Emodin Attenuates Acetaminophen-Induced Hepatotoxicity via Cgas-STING Pathway

a natural bioactive compound from acetaminophen The study explored the effects of emodin on APAP-induced hepatotoxicity and investigated the potential molecular mechanisms. The levels of of with both group and in group 0.05, 0.01). The also


Introduction
Acetaminophen (N-acetyl-p-aminophenol, APAP), a widely used acetanilide analgesic and antipyretic drug all over the world, which is primarily used against cold or in uenza-induced headache and fever [1,2].
Although APAP has been considered highly safe at a recommended dose, its intentional or unintentional overdose can cause severe nephrotoxicity and hepatotoxicity, leading to life-threatening acute kidney injury and liver failure [3,4]. Every year, more than 200 million people use APAP, APAP-induced acute hepatic failure results in 200 deaths. However, the treatment for APAP poisoning is mainly limited to Nacetyl-L-cysteine (NAC), which is a nonspeci c antidote that restores endogenous glutathione (GSH) [5]. Therefore, it is of great signi cance to explore the possible molecular mechanism of liver damage caused by APAP for its clinical application, as well as the potential therapeutic drugs against its toxicity.
In ammatory response and oxidative stress are considered to be the main mechanisms of APAP-induced liver failure [6,7]. When patients take an overdose of APAP, most drugs are metabolized in the liver by UDP-glucuronidase (UGT) and sulfotransferase (SULT) enzymes into non-toxic compounds, which are subsequently excreted in the urine and bile [8]. The remaining APAP is oxidized by CYP450 enzyme into a toxic intermediate metabolite, N-acetyl-p-benzo-quinoneimine (NAPQI), which can lead to the deplete of GSH and the generation of protein adduct in liver [9]. The depletion of GSH and the NAPQI adducts cause mitochondrial dysfunction and massive reactive oxygen species (ROS) secreted from the injury hepatocytes, which in turn leads to hepatocellular apoptosis [10,11]. Intercellular contents released from these damaged cells, called damage-associated molecular patterns (DAMPs), which can stimulate nonparenchymal cells to produce and release in ammatory mediators and chemokines [12]. Under the action of chemokines, a variety of immune and in ammatory cells, such as monocytes and neutrophils, are recruited into the liver and promote in ammatory responses via the activation of innate immune signal transduction pathways, resulting in the necrosis and apoptosis of liver cells [13]. Importantly, blocking oxidative stress and inhibiting in ammation are important targets for protecting hepatocytes from hepatotoxicity of APAP.
In the present study, we investigated the protective effects of emodin on APAP-induced liver injury, evaluated its activity of anti-in ammatory, anti-oxidative stress, and anti-apoptosis, and explored the role of cGAS-STING pathway in the bene cial effects of emodin.

Animal experiments
A total of 32 male (6-8 weeks) C57BL/6 mice weighting 17-23g were supplied by Weitonglihua Biotechnology Co., Ltd (Hangzhou, China) and kept in the Experimental Animal Center of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology. All animals were fed under the rearing room with a temperature of 22 ± 3℃, a relative humidity of 55 ± 5%, and a day-night cycle at 12 hours. After 7 days adaptive feeding, 32 mice were randomly distributed in four groups of 8 mice each, healthy control group (control), APAP group (APAP), Emodin low-dose group (Emo-L), and Emodin highdose group (Emo-H).
The mice in Emo-L, and Emo-H group were orally administration with emodin for consecutive 5 days (15 and 30 mg/ kg/day respectively). Emodin was dissolved in 40% polyethylene glycol (PEG). The control and APAP group received the same volume of vehicle. Two hours after the last emodin administration, APAP were used to (intraperitoneally injection at the concentration of 300 mg/kg body weight) induce acute hepatic injury. APAP was dissolving in saline. The control group received the same volume of vehicle. Twenty-four hours later, all the mice were euthanized with overdose of 1% pentobarbital sodium and specimens were collected immediately. All procedures of animal experiments were approved by Committee of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology.

Tissue collection
Serum samples were obtained from peripheral blood by centrifugation at 3000 rpm for 10 min at 4 ℃.
Half of left lobe of liver samples were also separated and collected. All above samples were stored at -80 ℃. Remaining left lobe of liver samples were xed with 4% paraformaldehyde and embedded in para n.
All samples were processed on ice as soon as possible to prevent protein degradation.

Hepatic histological analysis
Sections of xed hepatic samples embedded in para n were used for routine hematoxylin and eosin (H&E) staining. Sections were observed and photographed by the optical microscope (Olympus, Japan). Injury grades of hepatic samples were evaluated by Suzuki's score according to H&E staining results [20].

