Targeting the chromatin remodeler BAZ2B mitigates liver aging and fibrosis

Increasing age enhances the liver's vulnerability to nonalcoholic steatohepatitis (NASH) with fibrosis. Unraveling the intricate molecular relationship among aging, cellular senescence, and NASH is critical for developing effective treatments, yet it remains a daunting challenge. Here we report a vital epigenetic mechanism that links liver aging to NASH fibrosis. We find that up-regulation of a chromatin remodeler BAZ2B in a subpopulation of hepatocytes contributes to the pathological progression of NASH in patients. Genetic ablation or hepatocyte-specific knockdown of Baz2b rejuvenates hepatic cells and limits diet-induced steatohepatitis and liver fibrosis via preserving the peroxisome proliferator-activated receptor a (PPARa)-mediated lipid metabolism, which was impaired in both naturally aging and NASH mouse livers. Mechanistically, Baz2b down-regulates the expression of genes related to the PPARa signaling pathway by directly binding their promoter regions and reducing chromatin accessibility. Our study thus unravels the BAZ2B-PPARa-lipid-metabolism signal axis as a key pathway linking liver aging to NASH fibrosis, suggesting that BAZ2B is a potential therapeutic target for hepatic fibrosis. Overall design: For ChIP assays, chromatin immunoprecipitation was performed as described with some modifications 46. Briefly, for liver tissue, frozen samples were chopped into small pieces, immediately incubated in freshly prepared disuccinimidyl glutarate solution (2 mM in PBS), and rotated for 30 min at room temperature; for cell lines, samples were collected from 10 cm cultured dishes using cell scrapers. Then samples were washed in PBS buffer twice, crosslinked by 1% formaldehyde (in PBS with proteinase-inhibitor cocktail), and fixed on a nutator shaker for 10 min at room temperature. Fixed samples were quenched with 0.125 M glycine and washed with cold PBS buffer with protease inhibitors three times. Ground the liver tissues with a dounce homogenizer on ice, passed the homogenized tissue suspension through a 70 µM cell strainer to remove the connective tissue, and then washed samples with cold PBS buffer with protease inhibitors three times. Samples were then lysed by ice-cold lysis buffer (50 mM HEPES-KOH pH 7.5, 150 mM NaCl, 1 mM EDTA, 0.1% sodium deoxycholate, 1% Triton X-100, 1 mM PMSF and protease inhibitor cocktail) with 0.1% SDS. Lysates were digested using 0.5 µl micrococcal nuclease (MNase) (10011S, CST) in reaction buffer (50 mM Tris, 5mM CaCl2, pH 8.0) for 3 min at 37?°C, and then were stopped with 25 mM EDTA. About 10 µg of the digested chromatin was used for immunoprecipitation. Primary antibodies including anti-H3K4me3 (Abcam ab8580), anti-H3K27ac (Abcam ab4729), and anti-Flag (F1804, Sigma-Aldrich) were incubated with chromatin overnight at 4 ?C with rotation. The next day, 1.5 mg Dynabeas® Protein A beads (Invitrogen, 10001D) were added to each immunoprecipitated sample, and incubated at 4 ?C for 2 hours with rotation to collect chromatin/antibodies. DNAs were eluted from washed magnetic beads with incubation of SDS elution buffer (1% SDS,10 mM EDTA, 50 mM Tris-HCl) overnight at 65 °C. The remaining RNA and protein residues were removed by incubating with 3 mL of 0.5 mg/mL RNaseA at 37 °C for 15 min and then 3 mL of 20 mg/mL Proteinase K at 65 °C for 30 min. The ChIP-seq libraries were prepared using the NEBNext Ultra II DNA Library Prep Kit (New England BioLabs) and NEBNext Multiplex Oligos for Illumina (New England BioLabs). Sequencing was performed on an Illumina NovaSeq 6000. ATAC-seq Briefly, frozen liver samples were chopped into small pieces, and immediately incubated in homogenization buffer (10mM Tris, 0.25mM sucrose, 25mM KCl, 5mM MgCl2, 0.1 mM DTT, 1% BSA, 0.1% NP 40 with RNase inhibitor). Ground the liver tissue pieces with a dounce homogenizer on ice, passed the homogenized tissue suspension through a 70 µM and 40 µM cell strainer, washed once in cold Nuclei Wash Buffer (20% iodixanol in HB solution), and then resuspended samples in Blocking Buffer (10% BSA in PBS with RNase inhibitor). The segmentation reaction was performed using transposases (Vazyme, TD501) for 30 min at 37 °C. Segmented DNAs were purified and amplified using the TruePrepTM Index Kit (Vazyme, TD202). DNA sequencing was performed using an Illumina NovaSeq 6000. Total RNAs of livers were isolated and purified by using TRIzol (Invitrogen) and the RNeasy Mini kit (Qiagen), respectively. Then these RNA samples were subjected to library preparation and sequencing. The RNA-seq library was prepared by TruSeq Stranded Total RNA Sample Preparation kit (Illumina) for 150-bp paired-end sequencing. Poly-A-containing mRNAs were purified, fragmented to 300 bp nucleotides, and used for the synthesis of single-stranded cDNAs by random hexamer priming. Then the constructed single-stranded cDNAs were used as templates to synthesize the second-strand cDNA, leading to the generation of double-stranded cDNAs which were then amplified by polymerase chain reaction (PCR). Subsequently, qualified cDNA libraries were sequenced on an Illumina NextSeq500 platform. The HISAT aligner with –rna-strandness F 44 was used to map sequencing reads to the mouse reference genome GRCm38 (mm10) and mapped reads were counted using HTSeq to obtain raw read counts for each gene 45. The R package DESeq2 46 was used to identify the differentially expressed genes (DEGs) between groups. Genes with a P < 0.05 were designated as DEGs between groups.