Slow-Release H2S Donor Anethole Dithiolethione Protects Liver From Lipotoxicity by Improving Fatty Acid Metabolism

“Lipotoxicity” induced by free fatty acids (FAs) plays a central role in the pathogenesis of many metabolic diseases, with few treatment options available today. Hydrogen sulfide (H2S), a novel gaseous signaling molecule, has been reported to have a variety of pharmacological properties, but its effect on FAs metabolism remains unclear. The purpose of this study was to investigate the effect and mechanisms of anethole dithiolethione (ADT, a sustained-release H2S donor) on hepatic FAs metabolism. ADT was administered daily for 4 weeks in male Syrian golden hamsters fed a high fat diet (HFD), and FAs profiles of liver tissues were analyzed using GC-MS. The results showed that in HFD-fed hamsters, ADT treatment significantly reduced the accumulation of toxic saturated and monounsaturated fatty acids (C16:0, C18:0, C16:1, and C18:1n9), while increased the content of n-6 and n-3 series polyunsaturated fatty acids (C20:3n6, C20:4n6, and C22:6n3). Mechanistically, ADT obviously inhibited the overexpression of acetyl-CoA carboxylase1 (ACC1), fatty acid synthase (FAS), and stearoyl-CoA desaturase1 (SCD1), and up-regulated the levels of fatty acid transport proteins (FATPs), liver fatty acid binding protein (L-FABP), carnitine palmitoyltransferase 1α (CPT1α), fatty acid desaturase (FADS)1 and FADS2. Notably, ADT administration significantly promoted Mitofusin1-mediated mitochondrial fusion and fatty acid β-oxidation. These findings suggest that ADT plays a beneficial role by regulating the synthesis, desaturation, β-oxidation, uptake, binding/isolation, and transport of FAs. In conclusion, ADT is effective in improving FAs metabolic disorders and liver injuries caused by HFD, which renders ADT a candidate drug for lipotoxicity-induced diseases.