Biochemical assays
Serum levels of ALT, AST, ALP and ALB, and SOD, MDA, as well as the levels of GSH in liver tissues were tested by the biochemical kits.

Enzyme-linked immunosorbent assay (ELISA) analysis
The levels of IL-1β, TNF-α, IL-6, IL-10 in hepatic tissues were detected by ELISA kits followed the protocol offered by manufacture.

Protein extraction and western-blot analysis
Total proteins were extracted from hepatic tissues using RIPA buffer. The concentration of total proteins was quanti ed by bicinchoninic acid assay. Total proteins were subjected to 10-12% SDS-PAGE electrophoresis, and the proteins were transferred to PVDF membranes (Sigma, MA, USA). PVDF membranes were blocked with 5% non-fat milk for 1.5 hours at room temperature. Then the membranes were incubated with primary antibodies overnight at 4 ℃. PVDFs were incubated with corresponding secondary antibodies. Gray-scale values of straps was analyzed by Image J software.
TUNEL staining analysis TUNEL staining were used to detect the apoptosis rate of hepatocytes. Firstly, 6-micron sections of hepatic tissue were depara nized and rehydrated. The sections were treated with proteinase K for 15 mins, followed by incubation with TdT at 37℃ for 2 hours. Finally, results of TUNEL staining were observed and photographed by the optical microscope. The hepatic TUNEL positive cell number was assessed by Image Picture Pro software.

Statistical analysis
All experimental data in this study were come from at least 3 independent experiments and showed as means ± standard division (SD). One way ANOVA was performed to compare the differences among the four groups. A p ≤ 0.05 was considered statistically signi cant.

Results
Emodin alleviated APAP-induced liver injury Figure 2A shows that widely in ammatory in ltration in hepatic tissue, severe hepatocytes ballooning degeneration, and extensive hepatocytes necrosis in APAP group, which had not presented in control group. As is shown in Fig. 2B, Suzuki's score in APAP group increased compared to that in the control group (p < 0.05, p < 0.01). APAP group showed higher levels of serum ALT, AST, ALP, and lower levels of ALB compare to control group (p < 0.05, p < 0.01) (Fig. 2C-F). The above differences indicated the model of APAP-induced liver injury was successfully established in mice.
Compared to those indicators in APAP group, serum levels of ALB, ALT and ALP decreased, and the levels of ALB was upregulated in Emo-H group (p < 0.05). The results of Suzuki's score indicating that the hepatic injury was alleviated signi cantly under the administration of high-dose Emo (p < 0.05) (Fig. 2).
Emodin inhibited APAP-induced oxidative stress of liver tissues It is considered that hepatic damage is associated with the upregulation of oxidative stress. Our results exhibited that the levels of SOD, GSH in APAP group were downregulated and the levels of MDA was upregulated compared with control group (Fig. 3A-C) (p < 0.05, p < 0.01). However, the levels of SOD, MDA, and GSH were improved signi cantly in Emo-H group, which indicated that emodin could inhibit the oxidative stress caused by APAP-mediated liver injury. Previous studies indicated that transcription factor Nrf2 and its downstream proteins HO-1 and NQO1 exert anti-oxidant properties in cellular. So, we also detected the levels of Nrf2, HO-1, and NQO1 in hepatic tissues. Nrf2, HO-1, and NQO1 were downregulated in APAP group (Fig. 3D and E) (p < 0.05, p < 0.01), which indicated that they were failed to ful ll their protective roles in APAP-mediated hepatic injury. Nevertheless, the expression levels of Nrf2, HO-1, and Apoptosis is a programmed cell death procedure when cells confront harsh environment or suffered severe destroy. In APAP-induced hepatic injury, hepatocytes underwent strict oxidative stress and in ammation, which might lead to hepatocellular apoptosis ultimately. TUNEL staining is an effective method to label fragmented DNA emerged by cellular apoptosis. Figure 5A and B illustrated that there were more fragmented DNA in APAP group than those in control group and Emo-H group (p < 0.05, p < 0.01).
The results shown that the levels of Bax in APAP group were up-regulated (p < 0.01), and reduced in Emo-H group (p < 0.05). The levels of Bcl-2, a kind of antiapoptotic proteins, had opposite variation trends with Bax both in APAP group and in Emo-H group (p < 0.05, p < 0.01). The above results presented that APAPinduced hepatic injury also had severe hepatocellular apoptosis like other acute hepatic injury, and emodin could alleviate the apoptosis by rectifying the oxidative stress and in ammations.

