Adiponectin inhibits palmitic acid-induced NLRP3 inflammasome activation in hepatocytes through AMPK-JNK/ErK1/2-NFκB/ROS signaling pathways CURRENT STATUS: POSTED

Adiponectin, an adipose-derived adipokine, possesses a hepatoprotective role in various liver disorders. Inflammasome activation has been recognized to play a major role during the progression of non-alcoholic fatty liver diseases (NAFLD). However, the effect of adiponectin on NLRP3 inflammasome activation in liver and the exact mechanism remains largely unclear. Here, we assessed the effect of adiponectin on NLRP3 inflammasome activation and its potential molecular mechanisms through both in vivo and in vitro experiments. Male adiponectin-knockout (adiponectin-KO) mice and C57BL/6 (wild-type) mice were fed a 12 as an vivo Serum biochemical markers, liver and inflammasome-related gene and protein expression were determined. In addition, the hepatocytes isolated from rats were exposed to palmitic acid(PA) in the absence or presence of adiponectin and/or AMPK inhibitor. The activation of NLRP3 inflammasome was assessed by mRNA and protein expression. Furthermore, ROS production and related signaling pathways were also evaluated. The results indicated that the inhibitory effect of adiponectin on PA-mediated NLRP3 inflammasome activation was regulated by AMPK-JNK/ErK1/2-NFκB/ROS signaling with PA induced lipid droplet deposition and NLRP3 inflammasome expression in hepatocytes. To verify the roles of adiponectin in PA-mediated hepatic activation of NLRP3 inflammasome complexes, hepatocytes were pre-treated with adiponectin for 2 h and then exposed to PA for 24 h. It was found that PA induced the expression levels of NLRP3, caspase1, ASC, IL1β, IL18 and other inflammatory markers (e.g., TNFα and IL6) when compared to control group. However, treatment with adiponectin could alleviate PA-mediated NLRP3 inflammasome activation (Fig. 3B-H). These findings reveal that adiponectin exerts a protective effect on PA-stimulated NLRP3 inflammasome activation in hepatocytes.


Conclusion
Adiponectin inhibited PA-mediated NLRP3 inflammasome activation in hepatocytes. Adiponectin analogs or AMPK agonists could serve as a potential novel agent for preventing or delaying the progression of NASH and NAFLD. Background: Non-alcoholic fatty liver diseases (NAFLD) contains a broad histopathological spectrum ranging from steatosis alone to non-alcoholic steatohepatitis (NASH), which can result in hepatic cirrhosis and even liver cancer [1,2]. NASH is highly associated with obesity, insulin resistance, hypertension and diabetes mellitus [3]. In recent years, with the changes in human diet and lifestyle, the incidence rates of NAFLD and NASH are increased at an alarming rate, which has become the leading cause of chronic hepatic diseases [4,5]. At present, there is no effective therapeutic drug for NAFLD treatment. Thus, it is of great significance to understand the pathogenesis of NAFLD. The "two-hit" hypothesis on NASH development has been widely spread and well recognized [6]. Recently, new insights into the molecular mechanism of NASH have been provided. Numerous studies have suggested the involvement of inflammasome activation in the pathogenesis of NASH [7][8][9][10].
Inflammasomes are large cytoplasmic multi-protein complexes that response to both exogenous and endogenous alerting signals through intracellular NOD-like receptor (NLR) [11,12]. NLR family pyrin domain-containing 3 (NLRP3) inflammasome, composing of NLRP3, the adaptor protein ASC and the effector molecule pro-caspase1, is well characterized [13,14]. The activation of NLRP3 inflammasome usually requires two signals [15]: the prime signal that promotes the transcription of inflammasome components through NFκB signaling activation [16], and the second signal is initiated to activate NLRP3 for recruiting and interacting with caspase1 precursor (pro-caspase1) through the adaptor molecule ASC [17]. Pro-caspase1 is cleaved into active caspase1, which in turn promotes the secretion of mature IL-18 and IL-1β and triggers inflammatory responses.
Adiponectin is an adipose-derived adipokine [18]. Much attention has been attracted by adiponectin because of its insulin-sensitizing [19], anti-inflammatory and hepatoprotective properties [20,21]. 4 Previous studies have demonstrated that the plasma level of adiponectin is negatively correlated with NAFLD [18], and hypoadiponectinemia is independently associated with hepatic steatosis and inflammation in NASH patients [22]. Adiponectin deficiency can accelerate the progression of steatohepatitis in NASH mouse model and induce severe liver fibrosis [20,23]. However, the potential mechanism underlying the hepatoprotective effect of adiponectin has not been has yet to be fully elucidated. Therefore, this research aimed to explore the effects of adiponectin on palmitic acid (PA)mediated NLRP3 inflammasome activation in hepatocytes and its potential molecular mechanisms.

