All animal experiments were reviewed and approved by the Nanjing Medical University Ethics Committee on Humane Treatment of Experimental Animals (approval #: IACUC-1811060). The mice were maintained in an SPF environment with 12 h light/dark cycles and libitum access to food and water. All animal procedures conform to the guidelines on the protection of animals per the current NIH guidelines. Hepatocyte BRG1 conditional deletion mice were generated by crossing the Smarca4^f/f mice to the Alb-Cre mice as previously described (Hong et al., 2020; Dong et al., 2021; Kong et al., 2021). To induce non-alcoholic fatty liver disease (NAFLD), the mice were fed a high-fat high-carbohydrate diet (D12492, Research Diets) for 16 consecutive weeks or a methionine- and choline-deficient (MCD) diet (A02082002B, Research Diets) for eight consecutive weeks as previously described (Fan et al., 2020; Kong et al., 2021). Euthanasia was performed by intraperitoneal injection with pentobarbital sodium (150 mg/kg).

Primary hepatocytes were isolated from WT and BRG1^LKO mice and treated with palmitate (0.4 mM) for 12 h as previously described (Hong et al., 2020; Dong et al., 2021). The TBK1 promoter-luciferase constructs (Li et al., 2003) and the SRC promoter-luciferase constructs (Ritchie et al., 2000) have been described previously. Mutant constructs were generated by a QuikChange kit (Thermo Fisher, Waltham, MA, United States) and verified by direct sequencing. Small interfering RNAs purchased from Dharmacon. Transient transfection was performed with Lipofectamine 2000. Cells were harvested 48 h after transfection. Luciferase activities were assayed 24–48 h after transfection using a luciferase reporter assay system (Promega) as previously described (Chen et al., 2021; Liu et al., 2021a,b; Zhang et al., 2021). For luciferase assay/DHE staining/luminescence assay, the cells were plated at the density of ∼1 × 10⁵ cells/well in 12-well culture dishes. For RNA extraction-qPCR experiments, the cells were plated at the density of ∼2 × 10⁵ cells/well in p35 culture dishes. For protein extraction-Western blotting experiments, the cells were plated at the density of ∼5 × 10⁵ cells/well in p60 culture dishes. For ChIP assay, the cells were plated at the density of ∼2 × 10⁶ cells/well in p100 culture dishes. Palmitate was dissolved in pre-heated 0.1 N NaOH and diluted 1:10 in pre-warmed 12% BSA solution at the final stock concentration of 10 mM. Stock palmitate solution was diluted with culture media to a final concentration of 0.4 mM and added to the cells; equivalent volume of vehicles was added as a control.

Whole cell lysates were obtained by re-suspending cell pellets in RIPA buffer (50 mM Tris pH7.4, 150 mM NaCl, 1% Triton X-100) with freshly added protease inhibitor (Roche) as previously described (Chen et al., 2020a; Sun et al., 2020; Wu T. et al., 2020; Wu X. et al., 2020). Western blot analyses were performed with anti-TBK1 (1:1,000, Abcam, ab40676), anti-c-Jun (1:2,000, Proteintech, 24909-1), anti-c-Fos (1:2,000, Proteintech, 66590-1), anti-Sp1 (1:1,000, Abcam, ab227383), anti-phospho-S172-TBK1 (1:1,000, Cell Signaling Tech, 5483), anti-SRC (1:1,000, Abcam, ab16885), and anti-β-actin (1:5,000, Sigma, A2228) antibodies. The anti-phospho-Y179-TBK antibody was custom-made using the synthesized peptide TEE(pY)LHPDM-C as immunogen for injection in Balb/c mice as previously described (Li et al., 2017). For densitometrical quantification, densities of target proteins were normalized to those of β-actin as previously described (Chen et al., 2020b,c; Li et al., 2020a; Mao et al., 2020a). Data are expressed as relative protein levels compared to the control group which is arbitrarily set as 1.

RNA was extracted with the RNeasy RNA isolation kit (Qiagen). Reverse transcriptase reactions were performed using a SuperScript First-strand Synthesis System (Invitrogen) as previously described (Dong et al., 2020; Li et al., 2020b,c; Lv et al., 2020). Real-time PCR reactions were performed on an ABI Prism 7500 system with the following primers: Tbk1, 5′-CTCATCACAGCCTACGGAGAC-3′ and 5′-CGTGTTGGAA TTGGTTGAGTCG-3′; Src, 5′-CCGAGCGGCTTCTTTACCC-3′ and 5′-GCATGATACATGATGCGGTAGT-3′. Ct values of target genes were normalized to the Ct values of housekeeping control gene (18s, 5′-CGCGGTTCTATTTTGTTGGT-3′ and 5′-TCGTCTTCGAAACTCCGACT-3′ for both human and mouse genes) using the ΔΔCt method and expressed as relative mRNA expression levels compared to the control group which is arbitrarily set as 1.

