NPAS2 Compensates for Loss of CLOCK in Peripheral Circadian Oscillators

Abstract

Heterodimers of CLOCK and BMAL1 are the major transcriptional activators of the mammalian circadian clock. Because the paralog NPAS2 can substitute for CLOCK in the suprachiasmatic nucleus (SCN), the master circadian pacemaker, CLOCK-deficient mice maintain circadian rhythms in behavior and in tissues in vivo. However, when isolated from the SCN, CLOCK-deficient peripheral tissues are reportedly arrhythmic, suggesting a fundamental difference in circadian clock function between SCN and peripheral tissues. Surprisingly, however, using luminometry and single-cell bioluminescence imaging of PER2 expression, we now find that CLOCK-deficient dispersed SCN neurons and peripheral cells exhibit similarly stable, autonomous circadian rhythms in vitro. In CLOCK-deficient fibroblasts, knockdown of Npas2 leads to arrhythmicity, suggesting that NPAS2 can compensate for loss of CLOCK in peripheral cells as well as in SCN. Our data overturn the notion of an SCN-specific role for NPAS2 in the molecular circadian clock, and instead indicate that, at the cellular level, the core loops of SCN neuron and peripheral cell circadian clocks are fundamentally similar.

Fig 1. Clock^-/- SCN neurons show stable circadian rhythms in cultured SCN explants.

(A) Raster plots of mPer2^Luc bioluminescence intensity of individual wild-type (left, n = 26) and Clock^-/- (right, n = 26) SCN cells in slices. Each horizontal line represents a single cell, with time in days plotted left to right. Values above and below the mean are shown in red and green, respectively. (B) mPer2^Luc bioluminescence rhythms of two representative rhythmic individual neurons (black and gray curves) in organotypic SCN slice cultures from wild-type mice (left) and Clock^-/- mice (right). (C) Mean circadian period, amplitude, and sine wave goodness-of-fit of cellular mPer2^Luc rhythms, and the percentage of rhythmic neurons in organotypic SCN slice cultures from wild type (white), Clock^-/- (black), and Bmal1^-/- (patterned) mice. Bmal1^-/- data are from Ko et al [18]. Bmal1^-/- SCN slices showed unusual lability of period, not reflected in these summary statistics. Data are shown as mean ± SEM; ***p≤0.001 (student’s t-test); or as percentage of cells that were significantly rhythmic; WT: n (rhythmic/total) = 75/75; Clock^-/-: n = 79/79; Bmal1^-/-: n = 80/80.

Fig 2. The SCN oscillator network is responsible for stable rhythms in Clock^-/- SCN neurons.

(A) Raster plots of mPer2^Luc bioluminescence intensity of individual dispersed wild-type (left, n = 246) and Clock^-/- (right, n = 161) SCN cells. Data are presented as in Fig 1C. (B) mPer2^Luc bioluminescence rhythms of two representative rhythmic individual dispersed SCN neurons from wild type (left) and Clock^-/- mice (right). (C) Circadian period, amplitude, and sine wave goodness-of-fit of cellular mPer2^Luc rhythms, and the percentage of rhythmic neurons in dispersed SCN cultures from wild type (white), Clock^-/- (black), and Bmal1^-/- (patterned) mice. Bmal1^-/- data are from Ko et al [18]. Data are shown as mean ± SEM; ***p≤0.001 (student’s t-test); or as percentage of cells that are significantly rhythmic; WT: n (rhythmic/total) = 234/255; Clock^-/-: n = 138/208; Bmal1^-/-: n = 30/243.

Fig 3. Dispersed Clock^-/- fibroblasts show circadian rhythms comparable to those of dispersed Clock^-/- SCN neurons.

(A) Raster plots of bioluminescence intensity of individual dispersed wild-type (left, n = 320) and Clock^-/- (right, n = 121) SCN cells. Data are presented as in Fig 1C. (B) Images of mPer2^Luc expression of representative rhythmic wild-type (left) and Clock^-/- (right) fibroblasts. (C) Two representative mPer2^Luc bioluminescence rhythms of individual rhythmic wild-type (left) and Clock^-/- (right) fibroblasts. (D) Circadian period, amplitude, and goodness of fit of mPer2^Luc rhythms, and % of cells that were significantly rhythmic, for individual wild type (left) and Clock^-/- (right) fibroblasts. Data are shown as mean ± SEM; *p≤0.05, ***p≤0.001 (Student’s t-test); or % of cells rhythmic; ***p≤0.001 (Fisher’s exact test); WT: n (rhythmic/total) = 320/321; Clock^-/-: n = 121/163.

Fig 4. Peripheral organs of Clock^-/- mice can exhibit circadian rhythms in vitro.

(A) Two representative mPer2^Luc bioluminescence rhythms of rhythmic organotypic liver (black), lung (blue), kidney (red), and adrenal (green) slice cultures from wild type (left) and Clock^-/- (right) mice. After ~7 culture days, samples were treated with 10 μM forskolin (arrow). Y-axis scales are adjusted to amplitudes for better visualization of data. (B) Circadian mPer2^Luc rhythm period, amplitude, damping constant (days to reach 1/e of initial amplitude), and phase of first peak after forskolin treatment, and % of slices from wild type (unfilled bars) and Clock^-/- mice (filled bars) that were significantly rhythmic after forskolin treatment (culture days 8–14). Data are shown as mean ± SEM; *p≤0.05, **p≤0.01, ***p≤0.001 (student’s t-test); or % of slices rhythmic; n = 8.

Fig 5. Knockdown of Npas2 expression suppresses circadian rhythms in Clock^-/- fibroblasts.

(A) Fibroblasts dispersed from wild type and Clock^-/- mice were treated with lentiviral vectors carrying an Npas2-KD or scrambled DNA sequence, as well as a GFP reporter. Simultaneous fluorescence and bioluminescence images of a representative field show GFP expression marking transfected cells (left, green) and mPer2^Luc bioluminescence from both transfected and untransfected cells (middle, red). The overlay shows that circadian rhythms could be measured from both transfected (filled arrowheads) and untransfected (unfilled arrowheads) fibroblasts in the same culture dish. (B) Percentage of wild type (left) and Clock^-/- (right) fibroblasts treated with Npas2-KD lentiviruses (top) or scrambled lentiviruses (bottom) that were significantly rhythmic. ***p≤0.001 (Fisher’s exact test); number of cells are given in parentheses. (C) mPer2^Luc rhythms of representative individual dispersed untransfected (top) and transfected (bottom) Clock^-/- fibroblasts from the same culture dish.