Emodin attenuated the activity of cGAS-STING signaling pathway
As is shown in Fig. 6A and B, compare to Control group, the expression of cGAS-STING signaling pathway related proteins, including P-TBK1, P-IRF3, cGAS, and, STING signi cantly decreased in APAP group (p < 0.05, p < 0.01). This suggested that the cGAS-STING signaling pathway was activated in model mice. The expression of those proteins was signi cantly reduced (p < 0.05, p < 0.01) in Emo-H group, indicating that emodin could inhibit the hepatocellular injury caused by APAP via regulating the activity of cGAS-STING signaling pathway.

Discussion
APAP is one of the most widely used analgesic and antipyretic drugs around the world [21]. However, it has been reported that overdosage of APAP could cause liver injury even death [22,23]. Accumulation of NAPQI, one of the intermediate metabolites of APAP in liver, could actuate the liver injury by promoting oxidative stress and in ammation of hepatic tissues, which nally triggers the hepatocellular apoptosis [22,24,25]. In the present study, abnormal pathologic alternations, increased Suzuki's score, upregulated expression of AST, ALT and ALP, as well as downregulated ALB level in APAP group of mice indicated that the modelling of acute hepatic injury had been successfully established.
Emodin is the major component and one of quality control indexes of RP, a traditional Chinese herb [26-28]. Emodin has been reported that it has biologic activities and bene cial effects, such as hepatoprotective, anti-in ammatory responses, antibacterial, antivirus, as well as neuroprotective [19,[29][30][31]. Previous studies demonstrated that emodin could protect against APAP-induced hepatic injury via multiple targets, including cytochrome P450 (CYP), and AMP-activated protein kinase (AMPK)/ Yesassociated protein (YAP) signaling pathway [17]. Our study suggested that emodin attenuated APAPinduced hepatic injury by activating Nrf2 anti-oxidant pathway and inhibiting NLRP3 via downregulating cGAS-STING signaling pathway.
Oxidative stress is one of landmark events of APAP-induced acute hepatic injury [32]. In experiments of rodent model, routine dose of APAP were mainly involved in glucuronidation and sulfation, and the nontoxic metabolites were then excreted through vile and urine [33,34]. However, when excessive APAP is oxidated to NAPQI by cytochrome P450 (CYP), which binds with GSH to inhibit toxic responses [35,36]. The accumulation of NAPQI result in the depletion of GSH in liver, leading to the decrease of anti-oxidant enzyme activities, and the massive production of ROS [37]. ROS directly cause cytoplasmic vacuolation, hepatocyte apoptosis, and liver failure [38]. SOD, MDA and GSH are the commonly used indexes to measure the levels of intracellular oxidative stress, and SOD and GSH are involved in the anti-oxidant processes, the levels of MDA represent the extents of oxidative injury [39,40]. In our study, the levels of SOD and GSH were signi cantly increased with the treatment of emodin, the concentration of MDA was reduced. Anti-oxidant enzymes like HO-1 and NQO1, transcription factor Nrf2 is closely related to oxidative stress-associated cellular damage. The loss of Nrf2 in mice caused severe hepatic injury in chlorogenic acid induced acute liver injury [41]. Nrf2 can translocate to nucleus under the stimulation of ROS, and binds to anti-oxidant response element (ARE), leading to the transcription of anti-oxidant enzymes, including NQO1 and HO-1 [42,43]. Our results shown that emodin downregulated CYP2E1 expression and upregulated Nrf2, HO-1, NQO1 expression.
In APAP-induced hepatic injury, oxidative stress causes the activation of in ammatory related-signaling pathways, which further aggravates liver injury [44]. NLRP3 has been considered as an important proin ammatory factor activated by oxidative stress [45,46]. It has been reported that NLRP3 is one of potential in ammatory mediators in APAP-induced hepatic damage, partly because of lower levels of NQO1 in the liver [47,48]. Besides, immune cells in hepatic activated by damage-associated molecular patterns (DAMPs), which was conducted with mitochondrial DNA (mtDNA) and fragmented nucleus DNA as well as other proteins released from injured cell, could also be involved in hepatic in ammation [32,49]. In ammation in APAP-induced damage model is thought to be ampli ed by IL-1β, IL-6, and TNF-α, produced by Kupffer cells and hepatic dendric cells [12]. IL-10 is known as an anti-in ammatory cytokine and could suppress the acute hepatic injury, moreover, IL-10 de ciency mice showed more severe hepatic damage. In the present study, APAP was found to initiated the activation of NLRP3 in ammasome, which was inhibited by treatment with emodin.
