Hepatocyte SREBP signaling mediates clock communication within the liver

Abstract

Rhythmic intraorgan communication coordinates environmental signals and the cell-intrinsic clock to maintain organ homeostasis. Hepatocyte-specific KO of core components of the molecular clock Rev-erbα and -β (Reverb-hDKO) alters cholesterol and lipid metabolism in hepatocytes as well as rhythmic gene expression in nonparenchymal cells (NPCs) of the liver. Here, we report that in fatty liver caused by diet-induced obesity (DIO), hepatocyte SREBP cleavage-activating protein (SCAP) was required for Reverb-hDKO-induced diurnal rhythmic remodeling and epigenomic reprogramming in liver macrophages (LMs). Integrative analyses of isolated hepatocytes and LMs revealed that SCAP-dependent lipidomic changes in REV-ERB-depleted hepatocytes led to the enhancement of LM metabolic rhythms. Hepatocytic loss of REV-ERBα and β (REV-ERBs) also attenuated LM rhythms via SCAP-independent polypeptide secretion. These results shed light on the signaling mechanisms by which hepatocytes regulate diurnal rhythms in NPCs in fatty liver disease caused by DIO.

Keywords: Epigenetics; Hepatology; Homeostasis; Metabolism.

Figure 1. SREBP signaling is required for intraorgan communication from hepatocytes lacking REV-ERBs.

(A) UMAP visualization of cell clusters in livers from control and Reverb-hDKO mice fed a HFHSD for 4 weeks. (B) KEGG pathways enriched in the genes differentially expressed between hepatocyte cluster 2 versus cluster 1. (C) Number of DEGs in nonhepatocytes upon Reverb hDKO. (D) UMAP visualization of the overlapping cell cluster in livers from control, Reverb-hDKO, and Reverb/Scap-hTKO mice fed a HFHSD for 4 weeks. (E) Percentages of the indicated cell populations. (F and G) Heatmap of (F) and pathways enriched in (G) genes from LMs restored by Reverb/Scap hTKO. Hep, hepatocyte; Hb, hepatoblast; SC, stellate cell; EC, endothelial cell; B, B cell; T, T cell; erythroid, erythroid cell; Chol, cholangiocyte.

Figure 2. SREBP signaling is required for rhythmic transcriptomic remodeling in LMs upon loss of hepatocytic REV-ERBs.

(A) Relative mRNA expression of the indicated genes in livers from Reverb-hKO, and Reverb/Scap-hTKO mice fed a HFHSD for 4 weeks. The gene expression data are expressed as the mean ± SEM (n = 4–6 per group). (B) Number and percentage of control-, Reverb-hDKO–, and Reverb/Scap-hTKO–specific oscillating transcripts in hepatocytes. (C) Heatmap of rhythmic transcripts in hepatocytes or LMs dependent on hepatocytic REV-ERB. (D) Number and percentage of control-, Reverb-hDKO–, and Reverb/Scap-hTKO–specific oscillating transcripts in LMs. (E) Heatmap of enhanced rhythmic transcripts upon SCAP-dependent Reverb hDKO.

Figure 3. Ligands from hepatocytes reprogram REV-ERB–dependent oscillating enhancers in LMs.

(A and B) Heatmap (A) and IMAGE analysis (B) of rhythmic enhancers in LMs dependent on hepatocytic REV-ERBs. (C) Mean expression of putative target genes of JUND in isolated LMs from control liver. (D) Relative expression of Jund in LMs of livers from control, Reverb-hDKO, and Reverb/Scap-hTKO mice fed a HFHSD for 4 weeks. Data are presented as the mean ± SEM (n = 3 per time point). (E) Ligand-receptor pair analysis. Circle plots show links between the predicted ligands from hepatocytes (red) with their associated receptors from LMs (blue) associated with rhythmic transcripts in LMs dependent on hepatocytic REV-ERB potentially targeted by the ligand-receptor pairs. (F) Relative expression of Pdgfd in hepatocytes and Pdgfrb in LMs. Data are presented as the mean ± SEM (n = 3 per time point). (G) Relative expression of Jund in immortalized KCs with BSA (control) or PDGF treatment. Data are presented as the mean ± SEM (n = 3 per time point). The gene expression data in D, F, and G are expressed as the mean ± SEM (n = 4–6 per group). Amp, amplitude; adj.p, adjusted P value; Ctrl, control.

Figure 4. LM diurnal rhythms controlled by SCAP-dependent lipid metabolism in REV-ERB–depleted hepatocytes.

(A) Number and percentage of control-, Reverb-hDKO–, and Reverb/Scaph-DTKO–specific oscillating lipids. (B) TAG synthesis pathways of SCAP-dependent oscillating lipids whose species number was increased (red) or decreased (green) upon REV-ERB disruption. PI, phosphatidylinositol. (C) Heatmap of rhythmic lipids in LMs upon hepatocytic REV-ERB disruption independent of or dependent on SCAP. The rhythms of SCAP-independent lipids upon hepatocytic REV-ERB disruption are labeled in purple, and SCAP-dependent lipids are labeled in violet. The lipids whose abundance was increased upon hepatocytic REV-ERB disruption are highlighted in a light-green background. (D) GO terms enriched in gained rhythmic genes in LMs upon SCAP-dependent hepatocytic REV-ERB disruption.