Experimental animal model
The ethical approval for this study was obtained from the Animal Ethics Committee of our institution.
All animals were maintained and used in accordance with the guidelines and policies approved by the Animal Care and Use Committee of Shanghai Jiao Tong University School of Medicine. To induce the animal model of non-alcoholic steatohepatitis, 4-week-old male C57BL/6 and adiponectin-knockout (adiponectin-KO) mice were randomly assigned to two groups: normal diet feeding group and high-fatdiet (HFD) feeding (D12492, Research Diets) group. All animals were given unlimited access to water and food, and kept under a controlled temperature of 22-24 °C with a 12-h light/12-h dark cycle. The mice were then euthanized, and their liver tissue and plasma samples were collected at 0, 4, 8 and 12 weeks for further analyses.

Cell Isolation And Treatment
Hepatocytes were isolated from male C57BL/6 mice by a two-step (collagenase B and pronase E) perfusion method under ketamine/xylazine anesthesia as described previously [24]. The isolated hepatocytes were seeded in collagen-coated culture dishes at a density of 2*10^5 cells/ml and cultured in Dulbecco's modified Eagle's medium (HyClone, Logan, UT, USA) supplemented with 10% fetal bovine serum (Gibco, Carlsbad, CA, USA) and 1% (v/v) penicillin-streptomycin at 37 °C in 5% CO2 incubator for 48 h. The hepatocytes were then serum-starved for 6 h, followed by treatment with adiponectin (APN; 10 ug/ml, Biovendor, Nycodenz) for 2 h prior to 300 umol/ml palmitic acid (PA; Sigma, St. Louis, MO) exposure for 24 h. For the inhibition experiment, the hepatocytes were treated 5 with AMPK inhibitor (compound C, 10 um/M; Sigma) 2 h prior to adiponectin treatment. The treated hepatocytes and culture supernatant were collected for subsequent analyses.

Real Time PCR Analysis
Total RNA was extracted from the cultured hepatocytes or liver tissue by using TRIzol reagent (Invitorgen, Carlsbad, California, USA) based on the manufacturer's protocols. cDNA synthesis was performed using SuperScriptIII reverse transcriptase, random primers and 1 ug RNA (Invitrogen, Carlsbad, CA, USA). qPCR was conducted using Power SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA, USA). The expression level of each gene was normalized to the corresponding housekeeping gene β actin value, and presented as fold changes relative to controls.
All primers sequences are shown in Table 1. Equivalent amounts of total protein extracted from the hepatocytes or liver tissue were loaded onto 10% sodium dodecyl sulphate poly-acrylamide gel, and subsequently transferred onto polyvinylidene fluoride membranes (Millipore, Bedford, MA, USA). After blocking with 5% non-fat milk, the membranes were incubated overnight with primary antibodies at 4℃, followed by washing with 0.05% Tween-20/TBS. Following incubation with secondary antibodies, the resulting blots were visualized by an enhanced chemiluminescence kit (Pierce Perbio, Rochford, IL) based on the manufacturer's instructions. Finally, the protein bands were digitally scanned and quantitated using ImageJ software.