Chromatin immunoprecipitation (ChIP) assays were performed essentially as described before (Mao et al., 2020b; Yang et al., 2020a,b). In brief, chromatin in control and treated cells were cross-linked with 1% formaldehyde. Cells were incubated in lysis buffer (150 mM NaCl, 25 mM Tris pH 7.5, 1% Triton X-100, 0.1% SDS, 0.5% deoxycholate) supplemented with protease inhibitor tablet and PMSF. DNA was fragmented into ∼200 bp pieces using a Branson 250 sonicator. Aliquots of lysates containing 200 μg of protein were used for each immunoprecipitation reaction with anti-BRG1 (5 μg/reaction, Santa Cruz, sc-10768) or pre-immune IgG. Precipitated genomic DNA was amplified by real-time PCR. A total of 10% of the starting material is also included as the input. Data are then normalized to the input and expressed as% recovery relative the input as previously described (Chen et al., 2020b,c). All experiments were performed in triplicate wells and repeated three times.

Dihydroethidium (DHE) staining was performed essentially as previously described (Coarfa et al., 2020; Maity et al., 2020; Wang J. N. et al., 2020; Wang S. et al., 2020; Marti et al., 2021). Primary hepatocytes were stained with DHE (10 μM) at 37°C for 30 min. Fluorescence was visualized by con-focal microscopy (LSM 710, Zeiss) using an excitation wavelength between 480 and 520 nm and an emission wavelength between 570 and 600 nm. Quantifications were performed with Image J.

Quantitative measurements of intracellular ROS were performed with a ROS-Glo system (G8820, Promega). Briefly, a luminescence substrate solution was added to and incubated with cultured cells for 6 h followed by the addition of the diction solution. Luminescence was measured using a GloMax microplate reader (GM3000, Promega). Data were expressed as relative ROS levels compared to the control group.

Sample sizes reflected the minimal number needed for statistical significance based on power analysis and prior experience. Sample size for all animal experiments was ≥4 while all in vitro experiments were repeated at least three times with a power analysis showing β > 80. Data are expressed as mean ± standard deviation (SD). For comparison between two groups, two-tailed, unpaired Student’s t-test was performed. For comparison between more than two groups, one-way ANOVA with post hoc Scheffe analyses were performed using an SPSS package. Unless otherwise specified, p values smaller than 0.05 were considered statistically significant.

The effect of BRG1 deficiency on TBK1 expression was examined by the following experiments. Liver conditional BRG1 knockout (BRG1^LKO) mice and wild type (WT) littermates were fed a high-fat high carbohydrate (HFHC) diet for 16 weeks to induce non-alcoholic fatty liver disease (Figure 1A). As shown in Figure 1B, TBK1 mRNA expression was significantly up-regulated in the HFHC-fed livers compared to the chow diet-fed livers; BRG1 deletion, however, dampened in the induction of TBK1 expression by 30%. Western blotting showed that TBK1 protein levels were similarly down-regulated in the HFHC-fed LKO livers compared to the HFHC-fed WT livers (Figure 1C). Of note, TBK1 expression, at both mRNA (Figure 1B) and protein (data not shown) levels, was comparable between the chow-fed WT livers and the chow-fed LKO livers. In the second animal model, the BRG1^LKO mice and the WT mice were fed a methionine-and-choline deficient (MCD) diet for 8 weeks (Figure 1D). TBK1 mRNA expression was up-regulated by 1.88x fold in the WT livers but only 1.33x fold in the LKO livers equivalent to 29% suppression (Figure 1E). Similar observations were made with regard to the TBK1 protein expression (Figure 1F). Primary hepatocytes were isolated from WT and LKO mice and treated with palmitate (PA). PA treatment induced TBK1 expression more strongly in the WT cells than in the LKO cells (Figures 1G,H). Together, these data suggest that BRG1 may regulate TBK1 expression in the context of NAFLD.

