Berberine Inhibits Nod-Like Receptor Family Pyrin Domain Containing 3 Inflammasome Activation and Pyroptosis in Nonalcoholic Steatohepatitis via the ROS/TXNIP Axis

Berberine (BBR), an isoquinoline alkaloid originating from herbal plants, has been deemed beneficial for non-alcoholic fatty liver disease. Increasing evidence has demonstrated that Nod-like receptor family pyrin domain containing 3 (NLRP3) inflammasome activation and the subsequent pyroptosis contribute to the progression of non-alcoholic steatohepatitis (NASH). However, whether BBR impacts NLRP3 inflammasome activation and pyroptosis in NASH and the potential mechanism remains unclear. In the current study, we found that BBR significantly decreased lipid accumulation, ameliorated reactive oxygen species (ROS) and lipid peroxides, Tumor necrosis factor alpha (TNF-α) expression, and phosphorylation of Nuclear factor kappa B (NF-κB) p65 both in vivo and in vitro. In particular, BBR significantly inhibited NLRP3 expression, caspase-1 activity, and the pyroptosis executor, GSDMD-N, expression. In addition, BBR displayed similar inhibitory effects on NLRP3 inflammasome and pyroptosis with a decrease in ROS levels and TXNIP expression as N-acetyl-cysteine, a ROS scavenger, did. Whereas, the inhibitory effect of BBR on ROS, TXNIP expression, NLRP3 inflammasome activation and pyroptosis could be reversed by H2O2 in AML12 cells. This study demonstrates that BBR's inhibitory effect on NLRP3 inflammasome activation and pyroptosis may be mediated by ROS/TXNIP axis in vitro for the first time. Our findings suggest BBR is a potential candidate for the treatment of NASH.


INTRODUCTION
Nonalcoholic fatty liver disease (NAFLD) includes two pathologically distinct conditions with different prognoses: non-alcoholic fatty liver (NAFL) and non-alcoholic steatohepatitis (NASH), which can eventually progress to endstage liver diseases such as cirrhosis and hepatocellular carcinoma (Wong et al., 2014;Wong et al., 2016). Although the transition from NAFL to NASH has been considered the critical step in NAFLD development, understanding the mechanism and being able to clinically manage the disease require further investigations. Besides lifestyle changes, medications such as metformin, vitamin E, and Pioglitazone, are currently used to treat NAFLD in the clinic, but they often exert minor efficacy and considerable side effects (Wong et al., 2016;Sumida and Yoneda, 2018). Owing to these shortcomings, researchers are now focused on developing new therapeutics for NAFLD.
NLRP3 inflammasome complex, implicated in the progression of NAFLD from NAFL to NASH, contains NOD-like receptor pyrin domain containing 3 (NLRP3), a speck-like protein and damageassociated adaptor molecule with a caspase recruitment domain (ASC) and procaspase-1 as its effector molecule (He K. et al., 2017). NLRP3 inflammasome can be activated by many stimuli such as viruses, bacteria, free fatty acids (FFA), or reactive oxygen species (ROS). Once the complex is formed, inflammasomes activate caspase-1, which proteolytically activates the pro-inflammatory cytokines, interleukin-1b (IL-1b), and IL-18. Inflammasome activation also results in a rapid, pro-inflammatory form of cell death called pyroptosis (Latz et al., 2013).
Pyroptosis is a new form of programmed cell death resulting in cell swelling, membrane rupture, pore formation in the plasma membrane, extensive leakage of cytosolic contents, and high degree of inflammation (Shi et al., 2015;Shi et al., 2017). Pyroptosis is suggested to occur in cells (e.g., hepatocytes) where cytokine secretion is deemed infeasible following inflammasome-mediated caspase-1 activation (Miao et al., 2011). Gasdermin D (GSDMD), a major pyroptosis executor, exerts its function by releasing the N domain (GSDMD-N) cleaved by caspase 1/4/5/11. Recent studies have reported that GSDMD-mediated pyroptosis acts as an inflammatory link between NAFLD and NASH by controlling cytokine secretion, NF-ĸB activation, and lipogenesis (Shi et al., 2017;Xu et al., 2017).
Berberine (BBR), an isoquinoline alkaloid that originates from numerous herbal plants including Coptis chinensis, is used as a medicine for diarrhea and gut infections (Zhu et al., 2016). Recently, many other potential therapeutic effects of BBR for various diseases including central nervous system disorders, diabetes, cancer, cardiovascular disease, depression, hypertension, hypercholesterolemia, etc. have been explored (Kulkarni and Dhir, 2010;Vuddanda et al., 2010;Huang et al., 2018). BBR also displays a potent beneficial effect in NAFLD by improving insulin resistance via multiple processes, reducing lipid accumulation via regulating AMPK signaling, improving mitochondrial function, alleviating oxidative stress, reducing serum cholesterol, and regulating gut microenvironment. Nonetheless, whether BBR impacts NLRP3 inflammasome activation, the subsequent pyroptosis in NASH, and the potential mechanism whereby these occur, remain unclear. Considering the importance of NLRP3 inflammasome and pyroptosis in NASH, the above activities require exploration. Therefore, we examined the protective effects of BBR, particularly on NLRP3 inflammasome activation and pyroptosis, and explored the potential mechanism by which these occur. Our results showed that BBR inhibits NLRP3 inflammasome activation and the subsequent pyroptosis, and may be mediated by ROS/TXNIP axis in AML12 cells treated with methionine-choline deficient (MCD) medium plus lipopolysaccharide (LPS) or palmitic acid (PA).