INTRODUCTION
Fatty acids (FAs) are indispensable sources of energy in cells, and also important bioactive mediators involved in many homeostasis processes, including metabolism and regulating inflammatory immune responses (Masoodi et al., 2015). However, unlike adipocytes, which possess a strong capacity to store excessive free fatty acids (FFAs) in lipid droplets in the form of triglycerides, lipid overload in non-adipose tissues (such as heart, liver, kidney, skeletal muscle, and pancreatic b-cells) causes lipotoxicity, leading to cell dysfunction or death. Lipotoxicity has been reported to be mainly caused by long-chain saturated fatty acids (SFAs), especially palmitic acid (PA, C16: 0) (Leamy et al., 2013). Accumulating data suggest that abnormal FAs metabolism and its induced lipotoxicity are closely related to the risk of developing non-alcoholic fatty liver disease (NAFLD), diabetes, atherosclerosis, heart failure, and even multiple cancers (Afonso et al., 2016;Blucher and Stadler, 2017;Cansancao et al., 2018;Jiang et al., 2019;Zhou et al., 2019). Therefore, inhibition of FAs-related lipotoxicity represents a potential therapeutic strategy, which is of great interest.
The liver is the main metabolic organ of FAs, and the imbalance in its synthesis, uptake and disposal (mainly including mitochondrial oxidation and endoplasmic reticulum reesterification) will induce lipotoxicity, further leading to the body dysfunction. Mitochondrial b-oxidation, the most important metabolic pathway of FAs, is mainly regulated by rate-limiting enzymes such as carnitine palmitoyl-transferase 1a (CPT1a), which serves as a gatekeeper for FAs to enter mitochondria. In addition, it has been demonstrated that in starved mitochondrial fusion protein 1 knockout (Mitofusin1KO) mouse embryonic fibroblasts (MEFs), mitochondria were fragmented, reducing the b-oxidation rate and leading to lipid accumulation in lipid droplets (Rambold et al., 2015;Wrighton, 2015). However, there are no published data on the relationship between liver FAs metabolism and Mitofusin1 expression during nutrient oversupply.
Hydrogen sulfide (H 2 S), a well-known novel gaseous signaling molecule, is increasingly recognized as a crucial regulator of cardiovascular diseases. H 2 S in mammalian tissues is synthesized endogenously by three enzymes: cystathionine-b-synthase (CBS), cystathionine-g-lyase (CSE) and 3-mercaptopyruvate sulfurtransferase (3-MPST). Emerging data indicate that endogenous H 2 S level is significantly negatively correlated with abnormal lipid metabolism, including hyperlipidemia, NAFLD, and atherosclerosis Sun et al., 2015;Li et al., 2017). Exogenous H 2 S donors (NaHS, GYY4137, etc.) can significantly improve these metabolic diseases through multiple properties, such as anti-inflammatory, antioxidant, inhibiting foam cell formation, improving endothelial function, and activating liver autophagy (Liu et al., 2013;Sun et al., 2015;Durante, 2016;Li et al., 2017). And more notably, a recent study found that FFAs up-regulate the liver expression of 3-MPST, and subsequently inhibit the CSE/H 2 S pathway, impairing the endogenous synthesis of H 2 S, and leading to NAFLD (Li et al., 2017). However, the effect of H 2 S on FAs metabolism in the liver remains unclear.
At present, H 2 S-releasing "drugs" used in research have been largely limited to simple sulfide salts, mainly NaHS. However, NaHS, the so-called immediate-release H 2 S donor, releases excessive H 2 S instantaneously and therefore does not mimic the production of endogenous H 2 S . Anethole dithiolethione (ADT; 5-(4-methoxyphenyl)-3H-1,2-dithiole-3thione), clinically used as a hepatoprotective and choleretic drug, is also a prodrug of H 2 S (Wang et al., 2014;Szabo and Papapetropoulos, 2017). In vivo, ADT is rapidly metabolized into the active metabolite ADT-OH, with a half-life of about 3.1 and 4.4 h, respectively Wang et al., 2014). Compared with NaHS, ADT releases H 2 S and significantly increases serum H 2 S levels over a long period of time. In 2019, Dulac et al. proposed the first detailed mechanism for the production of H 2 S by ADT and ADT-OH in the presence of rat liver microsomes, NADPH and O 2 ( Figure 1) (Dulac et al., 2019). Therefore, this article aims to investigates how clinically available sustained-release H 2 S donor ADT affects hepatic fatty acid metabolism under high-fat diet (HFD) and explores its possible mechanisms.

MATERIALS AND METHODS
Drug ADT (Lot.N1017A) was purchased from Dalian Meilun Biotechnology Co., LTD., and its clinical dosage is 75 mg per day. The dosage of ADT in Syrian golden hamsters was calculated by human equivalent dose (HED) based on body surface area (Huang et al., 2015), with a conversion dose of 9.25 mg/kg. Therefore, the dose gradient of ADT in this study was set as: 5, 10, 20, and 40 mg/kg. ADT is insoluble in water, suspended in 0.5% sodium carboxymethylcellulose (CMC-Na) and 0.5% soybean lecithin, administered daily by gavage.

Animals and Diets
This study was reviewed and approved by the Laboratory Animal Care and Welfare Committee of Central South University (Approval No. 2018sydw0215). Fifty-four male Syrian golden hamsters (80-100 g) were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. (Qualified Certificate No. SCXK Jing 2016-0011). Hamsters were housed in IVC cages and free access to diet and water, which subjected to a 12-h light/ dark cycle with a relative humidity of 50% ± 10% and a temperature of 22°C to 25°C.

Experimental Design
After adaptation for one week, 54 hamsters were randomly divided into six groups, with 9 animals in each group. (1) control group: control diet with vehicle (0.5% CMC-Na and 0.5% soybean lecithin) treatment; (2) HFD group: high-fat diet with vehicle (0.5% CMC-Na and 0.5% soybean lecithin) treatment; (3-6) HFD+ADT group: high-fat diet with different doses of ADT. ADT and/or vehicle were administered to the hamster by gavage once a day for 4 weeks. After 4 weeks, all hamsters fasted overnight and were euthanatized at the end of the experiment. Blood samples were collected from the heart and kept at room temperature for 1 h before centrifugation at 3500 rpm for 10 min. The serum was separated and stored at −80°C until analysis. The left lobe of liver was fixed in 4% paraformaldehyde solution, and the rest was stored at −80°C after quick-freezing by liquid nitrogen.

