HNF4A defines tissue-specific circadian rhythms by beaconing BMAL1::CLOCK chromatin binding and shaping the rhythmic chromatin landscape

Transcription modulated by the circadian clock is diverse across cell types, underlying circadian control of peripheral metabolism and its observed perturbation in human diseases. We report that knockout of the lineage-specifying Hnf4a gene in mouse liver causes associated reductions in the genome-wide distribution of core clock component BMAL1 and accessible chromatin marks (H3K4me1 and H3K27ac). Ectopically expressing HNF4A remodels chromatin landscape and nucleates distinct tissue-specific BMAL1 chromatin binding events, predominantly in enhancer regions. Circadian rhythms are disturbed in Hnf4a knockout liver and HNF4A-MODY diabetic model cells. Additionally, the epigenetic state and accessibility of the liver genome dynamically change throughout the day, synchronized with chromatin occupancy of HNF4A and clustered expression of circadian outputs. Lastly, Bmal1 knockout attenuates HNF4A genome-wide binding in the liver, likely due to downregulated Hnf4a transcription. Our results may provide a general mechanism for establishing circadian rhythm heterogeneity during development and disease progression, governed by chromatin structure.


Field-specific reporting
Please select the one below that is the best fit for your research. If you are not sure, read the appropriate sections before making your selection.

Life sciences
Behavioural & social sciences Ecological, evolutionary & environmental sciences For a reference copy of the document with all sections, see nature.com/documents/nr-reporting-summary-flat.pdf

Life sciences study design
All studies must disclose on these points even when the disclosure is negative. By convention and in line with our other studies (PMID: 30530698), we used 2 or 3 biological replicates for NGS experiments, 3-4 biological replicates for RT-qPCR assays, and 4-5 biological replicates for circadian period calculations.
No data were excluded from the analyses.
All experiments used replicates which substantially agreed with each other.
Control or knockout mice were randomly allocated into the experiments after their genotypes were determined. In the liver histology experiments, for each tissue sample we examined 3 or 4 randomly selected imaging areas.
The investigators were not blinded during data collection (wet lab work) because blinding in not a common practice in the area of molecular biology. However, authors in charge of data analysis were blinded to the identity of experimental samples.
Antibodies  Note that full information on the approval of the study protocol must also be provided in the manuscript.

ChIP-seq Data deposition
Confirm that both raw and final processed data have been deposited in a public database such as GEO.
Confirm that you have deposited or provided access to graph files (e.g. BED files) for the called peaks.

Data access links
May remain private before publication.

Files in database submission
Genome browser session (e.g. UCSC)

Methodology
Replicates Sequencing depth Antibodies All antibodies used in the western blots have been validated by the manufacturers, stating that the antibodies recognize endogenous levels of the target proteins in mouse tissues. For all antibodies used in the ChIP-seq experiments, the manufacturers state that they recognize endogenous target proteins in both mouse tissues and human cell cultures and are suitable for ChIP-seq assays. Besides, our results demonstrated their strong activities.
The Hep3B cell line was a gift from Dr. Michael Karin at UCSD, originally purchased from ATCC. HEK 293T, U2OS, and HepG2 cell lines were directly purchased from ATCC.
All experiments in this study used low-passage cell cultures. The morphology of each cell line was consistent with images and descriptions on the ATCC website. RT-qPCR of liver cancer cell lines (Hep3B and HepG2) indicated high expression level of liver-specific factors.
All cell lines were periodically confirmed to be mycoplasma-free by using the PlasmoTest mycoplasma detection kit (InvivoGen #rep-pt1).
No commonly misidentified cell lines were used in the study.
All animal care and experiments were performed under the institutional protocols approved by the Institutional Animal Care and Use Committee (IACUC, #20826) at the University of Southern California. Hnf4a floxed mice (The Jackson Library #004665) were crossed with Albumin-Cre mice (The Jackson Library #003574) and Per2-luciferase reporter mice (The Jackson Library #006852) to obtain Hnf4afl/+;Alb-Cre+/-;Per2-luc+/+ and Hnf4afl/fl;Alb-Cre-/-;Per2-luc+/+ mice, which were then mated to obtain Hnf4afl/fl;Alb-Cre +/-;Per2-luc+/+ (HKO) and Hnf4afl/fl;Alb-Cre-/-;Per2-luc+/+ (Control) littermates. Bmal1 floxed mice (The Jackson Library #007668) were crossed with mice expressing Albumin-Cre (The Jackson Library #003574) to obtain Arntlfl/+;Alb-Cre+/-and Arntlfl/fl;Alb-Cre-/mice which were then mated to obtain Arntlfl/fl;Alb-Cre+/-(BKO) and Arntlfl/fl;Alb-Cre-/-(Control) littermates. In all experiments, male mice between 10 and 12 wk of age were used. In all experiments except the jet lag treatment, mice were housed in a room with controlled temperature of 21-23°C and humidity of 35-40% under a 12-h light/12-h dark (LD) cycle with free access to food and water. The chronic jet lag treatment was performed by housing experimental mice in the light-tight circadian cabinet and switching lighting conditions between two light onset schedules which are apart by 8 hours every three days from 7 to 11 wk of age.
The study did not involve wild animals.
The study did not involve samples collected from the field.
All animal care and experiments were performed under the institutional protocols approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Southern California.

No longer applicable
All ChIP-seq experiments used two or thee biological replicates. For ChIP-seq with mouse tissues, replicates were liver samples isolated from different mice. For ChIP-seq with cell lines, replicates were cells from different cell culture dishes. The replicates were agreeable with one another.
For each experiment, the total reads number was >25 million. >90% of the reads were uniquely mapped to the mouse or human genome. The reads were 150 bp long, single-end.