Cell Culture and Treatment
Immortalized mouse normal hepatocytes, AML12 cells were obtained from Stem Cell Bank, Chinese Academy of Sciences. AML12 cells were maintained at 37°C in DMEM/F-12 medium supplemented with 10% FBS, 1x ITS; and 40 ng/ml dexamethasone. At 70% confluence after serum starving for 12 h, cultured cells were treated with MCD medium for 24 h followed by a 12 h treatment with 1 mg/ml LPS or 1 mM PA for 24 h to establish the NASH cellular model and non-treated cells were used as control.
To assess the effect of BBR on MCD/LPS or PA-induced hepatocyte injury, two concentrations of BBR (10 or 20 mM) were added to the cells in combination with MCD/LPS or PA for 24 h. The selection of BBR concentration is based on our pilot experiment ( Figure S1) and other published study (Guo et al., 2016). To assess the role of oxidative stress in BBRmediated inhibition of NLRP3 inflammasome and pyroptosis in vitro, 10 mM NAC, a ROS scavenger, was applied in combination with MCD/LPS or PA for 24 h, and 200 nM H 2 O 2 , a ROS inducer, was applied in combination with BBR and MCD/LPS or PA for 6 h.

Animal Experiments
Fifteen 7-week old male C57BLKS/J mice were obtained from Experimental Animal Center of Southern Medical University (Guangzhou, China). All mice were maintained on a 12/12 h light/dark cycle with free access to food and water. The mice were divided to the following three groups (5 mice per group): (i) control group, received a methionine and choline sufficient (MCS) diet (Dyets, USA, #519581) for 5 weeks and treated with PBS via oral gavages for the last 2 weeks; (ii) MCD group, fed a methionine and choline deficient (MCD) diet (Dyets, USA, #519580) for 5 weeks and treated with PBS via oral gavages for the last 2 weeks; and (iii) MCD + BBR group, received MCD diet for 5 weeks and treated with BBR (100 mg/ kg body weight/d, suspension in PBS) via oral gavages for the last 2 weeks. At the age of 12 weeks, serum samples were collected prior to the sacrifice of mice and liver was harvested for further studies.
This study was approved by the ethical committee of Guangzhou Medical University, China. All the animal experiments complied with the standard ethical guidelines prescribed by the ethical committees mentioned above.