Biochemical and Histological Analysis
Serum glutamate alanine aminotransferase (ALT), aspartate aminotransferase (AST) and total bile acid (TBA) were measured by automatic biochemical analyzer (Hitachi 7600-210) in the laboratory of Second Xiangya Hospital. The fixed liver tissue was embedded in paraffin blocks, and 4-mm-thick slices were cut and stained with hematoxylin and eosin (HE) to observe histopathological changes.

Sample Pretreatment and FFAs Determination
FFAs in liver tissue were extracted and measured by GC-MS (Agilent 7890 A/5975 C) according to the method described by Tang et al. (2018). In short, 750 ml of dichloromethane and methanol (CH 2 Cl 2 : CH 3 OH = 1:2) mixture was added to liver tissue and homogenized, followed by centrifugation to obtain the supernatant. Then, 100 mM butylated hydroxytoluene (BHT, to prevent lipid peroxidation), 250 ml CH 2 Cl 2 and 250 ml water were sequentially added to the supernatant and vortexed for 30 s. We transferred the lower phase, centrifuged again to take the supernatant which then evaporated to dryness under nitrogen. The residue was dissolved in n-hexane, 10 ml internal standard (heptadecanoic acid) and 2 ml 0.5 M KOH-MeOH were added, followed by water bath heating at 60°C for 20 min. After cooling for 10 min, 3 ml of 12.5% H 2 SO 4 in methanol solution (to methylate the sample) was added and heated again at 60°C in the water bath for 1 h. After cooling the sample vial, 2 ml n-hexane and 1 ml saturated sodium chloride (NaCl) solution were added, and let stand for 10 min to stratify. The supernatant (hexane fraction) was then transferred for GC-MS analysis.
GC-MS analysis was performed on Agilent 7890 A/5975 C, with specific parameter settings modified according to the previously reported procedures. The samples were separated by VF-23 ms chromatographic column (Agilent: 30 m(length), 0.25 mm (inner diameter), 0.25 mm (film thickness)). FFA composition was determined based on the retention time of validated fatty acid methyl ester standard (Supelco 37, sigma). FFA content was expressed as a percentage of peak area.

Mitochondria Stain
The fixed liver tissue was embedded in a paraffin block and cut into 4-mm-thick sections. Heat shock protein 60 (HSP60, GB11243, 1:500), a mitochondrial chaperonin, was incubated with sections at 4°C overnight to label the mitochondria. CY3 goat anti-rabbit (GB21303,1:300) was used as the secondary antibody and incubated for 50 min. After washing three times with PBC, nuclei were stained with DAPI for 10 min. Finally, the morphology of mitochondria was observed by fluorescence microscope (NIKON ECLIPSE C1).

ROS Determination
The liver tissue was prepared into frozen sections (10 mm thick) and then incubated with ROS dye for 30 min. After washing with PBC for three times, nuclei were stained with DAPI for 10 min. Fluorescence was detected in a positive fluorescence microscope (NIKON ECLIPSE C1).

Statistical Analysis
Statistical comparisons of the data were performed by SPSS 18.0 software using one-way analysis of variance (ANOVA). The experimental results were presented as mean ± standard deviation (SD). P values < 0.05 were considered statistically significant.