Immunofluorescence Staining
The expression levels of NLRP3 in hepatocytes exposure to PA were analyzed by immunofluorescence.
All images were obtained using a fluorescence microscope (IX51, Olympus, Japan).

Determination Of Reactive Oxygen Species
The level of reactive oxygen species (ROS) in hepatocytes was measured using 2',7'dichlorofluorescein diacetate (DCFH-DA, ROS probe; D6883, Sigma). All images were acquired using a fluorescence microscope (IX51, Olympus, Japan). The excitation and emission wavelengths were fixed at 488 and 525 nm, respectively.

Detection Of Lipid Droplet Deposition
Hepatocytes were treated as described above and then fixed with 10% formalin. Lipid droplet deposition in cells was detected with BODIPY 493/503 (790389, Sigma), and then observed using a fluorescence microscope (IX51, Olympus, Japan).

Biochemical Analysis And Cytokine Assay
The levels of aspartate aminotransferase (AST), alanine aminotransferase (ALT), cholesterol (CHOL) and total triglyceride (TG) were detected with the commercial kits (Nanjing Jiancheng Bioengineering Institute). Cytokines in the supernatant or liver tissue were assessed using the commercially available enzyme-linked immunoassay (ELISA) kits. The ELISA kits for IL1β, IL18, TNFα and IL-6 were obtained from Elabscience (Wuhan, China).

Histological Analysis Of Liver Tissue
The liver tissue sections were subjected to hematoxylin and eosin (H&E) staining, oil red O (ORO) staining and immunohistochemistry dye by following the standard protocols as previously described.
Briefly, the paraffin-embedded liver sections were deparaffinized, rehydrated. stained with H&E or 7 ORO, and then observed under a microscope. For immunohistochemical staining, the liver sections were incubated with freshly prepared 3% hydrogen peroxide for 20 min. After washing with PBS, antigen retrieval was conducted in 0.01 M citric acid. The sections were submerged in 5% normal blocking serum for 30 min, and subsequently incubated with NLRP3, caspase1 or IL1b overnight at 4 °C. Following that, the sections were incubated with the corresponding secondary antibody for 1 h at room temperature. Lastly, all sections were examined using the microscope.

Statistical analysis
All data were presented as mean ± SEM. Student's t test or two-way ANOVA was employed to compare the statistical differences using SPSS statistics software (version 12.0 for Windows; SPSS, Inc., Chicago, IL, USA). A p-value of < 0.05 was deemed as statistically significant. All experiments were repeated at least 3 times.