Nile Red Staining
Intracellular lipids were stained with Nile Red (Sigma-Aldrich, cat. no. N3013) for 15 min after fixing with 4% paraformaldehyde for 30 min, and the nuclei stained with DAPI for an additional 5 min. Images were photographed with an inverted fluorescence microscope (EVOS FL, Thermo Fisher Scientific Inc., USA).

Measurement of Intracellular ROS Level
For the quantitative analysis of ROS production, cells were incubated with DCFH-DA fluorescence probe for 30 min using the manufacturer's manual. Fluorescence intensity was measured by flow cytometry (FACS Calibur, BD Biosciences) and analyzed using FlowJo software version 10.0 (Treestar, Inc., Palo Alto, CA, USA).

Measurement of Malondialdehyde
Measurement of malondialdehyde (MDA) levels in AML12 cells was performed using a Lipid Peroxidation MDA Assay Kit. MDA is an end-product of fatty acid peroxidation and can be measured via chromogenic reaction between MDA and Thiobarbituric acid (TBA). Briefly, 100 ml of cell lysate was mixed with 200 ml of MDA working solution followed by incubation in a 100°C water bath for 15 min, then cooled to 25°C. The mixture was centrifuged at 1,000 × g for 10 min, and 200 ml supernatant was used to measure absorbance at 532 nm. TBARS concentration was calculated from an MDA standard curve.

Quantitative PCR
Total RNA from cultured cells was isolated using Trizol and treated with DNase I. In brief, quantitative real-time PCR was carried out to determine lncRNA expression using the Prime Script RT Reagent Kit and SYBR Premix Ex Taq. Real-time PCR was performed on ABI Prism 7300 real-time PCR system (Applied Biosystems, Foster City, CA, USA). Quantification of target gene expression was performed using the relative quantification comparative CT method. The primer sequences used for real-time PCR include: TNF-a, 5'-GCCACCACG CTCTTCTGTCTAC-3' (forward) and 5'-GGGTCTGGG CCATAGAACTGAT-3' (reverse); b-actin: 5'-ATGCCACAGG ATTCCATACCCAAGA-3' (forward) and 5'-CTCTAG ACTTCGAGCAGGAGATGG-3' (reverse).

Assaying Caspase-1 Activity
Caspase-1 activity was detected using a caspase-1 activity assay kit according to the manufacturer's instructions. This assay was based on the ability of caspase-1 to convert acetyl-Tyr-Val-Ala-Asp p-nitroaniline (Ac-YVAD-pNA) to the yellow formazan product, p-nitroaniline (pNA). Thus, the activity of caspase-1 can be assessed by measuring the absorbance of pNA using a standard pNA curve. Fifty mg of total cytosolic protein was incubated in a 96-well microtiter plate with 20 nmol Ac-YVAD-pNA overnight at 37°C. The absorbance values at 405 nm were measured using the spectrophotometer, Varioscan Flash, and SkanIt Software Version 2.4.3. RE (Thermo Fisher Scientific Inc, USA). The production of pNA in the tested samples indicated the level of caspase-1 activation.

Measurement of ALT and AST
After centrifuged at 6,000 rpm for 10 min, serum levels of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) were then measured using Olympus AU5400 clinical biochemical analyzer (Tokyo, Japan).
The intensity of each band was quantified using ImageJ software and normalized to b-actin and adjusted relative to the MCD/LPS or PA treated group.

Statistical Analysis
Results are presented as means ± standard deviation. Data were analyzed using the Student's t-test (for two groups) or one-way ANOVA and Fish's Least Significant Difference (LSD) test (for more than two groups). P values < 0.05 were considered statistically significant. All analyses were performed using the Statistical Package for Social Sciences, version 20.0, for Windows (SPSS, Inc., Chicago, IL, USA).