Quantitative Changes of FFAs in Hamster Liver Treated With ADT
Quantitative changes of 13 FFAs detected by GC-MS in hamster liver in different experimental groups are shown in Table 3 (n = 9). These data indicated that total SFA and monounsaturated fatty acid (MUFA) increased significantly in the HFD group compared to the control group. However, there was no obvious change in total Polyunsaturated fatty acid (PUFA) between the two groups, with a slight increase in n-6 fatty acids and a decrease in n-3 fatty acids. ADT treatment significantly reduced total SFA and MUFA in the liver of hamsters fed a high-fat diet, and evidently up-regulated n-3 PUFA.
The specific changes of different FFAs between the experimental groups were further analyzed. C16:0 as well as C18:0, the two main SFAs, were increased in HFD, and this increase was subsequently inhibited by ADT treatment. C18:1n9, the main MUFA in the body, was significantly elevated in HFD, with concentrations nearly tripling compared to control levels, which can be reduced by ADT administration. For these three FFAs, ADT at 10 mg/kg was the most effective. In addition, C16:1 was also obviously up-regulated in HFD and was reduced

Gene
Forward primer Reverse primer in a dose-dependent manner by ADT. Meanwhile, the essential fatty acids (EFAs) including C18:2n6 and C18:3n3 were increased in HFD group and decreased in HFD + ADT group.

Effects of ADT on Synthesis, Desaturation, and Uptake of Fatty Acids in Liver
To investigate the molecular mechanism of ADT, we further quantitatively measured mRNA expression levels of genes related to FA synthesis, desaturation, and uptake in the liver (Figure 2). The expression of ACC1 and FAS was significantly increased in the HFD group and decreased dose-dependently in HFD + ADT groups ( Figure 2A). SCD1, a key rate-limiting enzyme that desaturates SFA into MUFA, was significantly elevated in the HFD group and attenuated by ADT administration. The changes of SCD1 were associated with changes in C16:1/C16:0 and C18:1n9/C18:0 ratios ( Figure 2B). On the contrary, FADS1 and FADS2, key rate-limiting enzymes in the desaturation of PUFA, were obviously down-regulated in HFD and dose-dependently increased by ADT treatment. Similarly, the changes in FADS1 and FADS2 were consistent with changes in C20:4n6/C18:2n6 and C22:6n3/C18:3n3 ratio ( Figure 2C).
Moreover, genes involved in liver FA uptake showed that the expression of CD36 was not obviously different among the groups, while the expression of FATP2, FATP4, and FATP5 was significantly reduced in HFD group but activated by ADT therapy ( Figure 2D).

Effects of ADT on FAs Transport, b-Oxidation, and Mitochondrial Fusion in Liver
Excessive free FAs in hepatocytes are bound/isolated by L-FABP and then transported to various cell compartments for rapid removal, mainly by mitochondrial b-oxidation. Therefore, the mRNA and protein expression levels of L-FABP, CPT1 and Mitofusin1 were quantitatively measured, and the liver mitochondria were stained to observe their fusion status ( Figure  3). As shown in Figure 3A, L-FABP, CPT1a, and Mitofusin1 were significantly decreased in HFD group, but remarkably raised in the HFD + ADT group. Furthermore, ADT treatment strikingly promoted mitochondrial fusion in HFD-fed hamster hepatocytes, which was consistent with the expression changes of Mitofusin1 ( Figure 3B).

Effect of ADT on ROS and Liver Function of HFD-Fed Hamsters
To establish the major defect of abnormal FAs metabolism driving by HFD, we performed fluorescence staining of liver ROS. The results showed that HFD significantly increased the level of ROS, while ADT administration dramatically reduced the excessive accumulation of ROS ( Figure 4A).