The next series of experiments were performed to evaluate the mechanism whereby BRG1 may regulate TBK1 expression. A TBK1 promoter-luciferase construct was transfected into primary hepatocytes isolated from WT and LKO mice. As shown in Figure 2A, PA treatment augmented the TBK1 promoter activity in the WT hepatocytes but barely so in the LKO hepatocytes, suggesting that BRG1 may directly regulate TBK1 transcription. Next, chromatin immunoprecipitation (ChIP) was performed to determine the region of the TBK1 promoter BRG1 may bind to. Precipitated DNA was amplified by four pairs of primers spanning the ∼1.2 kb region of the proximal TBK1 promoter: primers #1 amplified a region containing a putative E74-like factor (ELF) motif, primers #2 amplified a region containing a p53 motif and a cKrox motif, primers 3 amplified a region containing an activator protein 1 (AP-1) site, and primers #4 amplified a region with an NF-κB site; PA treatment induced strong binding of BRG1 to region #3 but not other regions of the TBK1 promoter (Figure 2B). Similarly, HFHC diet feeding (Figure 2C) or MCD diet feeding (Figure 2D) enhanced the association of BRG1 with region #3 of the TBK1 promoter in the murine livers. In order to verify whether or not AP-1 might indeed be necessary for recruiting BRG1 to the TBK1 promoter, endogenous AP-1, a heterodimer of c-Jun and c-Fos, was depleted with siRNAs (Figure 2E). AP-1 knockdown significantly dampened the binding of BRG1 to the TBK1 promoter (Figure 2F). Furthermore, mutation of the AP-1 site on the TBK1 promoter abrogated the induction of promoter activity by PA treatment (Figure 2G). Collectively, these data suggest that BRG1, possibly by interacting with AP-1, may directly activate TBK1 transcription.

Western blotting data showed that BRG1 deficiency, in addition to causing a decrease in total TBK1 expression, also led to suppression of TBK1 activity as indicated by reduced serine 172 (S172) phosphorylation (Figures 1C,F,H). Of note, tyrosine 179 (Y179) phosphorylation of TBK1 was similarly suppressed by BRG1 deficiency (Figure 3A). Li et al. (2017) have previously reported that c-SRC mediated Y179 phosphorylation of TBK1 serves as a prerequisite for S172 phosphorylation and hence TBK1 activation. Of interest, over-expression of c-SRC by adenovirus transduction normalized both Y179 phosphorylation and S172 phosphorylation of TBK1 in the BRG1-null hepatocytes (Figure 3A). We therefore hypothesized that BRG1 could potentially regulate SRC expression to influence TBK1 phosphorylation. QPCR (Figure 3B) and Western blotting (Figure 3C) showed that BRG1 deficiency resulted in a reduction of SRC expression in primary hepatocytes. In addition, BRG1 deletion in hepatocytes down-regulated SRC expression in the livers in the HFHC model (Figures 3D,E) and in the MCD model (Figures 3F,G).

The next set of experiments was designed to determine whether BRG1 could directly regulate SRC transcription. When a SRC promoter-luciferase construct was transfected into primary hepatocytes from WT and LKO mice, it was found that PA treatment led to more prominent induction of the SRC promoter in the WT hepatocytes than in the LKO hepatocytes (Figure 4A). Next, ChIP assay was performed in primary hepatocytes with or without PA treatment to pinpoint the region of the SRC promoter to which BRG1 might bind. As shown in Figure 4B, the proximal SRC promoter was divided into three regions and amplified by three sets of primers: primers #1 amplified the region containing a putative E26 transformation-specific (ETS) motif, primers #2 amplified the region with an interferon regulatory factor (IRF) motif and a CCAAT-enhancer binding protein (C/EBP) motif, whereas primers #3 amplified the region with a putative specificity protein 1 (Sp1) motif; PA treatment elicited strong binding of BRG1 to region #3 of the SRC promoter. ChIP assays performed with liver lysates from the HFHC diet-fed mice (Figure 4C) and the MCD diet-fed mice (Figure 4D) confirmed that BRG1 could be recruited to the SRC promoter region with an Sp1 motif. In order to authenticate the essentiality of Sp1 in BRG1 recruitment, endogenous Sp1 expression was depleted with siRNAs (Figure 4E). Sp1 depletion weakened significantly the binding of BRG1 to the SRC promoter (Figure 4F). Finally, mutation of this putative Sp1 motif within the SRC promoter canceled the induction of the promoter activity by PA treatment (Figure 4G).

Finally, attempts were made to correlate the observed regulation of TBK1 expression and activity by BRG1 to a functional readout relevant in the pathogenesis of non-alcoholic fatty liver disease. DHE staining showed that re-introduction of TBK1 or c-SRC via adenoviral delivery of ectopic expression vectors largely normalized ROS production in the BRG1-null hepatocytes bringing it to a level closer to the WT hepatocytes (Figure 5A). The restoration of ROS production by exogenous TBK1 or c-SRC over-expression in BRG1-null hepatocytes was confirmed by luminescence measurements of ROS (Figure 5B). On the contrary, knockdown of either TBK1 or c-SRC (Figure 5C) mitigated induction of ROS production by PA treatment in WT hepatocytes as measured by DHE staining (Figure 5D) and luminescence assay (Figure 5E).