Berberine Decreases PA or MCD Medium-Induced Lipid Accumulation in AML12 Cells
As lipid accumulation is a key feature of NAFLD, we established two NASH in vitro models to mimic hepatic steatosis and injury in NASH. AML12 cells were exposed to a 24-h MCD-medium treatment plus a 12 h LPS (1 mM) treatment or 24 h PA (1 mM) treatment and non-treated cells were used as control. Evidently, there was lipid accumulation in the MCD/LPS or PA treatment groups compared to the non-treated group by Nile red staining. We then explored the effect of BBR on lipid accumulation in AML12 cells by administering two doses (10 and 20 mM) in the presence or absence of MCD/LPS or PA. Both doses significantly and dose-dependently decreased lipid accumulation in AML12 cells in the presence of MCD medium/LPS or PA, as shown by Nile red staining and flow cytometric analysis (Figure 1).
Oxidative Stress in PA or MCD-Treated AML12 Cells is Ameliorated by BBR As oxidative stress contributes to the initiation and progression of liver injury in NAFLD, we determined the antioxidative effect of BBR by assessing intracellular ROS using a DCF assay. Figure 2A

Berberine Inhibits NFkB p65 Phosphorylation and TNFa Expression in AML12 Cells
TNF-a and NF-kB signaling play pivotal roles in the development and progression of NAFLD. Phosphorylation of p65 at Ser 536 leads to nuclear localization of the transcriptionally active complex and NF-kB mediated transactivation of several downstream genes. Thus, we examined the effect of BBR on TNF-a expression and NF-kB signaling by qRT-PCR and western blot, respectively. As shown in Figure 3A, qRT-PCR results revealed that TNF-a mRNA expression was significantly increased in the MCD/LPS group compared to the control group, and MCD/LPS-induced elevation was downregulated due to BBR in AML12 cells. Western blot showed that both TNF-a and phosphorylation of NF-kB p65 were increased by MCD/LPS treatment compared to the controls; these were inhibited by BBR, particularly the 20 mM dose ( Figure 3C). Consistent with the results obtained in MCD/ LPS-treated AML12 cells, mRNA and protein expression of TNF-a, as well as the phosphorylation level of NF-kB p65 were lower in the BBR treatment group compared to the PA treatment group (Figures 3B, D).

Berberine Suppresses NLRP3 Inflammasome Activation and Pyroptosis
In Vitro NLRP3 inflammasome activation may lead to GSDMD-driven pyroptosis, which plays a key role in the pathogenesis of NASH (Wree et al., 2014a). We aimed to determine whether BBR could attenuate NLRP3 inflammasome activation and the subsequent pyroptosis. As shown in Figure 4, we found that both MCD/LPS and PA treatment significantly increased NLRP3, pro-caspase 1 and caspase1 protein levels, and caspase-1 activity, indicating NLRP3 inflammasome activation. Moreover, GSDMD and its cleaved N domain (GSDMD-N), the pyroptosis executor, were also increased in the MCD/LPS and PA treatment groups. BBR treatment, however, dose-dependently ameliorated MCD/LPS or PA-induced activation of NLRP3 inflammasome and pyroptosis as shown by the decreased expression of NLRP3, pro-caspase 1 and caspase1, GSDMD-N protein, and capspase-1 activity. As the 20 mM concentration of BBR displayed more obvious protective effects on NASH, this was selected for use in the subsequent experiments.