Results:
Adiponectin deficiency aggravates liver injury and steatosis as well as sensitizes to HFD-induced NLRP3 inflammasome activation To figure out the effects of high-fat-diet (HFD) feeding on hepatic injury and steatosis in adiponectin-KO (APNKO)mice, several biochemical markers and liver histology were analyzed. The results indicated that the serum levels of AST, ALT, CHOL and TG were markedly increased with the prolongation of HFD feeding time, and appeared to be higher in APNKO mice than in wild-type mice ( Fig. 1A). H&E and ORO staining revealed severe inflammation, ballooned hepatocytes and worse hepatic steatosis in HFD mice in a time-dependent manner. Moreover, liver injury and fat deposition were more obvious in APNKO mice than in wild-type mice (Fig. 1B). To investigate whether NLRP3 inflammasome is involved in HFD-induced liver injury and steatosis under the condition of adiponectin deficiency, the expression levels of NLRP3 inflammasome and related inflammatory markers were detected. As shown in Fig. 1C, adiponectin deficiency promoted the gene expression of NLRP3, ASC, caspase1 and IL1β, IL18, TNFα and IL6 in HFD-fed APNKO mice compared to wild-type mice.
Additionally, the concentration of IL1β, IL18, TNFα and IL6 in liver tissue lysate were remarkably higher in adiponectin-KO mice than in wild-type mice (Fig. 1D). In accordance to the above findings, immunohistochemical staining of the liver tissue sections further confirmed that NLRP3, caspase1 and IL1β expression were remarkably upregulated in HFD-fed adiponectin-KO mice compared to wild-type mice (Fig. 1E). These data strongly indicate that adiponectin deficiency aggravates HFD-induced liver injury and steatosis as well as sensitizes to HFD-induced NLRP3 inflammasome activation and downstream cytokine production.
The effects of adiponectin deficiency on ROS production and NFκB/AMPK/MAPK signaling pathways To further elucidate the possible regulatory mechanism by which adiponectin deficiency aggravates HFD-induced NLRP3 inflammasome activation, the levels of ROS production as well as NFκB, AMPK and MAPK(JNK and ErK1/2) signaling pathways were determined. As presented in Fig. 2A, ROS production was significantly elevated in the liver of HFD-fed adiponectin-KO mice compared to wildtype mice. Besides, NFκB expression was markedly upregulated in HFD-fed adiponectin-KO mice compared to wild-type mice (Fig. 2B). On the contrary, the phosphorylation levels of AMPK, JNK and ErK1/2 were markedly reduced in adiponectin deficiency mice, while there was no remarkable alteration in the protein levels of total AMPK, JNK and ErK1/2 ( Fig. 2C-E). These results suggest that adiponectin deficiency promotes HFD-induced NLRP3 inflammasome activation pathways (NFκB and ROS) and attenuates AMPK/JNK/ErK1/2 signaling pathways.
Adiponectin Alleviates PA-mediated NLRP3 Inflammasome Expression In Hepatocytes Taking into account that adiponectin deficiency aggravates HFD-mediated NLRP3 inflammasome expression, we further determine the effects of PA and adiponectin on NLRP3 inflammasome activation in hepatocytes. As presented in Fig. 3A, treatment with PA induced lipid droplet deposition and NLRP3 inflammasome expression in hepatocytes. To verify the roles of adiponectin in PAmediated hepatic activation of NLRP3 inflammasome complexes, hepatocytes were pre-treated with adiponectin for 2 h and then exposed to PA for 24 h. It was found that PA induced the expression levels of NLRP3, caspase1, ASC, IL1β, IL18 and other inflammatory markers (e.g., TNFα and IL6) when compared to control group. However, treatment with adiponectin could alleviate PA-mediated NLRP3 inflammasome activation (Fig. 3B-H). These findings reveal that adiponectin exerts a protective effect on PA-stimulated NLRP3 inflammasome activation in hepatocytes.

