminescence signals to exponential functions. Error bars show mean ± SEM (n = 4). *P < 0.05, **P < 0.01, Student’s t test. (D) HEK293T cells were transfected with myc-PER2 and FLAG-PER3 expression vectors. At 40 h after transfection, the cells were treated with 100 µg/mL CHX for 4 or 8 h. PER2 protein levels at time point 0 were set to 1, and relative PER2 protein levels were plotted. Error bars show mean ± SEM (n = 3). *P < 0.05, Student’s t test.

Fig. S3.

PER1 protein levels in hPER3 Tg mice. (A) Representative Western blots of liver extracts of hPER3-WT-Tg and hPER3-P415A/H417R-Tg mice obtained on LL day 19. LMNB1 and β-actin were used as loading controls in the nuclear and cytoplasmic fractions, respectively. (B) Quantification of the Western blots in A. Expression levels were normalized to average levels in hPER3-WT-Tg mice. Results are expressed as mean ± SEM. ***P < 0.001, Student’s t test.

Fig. 5.

hPER3-P415A/H417R-Tg mice show depression-like behavior on the tail suspension test. (A and B) Duration of immobility in the tail suspension test for WT vs. hPER3-P415A/H417R-Tg mice (A) and for WT vs. hPER3-WT-Tg mice (B). (C) Quantification of the effects of imipramine on immobility of WT vs. hPER3-P415A/H417R-Tg mice in the tail suspension test. Error bars show mean ± SEM (n = 4–22). **P < 0.01, Student’s t test.

Fig. S4.

hPER3-P415A/H417R-Tg mice exhibit decreased sleep in LD cycles compared with hPER3-WT-Tg and WT mice. EEG analyses for WT (light gray) vs. hPER3-P415A/H417R-Tg (black) mice (A and C) and WT (light gray) vs. hPER3-WT-Tg (dark gray) mice (B and D) were used to quantitate sleep under baseline conditions (A and B) and after sleep deprivation (C and D). (A1 and B1) Percentage of time spent awake. (A2 and B2) Percentage of time spent in NREM sleep. (A3 and B3) Percentage of time spent in REM sleep. Quantification of mean episode duration and number of episodes for (A4 and B4) wakefulness, (A5 and B5) NREM sleep, and (A6 and B6) REM sleep. Average spectral power was calculated for each of the 6-h periods (from left to right: L1, L2, D1, and D2) as indicated by the horizontal bars in (A1–A3 and B1–B3). (A7 and B7) NREM δ and (A8 and B8) REM θ powers were compared for the two genotypes. (C1 and D1) Percentage of time spent awake. (C2 and D2) Percentage of time spent in NREM sleep. (C3 and D3) Percentage of time spent in REM sleep. Average spectral power was calculated for each of the 6-h periods (from left to right: L1/SD, L2, D1, and D2) as indicated by the horizontal bars in C1–C3 and D1–D3. (C4 and D4) NREM δ and (C5 and D5) REM θ powers were compared for the two genotypes. (C6 and D6) Percent changes of time after sleep deprivation compared with the baseline condition for NREM and REM in L2, D1, and D2 for the two genotypes. (C7 and D7) Cumulative NREM and REM sleep loss and gain compared with baseline conditions for the sleep deprivation experiment. (C8 and D8) Analysis of spectral sleep power changes compared with baseline conditions for L2, D1, and D2. Error bars represent SEM (n = 7). *P < 0.05, Student’s t test.

Fig. S4.

hPER3-P415A/H417R-Tg mice exhibit decreased sleep in LD cycles compared with hPER3-WT-Tg and WT mice. EEG analyses for WT (light gray) vs. hPER3-P415A/H417R-Tg (black) mice (A and C) and WT (light gray) vs. hPER3-WT-Tg (dark gray) mice (B and D) were used to quantitate sleep under baseline conditions (A and B) and after sleep deprivation (C and D). (A1 and B1) Percentage of time spent awake. (A2 and B2) Percentage of time spent in NREM sleep. (A3 and B3) Percentage of time spent in REM sleep. Quantification of mean episode duration and number of episodes for (A4 and B4) wakefulness, (A5 and B5) NREM sleep, and (A6 and B6) REM sleep. Average spectral power was calculated for each of the 6-h periods (from left to right: L1, L2, D1, and D2) as indicated by the horizontal bars in (A1–A3 and B1–B3). (A7 and B7) NREM δ and (A8 and B8) REM θ powers were compared for the two genotypes. (C1 and D1) Percentage of time spent awake. (C2 and D2) Percentage of time spent in NREM sleep. (C3 and D3) Percentage of time spent in REM sleep. Average spectral power was calculated for each of the 6-h periods (from left to right: L1/SD, L2, D1, and D2) as indicated by the horizontal bars in C1–C3 and D1–D3. (C4 and D4) NREM δ and (C5 and D5) REM θ powers were compared for the two genotypes. (C6 and D6) Percent changes of time after sleep deprivation compared with the baseline condition for NREM and REM in L2, D1, and D2 for the two genotypes. (C7 and D7) Cumulative NREM and REM sleep loss and gain compared with baseline conditions for the sleep deprivation experiment. (C8 and D8) Analysis of spectral sleep power changes compared with baseline conditions for L2, D1, and D2. Error bars represent SEM (n = 7). *P < 0.05, Student’s t test.