Berberine Suppresses NLRP3 Inflammasome Activation and Pyroptosis via the ROS/TXNIP Axis
It has been reported that ROS can trigger TXNIP to dissociate from TRX and bind to NLRP3, resulting in NLRP3 activation (Zhou et al., 2010). Based on the above findings, we speculated that BBR may inhibit NLRP3 inflammasome and pyroptosis in NASH via the ROX-TXNIP axis. To explore the potential underlying mechanism, NAC, a ROS scavenger, was introduced into the experiment and the inhibitory effect of BBR and NAC on oxidative stress, NLRP3 inflammasome activation, cellular pyroptosis, and lipid accumulation was compared. As shown in Figure 5, BBR (20 mM) and NAC (10 mM) decreased the cellular ROS levels in MCD/LPS or PAtreated AML12 cells to a similar degree. As expected, TXNIP, NLRP3, pro-caspase 1, caspase1, and GSDMD-N protein levels and caspase-1 activity were suppressed by NAC, a result similar to BBR. However, as shown by Nile red staining, NAC failed to alleviate MCD/LPS or PA-induced lipid accumulation as BBR.
These results indicate that the ROS-TXNIP axis may be involved in BBR's inhibitory effect on NLRP3 inflammasome and the subsequent cellular pyroptosis. To further confirm these findings, H 2 O 2 , a ROS inducer, was administered to the BBR treatment group by exposing the AML12 cells to H 2 O 2 for 6 h. We observed that there was no lowering in cellular ROS levels in this group compared to the MCD/LPS or PA group, indicating the abolition of the anti-oxidative effect of BBR ( Figures 6A, D). Western blot showed that the decreases in TXNIP, NLRP3, pro-caspase 1, caspase1, GSDMD-N protein levels, and caspase-1 activity   caused by BBR in MCD/LPS or PA-treated AML12 cells were reversed, and the activation of NLRP3 inflammasome and pyroptosis was rescued to varying degrees by H 2 O 2 ( Figure  6). While, Nile red staining showed that H 2 O 2 administration fail to reverse the BBR-induced amelioration of lipid accumulation, indicating that ROS may not involve in BBR-mediated improvement of lipid accumulation in NASH ( Figure 6G). Taken together, these data reveal that the inhibitory effect of BBR on NLRP3 inflammasome and the subsequent cellular pyroptosis in MCD/LPS or PAtreated AML12 cells may be mediated by the ROS/ TXNIP axis.

Berberine Ameliorated Steatohepatitis in MCD-Fed Mice
To validate the findings in AML12 cells, we examined the protective effect of BBR in MCD-induced steatohepatitis mice. As shown in Figure 7, upon feeding MCD diet for 5 weeks, lipid droplets and inflammatory foci within the lobule in MCD-fed mice were observed by HE staining ( Figure 7A). Oil red O staining of the liver sections further confirmed the fat accumulation in MCD-fed mice ( Figure 7B). Liver injury was occurred as reflected in the elevation of ALT and AST ( Figure  7D). These results indicated steatohepatitis was developed in MCD-fed mice. However, 2-week BBR treatment significantly ameliorated hepatic steatosis and liver injury induced by MCD diet. And the NAS system also revealed that the liver histology in the BBR-administered group was improved ( Figure 1E). Furthermore, Western blot showed that the level of TXNIP, NLRP3 inflammasome, pyroptosis, phosphor p65, and TNFa was increased in the liver of MCD group. Whereas 2-week BBR treatment to MCD-fed mice significantly decreased the expression of TXNIP, NLRP3, caspase-1, GSDMD-N, phosphor p65, and TNFa ( Figures 7C, F), which was consistent with the findings in AML12 cells. Taken together, our in vivo study indicated that BBR may ameliorated steatohepatitis in MCD-fed mice.