DISCUSSION
FFA-induced lipotoxicity plays a pathological role in the occurrence and development of many metabolic diseases, however, current therapeutic strategies are limited. Our results indicated that ADT administration significantly reduced the concentration of toxic SFA and MUFA in HFD hamster liver, and increased the level of n-3 PUFA. In addition, ADT obviously decreased serum ALT, AST, TBA and liver ROS levels, and improved hepatic steatosis. These results confirmed that ADT can effectively improve FAs metabolic disorders and Liver damages caused by HFD, making ADT a promising candidate drug for lipotoxicity therapy.
-mediated inflammation, and activation of death receptors (Chavez and Summers, 2003;Anderson et al., 2012;Salvado et al., 2015;Win et al., 2015). In this study, HFD-fed hamster livers showed increased level of SFAs, especially PA, while ADT treatment significantly inhibited this overaccumulation. This may partially explain the protective effect of ADT on liver injury, and we further explored the molecular mechanisms underlying these changes.
Hepatic SFAs are mainly derived from de novo lipogenesis (DNL), adipose tissue and diets. Under physiological conditions, free FAs are bound/sequestered by L-FABP and then transported to oxidative (peroxisomes, mitochondria) or stored (endoplasmic reticulum, lipid droplets) organelles for rapid removal, minimizing the toxic effects of excess free FAs (Smathers et al., 2013;Wang et al., 2015). Mitochondrial boxidation is the most important metabolic pathway of FAs. Of note, Rambold et al. recently found that in nutrient-deficient Mitofusin1 knockdown MEFs, only fused mitochondria can efficiently take up FAs and ensure its homogenous distribution A B C FIGURE 4 | Effect of ADT on ROS and liver function of HFD-fed hamsters. (A) Fluorescent images of ROS in hamster liver (magnification, ×200). (B) Serum levels of ALT, AST, and TBA in each group of hamsters (n = 9). ALT, alanine aminotransferase; AST, aspartate aminotransferase; TBA, total biliary acid. Data are presented as mean ± SD. Differences between groups were assessed by one-way analysis of variance (ANOVA) for multiple comparisons followed by LSD or Dunnett's t-test for post hoc tests. ## p < 0.01, and ### p < 0.001 versus control. *p < 0.05, **p < 0.01, and ***p < 0.001 versus HFD. (C) Histopathological examination of hamster liver sections by HE staining (magnification, ×100). throughout the mitochondria, maximizing the use of FAs for boxidative reactions (Rambold et al., 2015). Therefore, this study investigated the expression of genes involved in FAs metabolism, including synthesis, oxidation, uptake, and transport. We found that HFD induced overexpression of ACC1 and FAS, while the levels of L-FABP, CPT1a, and Mitofusin1 were significantly down-regulated, and mitochondria were also fragmented. Increased synthesis, weakened binding/transportation, and impaired b-oxidation are concurrently operative to provide a continuous excess of ROS and free FAs, especially PA. These excess PA and its toxic metabolites can activate various intracellular reactions, leading to lipotoxic stress in the endoplasmic reticulum and mitochondria, which in turn exacerbated the accumulation of free FAs. Compared with HFD group, ADT administration significantly inhibited ACC1 and FAS, up-regulated L-FABP, CPT1a, and Mitofusin1, and promoted mitochondrial fusion, thereby effectively reducing the accumulation of toxic SFAs and improving liver damage.
Appropriate intracellular balance of SFAs and MUFAs is regulated by SCD1 (Singh et al., 2015). SCD1 catalyzes the desaturation of SFA to MUFA, mainly producing oleic acid (OA, C18:1n9). It has been reported that OA is mainly incorporated into relatively inert triacylglycerol (TAG) and stored in lipid droplets, so metabolizing SFA to OA via SCD1 reduces the excessive formation of toxic lipid intermediates (diacylglycerol and ceramides) (Man et al., 2006;Palomer et al., 2018). In this study, long-term HFD induced SCD1 and significantly increased the concentration of OA, which may be due to the body's adaptive results in the face of ongoing metabolic stress. However, this relative protection by channeling the SFA into the less toxic lipid pool is only a temporary measure, as the TG buffer capacity in liver tissue is limited. Continuous overproduction of OA results in liver lipid accumulation, steatosis, and low-grade chronic inflammation, thereby potentiating the metabolic syndrome (Ricchi et al., 2009;Singh et al., 2015). In addition to effectively reducing the accumulation of toxic SFA, our results indicated that ADT treatment also inhibited SCD1 overexpression and significantly decreased the content of MUFA (mainly OA).