APAP-induced liver damage causes hepatocyte death via necrosis and apoptosis [50]. Bax and Bcl-2 can regulate the of progression apoptosis [51]. The excessive APAP-adducts have been proved to promote hepatocellular apoptosis [52]. We found that emodin alleviated hepatocyte necrosis and apoptosis induced by APAP, as well as inhibited Bax/Bcl-2 ratio. Above results indicated that emodin could inhibit APAP-induced hepatic injury via multiple processes.
Cyclic GMP-AMP synthase (cGAS), a sensor of DNA, which could be activated by virus DNA or aberrant intracellular DNA [53]. cGAS can recognize DNA via electrostatic action and hydrogen-bonding interaction [54], and catalyze the synthesis of Cyclic guanosine monophosphate-adenosine monophosphate (cGAMP) from adenosine triphosphate (ATP) and guanosine triphosphate (GTP) after recognizing DNA [55]. Stimulator of interferon genes (STING) is found on the outer mitochondrial membrane and endoplasmic reticulum in the form of a dimer in inactivated state [56]. Dimer STING can bind with cGAMP catalyzed by cGAS and subsequently translocated to the vesicles around perinuclear region from endoplasmic reticulum by Golgi body [57]. TANK-binding kinase 1 (TBK1) were enrolled into the vesicles to phosphorylated and activated STING [58]. Furthermore, phosphorylated STING can phosphorylate transcription factor interferon regulatory factor 3 (IRF3) [58], one of the most important downstream transcription factors of cGAS-STING signaling pathway, which is closely related to in ammation and apoptosis [59]. Finally, phosphorylated IRF3 enter the nucleic and promote the transcription of IFN-α [58]. In addition to cGAMP, STING can also be activated by second messenger cyclic guanosine monophosphate (cGMP) and cyclic adenosine monophosphate (cAMP) [60].
It is noteworthy that cGAS-STING signaling pathway also participate in multiple kinds of acute and chronic hepatic injury, including radiation-induced liver injury [61], non-alcoholic fatty liver disease (NAFLD) and high fat diet-associated hepatic injury [62], and hepatitis B virus (HBV) infection-associated liver injury [63]. However, it is interesting that STING mainly expressed in hepatic nonparenchymal cells like intrahepatic macrophages instead of hepatic parenchymal cells, which resulted that the resistance to HBV infection is failed in hepatocellular [63]. It plays vital role in APAP-induced hepatic injury that necrosis hepatocellular released generous mtDNA and fragmented nucleus DNA to intracellular space, and the DNA might amplify the hepatic injury as DAMPs [64]. The cGAS-STING signaling pathway, is associated with innate immune response and DNA recognition. Araujo et al. found that the activation of cGAS-STING signaling pathway plays an important role in APAP-induced hepatic injury [65]. The levels of cGAS and STING were upregulated in hepatic parenchymal cells, and the levels of STING were also consistently increased in hepatic nonparenchymal cells. Simultaneously, massive mtDNA accumulated in extracellular space, which is one of the causes of the activation of cGAS-SITNG signaling pathway in hepatocellular. Activated cGAS-STING signaling pathway in hepatic parenchymal cells promotes in ammation, apoptosis, and necrosis of hepatic tissues [65]. In addition, hepatic nonparenchymal cells with activated cGAS-STING signaling pathway secret IFN-α, which also aggravates the liver damage. Therefore, inhibiting the cGAS-STING signaling pathway is a potential therapeutic method of APAPinduced hepatic injury. Our study shown that emodin could inhibit the expression of cGAS, STING, P-IRF3, and P-TBK1 in liver tissues. These results suggested that the protective effect of emodin on APAPinduced liver damage might be associated with the inhibition of the cGAS-STING signaling pathway.

Conclusion
In the present study, the results shown that administration of emodin attenuated APAP-induced liver injury mainly by alleviating hepatic pathological damage, apoptosis, oxidative stress, and inhibiting the in ammatory response. Additionally, we found that emodin suppressed the cGAS-STING signaling pathway against APAP-induced in ammatory responses and apoptosis. This study supports more evidences for the application of emodin and RP. Building on prior research, it is reasonable to speculate that emodin might be a potential candidate for the prevention and treatment of APAP.

Declarations COMPETING INTERESTS
None.

DATA AVAILABILITY
The data used to support the ndings of this study are included in the paper.
Ethics Approval and Consent to Participate.
All experiment procedures were approved and carried out in accordance with Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology Care Committee guidelines.