Biological clocks coordinate the spatiotemporal organization of metabolism with diurnal time cues, and bulk-tissue RNA sequencing studies in mammals have revealed widespread circadian (about 24 h) gene expression in mice and non-human primates. Yet whether and how cell-type-specific organization of circadian gene expression is achieved in metabolic tissues has remained unclear, largely because bulk-tissue profiling averages away the cellular heterogeneity of organs such as liver and skeletal muscle.
A new study published in Life Metabolism by Professor Min-Dian Li at Army Medical University and Professor Dingbang Ma at the Interdisciplinary Research Center on Biology and Chemistry, Chinese Academy of Sciences, now provides evidence that circadian transcription in these tissues is organized in a cell-type-specific manner. The team employed time-resolved single-nucleus RNA sequencing (snRNA-seq) to profile circadian transcription across cell types in skeletal muscle and liver from female mice, and identified cell-type-specific rhythmic gene programs, phase-specific pathway activities, and candidate circadian transcription factors in zonated hepatocytes and metabolically distinct subtypes of myonuclei.
Using snRNA-seq of gastrocnemius muscle and liver sampled every 3 hours over a 24-hour period under constant darkness, the researchers profiled circadian transcription across cell types, including hepatocytes resolved along the portal–central vein axis (Zones 1–4) and myonuclei resolved into subtypes including type I, IIa, IIb, and IIx fibers.
Rather than a uniform tissue-wide rhythm, they found that rhythmic gene content varied by cell identity. Each subtype contained 1,300–3,000 (6%–13%) and 600–1,600 (4%–8%) rhythmic genes in liver and skeletal muscle, respectively. Zonated hepatocytes exhibited substantially greater circadian rhythmicity than nonparenchymal cells, and type IIb myonuclei exhibited the largest number of rhythmic genes among myonuclei.
Further experiments revealed that metabolically distinct subtypes activated different pathways at different phases. Periportal Zone 1 hepatocytes showed cAMP signaling, FOXO signaling, xenobiotic metabolism, and nutrient metabolism active at the inactive and active phases, respectively, whereas pericentral Zone 4 hepatocytes upregulated HIF-1 signaling, glycolysis, and pyrimidine metabolism during the active phase. In muscle, oxidative type IIa myonuclei showed robust activity of oxidative phosphorylation, the citric acid cycle, and PPAR and PI3K-AKT signaling in the late active phase, while glycolytic type IIb myonuclei activated glycolysis/gluconeogenesis and ribosomal pathways in the inactive phase.
When the team applied SCENIC regulon analysis, they observed candidate circadian regulators specific to these subtypes, including Srebf1, Foxo1, and Tead3 in Zone 1 hepatocytes, selectively enriched and rhythmically expressed nuclear receptors including Nr1h4 (FXR) and Ppara in hepatocytes, and Ppara, Esrrg, and Klf15 linked to oxidative myonuclei, with Tbx15 associated with glycolytic myonuclei. A validation cohort of adult female wild-type or Clock mutant mice fed a high-fat diet for 12 weeks provided independent bulk tissue-level evidence that several of these regulators, including Srebf1, Foxo1, Nr1h4, Ppara, Klf15, and Tbx15, are regulated by the circadian clock system. Consistent with this, the rhythmic programs differed substantially between subtypes: only 29 rhythmic genes were shared among myonuclei subtypes (including core clock genes such as Arntl, Clock, and Per2), whereas 1,481 rhythmic genes were shared across hepatocyte zones.
Together, these findings challenge the view that circadian transcription in metabolic tissues is a largely uniform, tissue-level phenomenon, showing instead that liver zonation and myonuclear metabolic type are associated with differential patterns of circadian metabolic gene expression and linked to circadian activities of transcription factors. The work identifies a cell-type-resolved circadian landscape and candidate transcription factors, including druggable metabolic nuclear receptors, suggesting potential circadian therapeutic windows, and provides a resource (accessible via CircaMetDB, www.circametdb.org.cn) for investigating cell-type-specific circadian transcriptional programs.
