A non-canonical E-box within the MyoD core enhancer is necessary for circadian expression in skeletal muscle

Abstract

The myogenic differentiation 1 (MyoD) gene is a master regulator of myogenesis. We previously reported that the expression of MyoD mRNA oscillates over 24 h in skeletal muscle and that the circadian clock transcription factors, BMAL1 (brain and muscle ARNT-like 1) and CLOCK (circadian locomotor output cycles kaput), were bound to the core enhancer (CE) of the MyoD gene in vivo. In this study, we provide in vivo and in vitro evidence that the CE is necessary for circadian expression of MyoD in adult muscle. Gel shift assays identified a conserved non-canonical E-box within the CE that is bound by CLOCK and BMAL1. Functional analysis revealed that this E-box was required for full activation by BMAL1/CLOCK and for in vitro circadian oscillation. Expression profiling of muscle of CE(loxP/loxP) mice found approximately 1300 genes mis-expressed relative to wild-type. Based on the informatics results, we analyzed the respiratory function of mitochondria isolated from wild-type and CE(loxP/loxP) mice. These assays determined that State 5 respiration was significantly reduced in CE(loxP/loxP) muscle. The results of this work identify a novel element in the MyoD enhancer that confers circadian regulation to MyoD in skeletal muscle and suggest that loss of circadian regulation leads to changes in myogenic expression and downstream mitochondrial function.

Figure 1.

Circadian oscillation of MyoD mRNA is dampened in skeletal muscle tissues of the CE^loxP/loxP mouse. Adult C57BL/6 J and CE^loxP/loxP mice were entrained to 12 h light/12 h dark conditions for 2 weeks. Thirty hours after lights were turned off (CT18), TA muscles were collected every 4 h from CT18 to CT50. Transcript levels were determined by RT–PCR. Bmal1, Per2 and MyoD mRNA levels were normalized to Rpl26 mRNA levels. (A) Bmal1 and Per2 expression in wild-type mice TA muscle; (B) Luminescence recording from Per2::Luc soleus muscles. The data for the graphs represent the mean values ± SEM of raw data for five animals. (C) Bmal1 and Per2 expression in CE^loxP/loxP mice TA muscle; (D) Luminescence recordings from Per2::Luc × CE^loxP/loxP soleus muscles. The data for the graphs represent the mean values ± SEM of raw data for five animals. (E) MyoD expression in wild-type and CE^loxP/loxP mice TA muscle; (F) Average period length data of bioluminescence from Per2::Luc and Per2::Luc × CE^loxP/loxP soleus muscles. The graphs in (A, C and E) display the results as the mean value ± SEM for three independent samples for each genotype. The gray bars indicate the subjective circadian light period and the black bars indicate subjective circadian dark period. For each respective gene, one-way ANOVA followed by Tukey's post-hoc test was performed to determine significant differences across all time points. For circadian analysis of gene expression, the JTK_CYCLE program was used.

Figure 2.

The non-canonical E-box of the MyoD CE is required for CLOCK and BMAL1 binding and activation of the MyoD promoter by BMAL1: CLOCK. EMSA analysis of CLOCK and BMAL1 binding at the E-boxes located within the MyoD CE was performed with nuclear extract derived from C2C12 cells over-expressing Clock and Bmal1 and an oligonucleotide probe containing one of the five CE E-boxes. (A) The DNA:protein complexes formed using a probe containing one of the canonical E-boxes (E1, E2, E3 or E4). (B) The DNA:protein complexes formed using the non-canonical E-box and the mutant of the non-canonical E-box. The arrow indicates the specific DNA:protein complex that contained CLOCK and BMAL1. (C) A wild-type (CE-4.7MyoD:luciferase) or mutant (CEmt-4.7MyoD:luciferase) reporter gene construct was transiently co-transfected with Clock plus Bmal1 expression plasmids into C2C12 myoblasts. Luciferase activity was measured 24–48 h later. The histogram displays the results from three independent experiments with values presented as mean ± SEM. A one-way ANOVA test indicated CE-4.7MyoD:luciferase reporter gene activation by CLOCK plus BMAL1 (empty bar) was significantly greater than control (gray bar) denoted by asterisks.

Figure 3.

Serum shock is sufficient to synchronize circadian gene expression in C2C12 myotubes. Stable clones of C2C12 cells (CE-4.7MyoD) were differentiated for 5 days prior to serum shock. RNA was isolated at the indicated time post-serum shock and used to determine the expression level of Bmal1 and MyoD by RT-PCR. The target transcript levels were normalized to Rpl26 transcript levels and then normalized to the highest point across all time points. The data for the graphs represent the mean values ± SEM for three independent clones for the reporter gene. The expression of endogenous Bmal1 (A), endogenous MyoD (B) all displayed a circadian oscillation as indicated by significant differences in peak and trough expression. For each respective gene, one-way ANOVA followed by Tukey post-hoc test identified significant difference (P < 0.05) between peak expression and other time points as indicated by an asterisk. For circadian analysis of gene expression the JTK_CYCLE program was used.

Figure 4.

Mutation of the MyoD CE non-canonical E-box disrupts circadian oscillation in bioluminescent assay. Stable clones of C2C12 cells (carrying the Bmal1 promoter:luciferase, CE-4.7MyoD:luciferase or CEmt-4.7MyoD:luciferase reporter) were differentiated for 5 days prior to serum shock. Then the cultures were incubated in Lumicycle at 37°C and bioluminescence was recorded. The graphs shown here are baseline subtracted results as the mean value ± SEM from three independent clones for the reporter gene. (A) Bmal1 promoter:luciferase reporter. (B) CE-4.7MyoD:luciferase reporter. (C) CEmt-4.7MyoD:luciferase reporter. The graphs display the results as the mean value ± SEM from three independent clones for each reporter. For circadian analysis of gene expression, the JTK_CYCLE program was used.

Figure 5.

Respiration rates are lower in the mitochondria of the CE^loxP/loxP than in wild-type. Gastrocnemius muscles from 5- to 6-month-old mice were isolated from wild-type (empty bars) or the CE^loxP/loxP (gray bars) and used for the determination of respiration rates. The data for the graphs represent the mean values ± SEM for four animals.