DISCUSSION
Although many researched have been performed to explore the therapeutic effect and mechanism of BBR in diseases, its effect on NLRP3 inflammasome-mediated pyroptosis in NASH and the mechanism whereby this occurs, remain unknown. Through our work, we discovered that BBR not only reduces lipid accumulation, alleviates oxidative stress, and inhibits NK-kB signaling and TNF-a expression, but significantly inhibits NLRP3 inflammasome activation and the subsequent pyroptosis. Mechanistically, we further demonstrated that its inhibitory effect on NLRP3 FIGURE 6 | The inhibitory effect of berberine (BBR) on nod-like receptor family pyrin domain containing 3 (NLRP3) inflammasome activation and pyroptosis was reversed by H 2 O 2 . AML12 cells were exposed to methionine-choline deficient (MCD)/lipopolysaccharide (LPS) or PA in the presence or absence of 20 mM BBR for 24 h. H 2 O 2 (200 mM) was added to cells in the final 6 h in the indicated group. (A, D) The proportion of cells producing reactive oxygen species (ROS) detected by flow cytometry using DCF-DA assay was compared to control group. (B, E) Caspase-1 activity was determined using a caspase-1 activity assay. (C, F) Western blot for TXNIP, NLRP3, pro-caspase 1, caspase 1, GSDMD and GSDMD-N in AML12 cells. (G) Representative images of Nile red staining of AML12 cells. Data were expressed as mean ± SD of three independent experiments. & p < 0.05 vs control group, *p < 0.05 vs MCD+LPS or PA group, # p < 0.05 vs MCD+LPS+BBR or PA +BBR group. MCD, Methionine-choline deficient; LPS, lipopolysaccharide; PA, palmitic acid; ROS, reactive oxygen species; GSDMD, gasdermin D.
In the current study, we established two distinct-mechanical NASH in vitro models by exposing AML12 cells to PA or MCD medium plus LPS and one in vivo model by feeding MCD diet to C57BL/6J mouse for 5 weeks. PA, accounting for 20-30% of total fatty acids in the human body (Carta et al., 2017) and the cause of lipotoxicity in hepatocytes, is widely used to establish NASH in vitro models (Ji et al., 2018). Deprivation of methionine and choline is known to cause many disturbances in hepatic lipid metabolism due to the decreased availability of methionine and choline for phospholipid synthesis (Rinella et al., 2008;Pickens et al., 2009). Furthermore, LPS, which plays a key role in the pathogenesis of NASH as one of the multi-hits, can cause inflammation in cells. Our results showed that significant lipid accumulation, oxidative stress, and activation of inflammation- associated signaling were observed in the in vitro models, and histological features such as lipid accumulation, inflammatory infiltration, and liver injury were occurred in the in vivo model, facilitating the investigation performed to determine the effect and molecular mechanism of BBR on NASH.
Recently, increasing evidence have suggested a critical role of GSDMD-mediated pyroptosis in liver diseases (Kofahi et al., 2016;Beier and Banales, 2018;Khanova et al., 2018). Xu et al. demonstrated that GSDMD-N correlates with NAS and fibrosis in NASH patients. Mechanically, GSDMD-N promoted NASH development by mediating hepatic cytokine secretion, macrophage infiltration, NF-ĸB signaling pathway, and lipogenic gene expression . In the present study, we revealed that BBR can reduce GSDMD-N both in vitro and in vivo. Furthermore, to the best of our knowledge, this is the first study to reveal this discovery. The roles of BBR on pyroptosis may vary in different diseases. Li et al. (2017) discovered that BBR significantly increased ATP-induced pyroptosis and increased the release of caspase-1, p10, and IL-1b in macrophages. A research in 2016 (Chu et al., 2016) revealed that BBR could induce pyroptosis in HepG2 cells via activating caspase-1.
canonical caspase-1, cleaved GSDMD and played a pivotal pathogenic role (Khanova et al., 2018). In our study, however, elevation of GSDMD-N was mainly caused by the activation of NLRP3-ASC-caspase-1 inflammasome in the NASH cellular model.
It is well known that activation of the NLRP3 inflammasome in the liver promotes the progression of NASH (Csak et al., 2011). During the development of experimental and clinical NAFLD, the expression of hepatic NLRP3 increases (Csak et al., 2014;Wree et al., 2014a;Wree et al., 2014b;Szabo and Petrasek, 2015). Thus, the development of fatty liver disease in mice deficient in NLRP3 (Nlrp3−/−) or its essential components (Asc−/− and Casp1−/− mice) can be avoided if they consume a high fat or nutrient-deficient diet (Dixon et al., 2012;Wree et al., 2014b). Mridha et al. reported that a p h a r m a c o l o g i c b l o c k a g e o f N L R P 3 r e d u c e s l i v e r inflammation and fibrosis in experimental NASH in mice (Mridha et al., 2017), indicating NLRP3 is a potential target of NASH treatment.