Fig. S4.

hPER3-P415A/H417R-Tg mice exhibit decreased sleep in LD cycles compared with hPER3-WT-Tg and WT mice. EEG analyses for WT (light gray) vs. hPER3-P415A/H417R-Tg (black) mice (A and C) and WT (light gray) vs. hPER3-WT-Tg (dark gray) mice (B and D) were used to quantitate sleep under baseline conditions (A and B) and after sleep deprivation (C and D). (A1 and B1) Percentage of time spent awake. (A2 and B2) Percentage of time spent in NREM sleep. (A3 and B3) Percentage of time spent in REM sleep. Quantification of mean episode duration and number of episodes for (A4 and B4) wakefulness, (A5 and B5) NREM sleep, and (A6 and B6) REM sleep. Average spectral power was calculated for each of the 6-h periods (from left to right: L1, L2, D1, and D2) as indicated by the horizontal bars in (A1–A3 and B1–B3). (A7 and B7) NREM δ and (A8 and B8) REM θ powers were compared for the two genotypes. (C1 and D1) Percentage of time spent awake. (C2 and D2) Percentage of time spent in NREM sleep. (C3 and D3) Percentage of time spent in REM sleep. Average spectral power was calculated for each of the 6-h periods (from left to right: L1/SD, L2, D1, and D2) as indicated by the horizontal bars in C1–C3 and D1–D3. (C4 and D4) NREM δ and (C5 and D5) REM θ powers were compared for the two genotypes. (C6 and D6) Percent changes of time after sleep deprivation compared with the baseline condition for NREM and REM in L2, D1, and D2 for the two genotypes. (C7 and D7) Cumulative NREM and REM sleep loss and gain compared with baseline conditions for the sleep deprivation experiment. (C8 and D8) Analysis of spectral sleep power changes compared with baseline conditions for L2, D1, and D2. Error bars represent SEM (n = 7). *P < 0.05, Student’s t test.

Fig. S4.

hPER3-P415A/H417R-Tg mice exhibit decreased sleep in LD cycles compared with hPER3-WT-Tg and WT mice. EEG analyses for WT (light gray) vs. hPER3-P415A/H417R-Tg (black) mice (A and C) and WT (light gray) vs. hPER3-WT-Tg (dark gray) mice (B and D) were used to quantitate sleep under baseline conditions (A and B) and after sleep deprivation (C and D). (A1 and B1) Percentage of time spent awake. (A2 and B2) Percentage of time spent in NREM sleep. (A3 and B3) Percentage of time spent in REM sleep. Quantification of mean episode duration and number of episodes for (A4 and B4) wakefulness, (A5 and B5) NREM sleep, and (A6 and B6) REM sleep. Average spectral power was calculated for each of the 6-h periods (from left to right: L1, L2, D1, and D2) as indicated by the horizontal bars in (A1–A3 and B1–B3). (A7 and B7) NREM δ and (A8 and B8) REM θ powers were compared for the two genotypes. (C1 and D1) Percentage of time spent awake. (C2 and D2) Percentage of time spent in NREM sleep. (C3 and D3) Percentage of time spent in REM sleep. Average spectral power was calculated for each of the 6-h periods (from left to right: L1/SD, L2, D1, and D2) as indicated by the horizontal bars in C1–C3 and D1–D3. (C4 and D4) NREM δ and (C5 and D5) REM θ powers were compared for the two genotypes. (C6 and D6) Percent changes of time after sleep deprivation compared with the baseline condition for NREM and REM in L2, D1, and D2 for the two genotypes. (C7 and D7) Cumulative NREM and REM sleep loss and gain compared with baseline conditions for the sleep deprivation experiment. (C8 and D8) Analysis of spectral sleep power changes compared with baseline conditions for L2, D1, and D2. Error bars represent SEM (n = 7). *P < 0.05, Student’s t test.

Fig. 6.

PER3^−/− mice demonstrate depression-like behavior in short photoperiods. (A) Duration of immobility of WT vs. Per3^−/− mice in the tail suspension test under LD 12:12 and LD 4:20. (B and C) Latency to immobilization (B) and duration of immobility (C) of WT vs. Per3^−/− mice in the forced swim test under LD 12:12 and LD 4:20. (D) Treatment with imipramine (20 mg/kg i.p.) rescues the tail suspension phenotype in Per3^−/− mice under LD 4:20. Error bars show mean ± SEM (n = 6 per group). (E) Treatment with imipramine (20 mg/kg i.p.) rescues the forced swim phenotype in Per3^−/− mice under LD 4:20. Error bars represent mean ± SEM (n = 6 each genotype). (F) Amount of sucrose water intake in WT and Per3^−/− mice under LD 4:20. Error bars represent mean ± SEM (n = 4–22). *P < 0.05, **P < 0.01, ***P < 0.005, Student’s t test.