9
AMPK inhibitor attenuates adiponectin-mediated hepatoprotective effects against PA-mediated NLRP3 inflammasome activation A previous study has suggested that AMPK can play a vital role in the biological activities of adiponectin [25]. The results of in vivo experiments showed that the level of p-AMPK was reduced in adiponectin-KO mice fed a HFD. Thus, in the in vitro study, the effects of AMPK signaling pathway on NLRP3 inflammasome suppression by adiponectin were assessed with Compound C (10 um/ml, an AMPK inhibitor). It was found that the mRNA expression levels of NLRP3 (Fig. 4A), Caspase1 (Fig. 4B), ASC (Fig. 4C), IL1β (Fig. 4D), IL18 (Fig. 4E) TNFα (Fig. 4F) and IL6 (Fig. 4G) were markedly upregulated in Compound C + adiponectin + PA group compared to adiponectin + PA group (Fig. 4), suggesting that compound C can disrupt the protective role of adiponectin against PA-mediated NLRP3 inflammasome activation. To further confirm these findings, the protein expression levels of NLPR3, pro-capase1, caspase1 (P10), pro-IL1β and IL1β in hepatocytes were detected. Consistent with the above data, adiponectin abolished PA-mediated expression of NLRP3 inflammasome protein complexes. On the contrary, Compound C reversed the inhibitory effect of adiponectin on PA-mediated NLRP3 inflammasome activation in hepatocytes (Fig. 5A-D), In support of the above-mentioned data, similar results were also observed with regard to the concentrations of IL1β, IL18, TNFα and IL6 in cell culture supernatant (Fig. 5E-H).
Adiponectin plays a protective role against PA-mediated NLRP3 inflammasome activation via AMPK-JNK/Erk-NFkB/ROS signaling pathways Based on the above data, we speculated that adiponectin may inhibit PA-mediated inflammasome activation through AMPK signaling pathway. Therefore, we next determined the expression levels of AMPK and other related signaling pathways. As presented in Fig. 6A, adiponectin increased the phosphorylation levels of AMPK, JNK and ErK1/2 in hepatocytes. Our previous work has indicated that AMPK can act as an upstream regulator of JNK and Erk1/2 in hepatic stellate cells [26].Thus, we also detected the activities of JNK and Erk1/2 following the exposure to AMPK inhibitor. Our results demonstrated that Compound C markedly decreased adiponectin-induced expression levels of p-AMPK, p-JNK and p-Erk1/2 ( Fig. 6A-D), suggesting that AMPK can regulate p-JNK and p-Erk1/2 as upstream signaling pathways. Given that PA could induce NFκB phosphorylation and ROS production ( Fig. 6A and Fig. 6E-F) that act as the regulators of NLRP3 inflammasome activation [12], we further assessed the effects of adiponectin and Compound C on PA-mediated NFκB and ROS expression, It was found that adiponectin inhibited PA-mediated NFκB phosphorylation and ROS production in hepatocytes, and Compound C reversed the suppressive effects of adiponectin on PA-mediated NFκB expression and ROS production (Fig. 6F). These findings reveal that adiponectin can suppress PAmediated NLRP3 inflammasome activation in hepatocytes via AMPK-JNK/Erk1/2-NFκB/ROS signaling pathways.

Discussion
Accumulating evidence has suggested that adiponectin exerts beneficial effects on hepatic disorders [20,23], including steatohepatitis and liver fibrosis [26,27]. However, the potential mechanism responsible for these effects has been not yet fully elucidated. Recently, several studies have shown that adiponectin and adiponectin receptor agonist (adipoRon) exhibit protective effects against diabetes-related vascular disorders or nephropathy by suppressing NLRP3 activation [28,29].
However, it remains unclear whether adiponectin can affect NLRP3 inflammasome expression in hepatic diseases and its potential underlying mechanism. Therefore, in this study, we explored the effect and potential mechanism of adiponectin on PA-mediated NLRP3 inflammasome activation Adiponectin plays important roles in the metabolic functions of adipose tissue, skeletal muscle and liver [30]. Previous studies have shown that the serum levels of adiponectin are reduced in NASH patients [31], and hypoadiponectinemia is negatively associated with steatosis and inflammation in NAFLD patients [32]. Adiponectin-deficient mice exist hepatic excessive steatosis and necroinflammation in a mouse model of nonalcoholic steatohepatitis [20,23]. Consistent with these findings, our in vivo study also demonstrated that adiponectin deficiency aggravated HFD-induced liver injury, such as elevated serum ALT/AST levels, promoted lipid droplet accumulation and increased ROS production. In addition, it was found that adiponectin deficiency promoted HFDinduced NLRP3 inflammation activation and NFκB expression, and attenuated the phosphorylation levels of AMPK, JNK and Erk1/2. Several studies have indicated that NLRP3 inflammasome can be activated by endotoxin, ROS, uric acid, etc. [16,33], and adiponectin may prevent the progression of steatohepatitis by regulating oxidative stress [23]. Therefore, we speculated that the aggravating effect of adiponectin deficiency on NLRP3 inflammasome activation was due to the increased ROS production and activated NFκB signaling pathway. The underlying molecular mechanisms were further explored in vitro.
Hepatocytes were isolated from mice and cultured for 48hours, then cells were serumstarved for 6 hours and exposed to palmitic acid (PA, 300umol/ml) for 24 hours. For  were measured by ELISA. Results was presented as mean ± SEM, all experiments were performed at least three times and at least in triplicate. ***p < 0.001, **p<0.01, *p<0.05.