Increasing studies have reported that BBR exhibits an inhibitory effect on NLRP3 inflammasome and is a potential candidate for the treatment of various diseases such as atherosclerosis , gouty arthritis (Dinesh and Rasool, 2017), and liver injury (Vivoli et al., 2016). A previous study found that BBR inhibits palmitate-induced NLRP3 inflammasome activation by triggering autophagy in macrophages (Zhou et al., 2017), and in our study, besides macrophages, BBR inhibited NLRP3 inflammasome activation as shown by the decrease in NLRP3 protein levels and caspase-1 activity in MCD/LPS or PA-treated hepatocytes and MCD-fed mice liver.
NLRP3 inflammasome can be activated by many DAMPs and PAMPs, including FFA, LPS, and ROS (Abais et al., 2015); however, the mechanism used by BBR to suppress MCD/LPS or PA-induced NLRP3 inflammasome activation was unclear. Following an increase in ROS, TXNIP dissociates from TRX and binds to NLRP3, leading to NLRP3 activation (Zhou et al., 2010). Owing to this, we hypothesized that BBR may suppress NLRP3 inflammasome activation and subsequent pyroptosis via the ROS-TXNIP pathway; this is because ROS was decreased by BBR. Our results revealed that NAC, a ROS scavenger, has a similar effect on NLRP3 inflammasome and pyroptosis to BBR as it suppresses the ROX-TXNIP pathway. Besides, a rescue experiment further confirmed that ROS-TXNIP axis played a key role in the inhibitory effect of BBR on NLRP3 inflammasome and pyroptosis in NASH.
Although two in vitro models and one in vivo model were employed in our study, more in vivo studies using different models are still needed to further confirm the protective effects of BBR on NLRP3 inflammasome and pyroptosis in NASH. MCD diet is widely employed in the NASH animal studies and can induce the histological features of steatohepatitis in human NASH. However, its metabolic context such as obesity and insulin resistance are quite different from human NASH (Ibrahim et al., 2016;Farrell et al., 2019). Since no current model can reproduce all major features of severe human NASH, it is recommended at least two different models such as fast food diet and CDAA/HFD/0.2% cholesterol should be used in the future studies, since no murine model depicts all features of human NASH (Farrell et al., 2019).
Several studies have explored the therapeutic effects of BBR on NASH in murine models (Qiang et al., 2016;He Q. et al., 2017;Zhao et al., 2017;Feng et al., 2018;Luo et al., 2019) and humans (Chang et al., 2016;Wei et al., 2016); Qiang et al. demonstrated that demethyleneberberine attenuated lipid accumulation and inflammatory factors secretion with activation of AMPK and inhibition of oxidative stress in MCD diet feeding mice and db/db mice (Qiang et al., 2016). Feng et al. compared the effects of BBR, curcumin, and lovastatin on NAFLD rats and found that BBR in combination with curcumin exhibited better ameliorative effects on rats with non-alcohol fatty liver than lovastatin (Feng et al., 2018). A meta-analysis included 6ix randomized controlled trials involving 501 patients revealed that BBR has positive efficacy on blood lipids, blood glucose, liver function, insulin resistance, and fatty liver condition of NAFLD patients (Wei et al., 2016). However, these studies were mainly focused on the effects and mechanisms of BBR in the regulation of lipid metabolism and hepatic injury in NAFLD. The roles of BBR in the regulation of NLRP3 inflammasome and pyroptosis in NASH in vivo has not been clarified, which needs to be determined in the future studies.
In conclusion, we have provided new molecular insights into the protective mechanism of BBR on NASH through its inhibition of NLRP3 inflammasome and the subsequent pyroptosis, in a ROS-dependent manner. Our findings suggest that BBR is a potential candidate for the treatment of NASH.

DATA AVAILABILITY STATEMENT
The datasets generated for this study are available on request to the corresponding authors.

ETHICS STATEMENT
The animal study was reviewed and approved by Animal Ethics Committee of the Second Affiliated Hospital of Guangzhou Medical University.

AUTHOR CONTRIBUTIONS
WM and DZ performed in vitro experiments and generated data. YX: study design, data analysis, and manuscript preparation. JX and LY performed in vivo experiments and generated data. GY, ZW, QL, and JL: data analysis and figure preparation. TY, CG, and SL: data analysis and interpretation. YZ: generated idea and study design. HY: generated idea, study design, data analysis, and manuscript writing.