Unlike SFAs and MUFAs, PUFAs have important physiological functions in the body and are associated with beneficial health effects. Emerging evidence shows that excessive intake of HFD reduces the levels of AA (C20:4n6), EPA (C20:5n3), and DHA (C22:6n3), enhances the production of pro-inflammatory arachidic acids and reactive oxygen species (ROS), leading to pro-inflammatory status (Das, 2018;Gai et al., 2018). AA, EPA and DHA are endogenously derived from essential fatty acids (EFAs: LA, C18:2n6; ALA, C18:3n3) through a successive series of desaturation and chain extension steps. The rate-regulating conversion steps in this process are mediated by D-6 desaturase (FADS2) and D-5 desaturase (FADS1) (Koletzko et al., 2019). Our results also confirmed that HFD impaired the desaturated metabolic pathway of EFAs: the ratio of AA/LA and DHA/ALA in HFD hamsters was down-regulated, and the expressions of FADS1 and FADS2 were significantly decreased. In this study, the liver concentration of EPA was too low for accurate quantification. ADT administration obviously up-regulated FADS1 and FADS2, increasing the conversion of EFA to AA and DHA.
In conclusion, this study demonstrates that ADT is effective in improving HFD induced fatty acid metabolism disorders and liver injuries. Specifically, ADT may exert protective effects by: (i) inhibiting overexpression of ACC1, FAS, and SCD1, thereby reducing endogenous synthesis of toxic palmic acid (C16:0) and oleic acid (C18:1); (ii) up-regulating FATPs, L-FABP, CPT1a, and Mitofusin1, thus enhancing the b-oxidation of liver fatty acids; (iii) activating FADS1 and FADS2 to improve the A B FIGURE 5 | Changes of FAs metabolism in hamster liver of HFD group (A) and HFD+ADT group (B). Compared with the HFD group, ADT administration significantly inhibited the overexpression of ACC1, FAS and SCD1, activated FADS1 and FADS2, thereby reducing endogenous synthesis of PA, SA and OA, and improved the desaturated metabolic pathway of EFAs. In addition, ADT upregulated the levels of FATPs, L-FABP, and CPT1a, thus increasing the uptake, binding/ isolation, transport, and b-oxidation of FAs. More importantly, ADT obviously increased the expression of Mitofusin1 to promote mitochondrial fusion state and maximal b-oxidation. PA, palmitic acid; SA, stearic acid; OA, oleic acid; EFAs, essential fatty acids, including LA and ALA; MFN1, Mitofusin1. desaturated metabolic pathway of EFAs (outlined in Figure 5). These results may provide important new insights into the role of ADT in hepatic FAs metabolism and perhaps new therapeutic strategies for lipotoxicity. However, further researches are still needed to evaluate the potential application of ADT in the clinical treatment of lipotoxicity-induced metabolic diseases (eg, NAFLD). Moreover, there is one further point to make. The clinically available H 2 S sustained-release donor ADT is a prodrug, and it is difficult to determine whether its regulatory effect on liver FFAs depends on itself or H 2 S. In our opinion, H 2 S largely mediated the FFAs regulation of ADT. First, ADT is rapidly demethylated into ADT-OH in vivo, and their dithiolethione pharmacophore is cleaved by liver microsomes to release H 2 S (Wang et al., 2014;Szabo and Papapetropoulos, 2017). Second, more and more evidences show that H 2 S mediates the biological effects of ADT and ADT-OH. For example, Zhang et al. found that ADT and ADT-OH can activate Nrf2 signaling through the Ssulfhydration of Keap1 (Zhang and Munday, 2008). In 2013, Yang et al. used NaHS and CSE knockout mice (lacking H2S production), indicating that H2S also activated Nrf2 by Ssulfhydration of Keap1 . Besides, both ADT-OH and NaHS can enhance AMPK activation in brain regions where microglia cells are over-activated under LPS stimulation (Zhou et al., 2014). Therefore, these results support that H 2 S released from ADT or ADT-OH mediates its biological effects, but further studies are needed.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

ETHICS STATEMENT
The animal study was reviewed and approved by the Laboratory Animal Care and Welfare Committee of Central South University.

AUTHOR CONTRIBUTIONS
YL and CZ contributed to the conception and design of this study. CZ and NY conducted the experiments. MT, HC, WL, NY, and CZ contributed to the detection of liver free FAs and related gene levels. CZ and WL performed data processing and statistical analysis. CZ wrote the first draft of the manuscript. All authors contributed to the article and approved the submitted version.