Salvia-Nelumbinis naturalis improves lipid metabolism of NAFLD by regulating the SIRT1/AMPK signaling pathway

Nonalcoholic fatty liver disease (NAFLD) is the liver injury induced by metabolic stress, which is highly correlated with insulin resistance, oxidative stress, and genetic susceptibility. This disease ranges from nonalcoholic fatty liver (NAFL) with simple fatty infiltration of the liver to nonalcoholic steatohepatitis (NASH) with inflammation and hepatocyte ballooning in addition to diffuse steatosis, eventually leading to cirrhosis and hepatocellular carcinoma (HCC) [1]. In recent years, the prevalence of NAFLD increased quickly, afflicting around 25% of the population worldwide. Approximately 20% -30% of NAFL patients could develop into NASH, and 25% of them may further progress to cirrhosis [2].

The pathogenesis of NAFLD is complicated and has not been completely clear by far. Excess fatty acids generated in the liver or fluxed from adipose lipolysis, if their disposal by oxidation or release of VLDL is impaired or overwhelmed, can generate lipotoxic lipid species. Such lipids then cause mitochondria dysfunction, endoplasmic reticulum stress, inflammation, apoptosis, and activation of stellate cells. Thus dysfunctional lipid metabolism triggers subsequent liver injury [3, 4]. Recent studies have shown that AMP-activated protein kinase (AMPK) and Sirtuin1 (SIRT1) are the key enzymes responsible for energy homeostasis by regulating fatty acid metabolism [5, 6]. AMPK functions as an energy switch in regulating the synthesis and β-oxidation of fatty acids by regulating the associated gene expression and activation [7]. SIRT1 is an NAD⁺-dependent protein deacetylase, which functions as an important regulator of energy hemostasis and lipid metabolism in mammals [8]. SIRT1 activity relies on AMPK and was also shown to regulate AMPK activation in NAFLD [9, 10].

Lifestyle changes have shown benefit for NAFLD. However, it is challenging to sustain in a long term critically, and currently, there are no approved efficient drugs for the treatment of NAFLD and NASH [11]. In recent years, traditional Chinese medicine (TCM) has displayed effects in the treatment of NAFLD. Salvia–Nelumbinis naturalis (SNN) is the extract of a compound formula of Chinese medicine (also named Jiang Zhi Granule), composed of five medicinal herbs, namely, Salvia miltiorrhiza Bunge., Nelumbo nucifera Gaertn., Gynostemma pentaphyllum (Thunb.) Makino, Polygonum cuspidatum Sieb. et Zucc., and Artemisia capillaris Thunb. It has been used to treat NAFLD in China, with significant effects of alleviating hepatic steatosis with few side effects [12]. Previous in vivo studies have confirmed the effect of SNN on NAFLD, alleviating high-fat diet (HFD)-induced or methionine-choline-deficiency (MCD) diet-induced hepatosteatosis and lipotoxic liver injury as well as serum levels of transaminases and lipid in male C57BL/6 mice or SD rats [13‐16]. In vitro studies have also indicated that SNN could reduce lipid droplet accumulation and increase the resistance to damage of hepatocytes induced by free fatty acids [17, 18]. Moreover, the underlying mechanisms were partially unraveled, including improving insulin resistance, enhancing autophagy and anti-oxidative stress, and inhibiting transcription of liver X receptor α (LXR-α)-mediated sterol regulatory element binding protein-1c (SREBP-1c), et al [13, 15‐17].

The pathophysiology of NAFLD is regarded to be constituted by multiple parallel hits, leading to an imbalance of energy metabolism in the liver, mostly in the form of carbohydrates and fat [4, 19]. The compound formulae of TCM usually have the advantage of multi-target effects. Therefore, in the present study, through the intervention experiment of the HFD-induced NAFLD mice model and fatty acid-induced hepatocytes with SNN, we tried to investigate the mechanism of SNN based on the lipid metabolism-regulating pathway, SIRT1/AMPK. This will help in understanding the therapeutic effect of the traditional herbal formula and promoting its further clinical application.

HFD-induced mouse or rat model is commonly used for the research of pathogenesis and drug development of NAFLD due to its similar pathogenesis with human NAFLD. In this study, male C57BL/6J mice fed with HFD were used to establish the NAFLD model. changes in serum biochemical indexes and liver pathology were after the 4th, 8th, 12th, and 16th week, respectively. The dynamic monitored results showed that, with the time elongation, the body weight, liver weight, serum lipid, and transaminases of HFD-fed mice increased more seriously than those of the control mice. It has been reported that the prevalence of NAFLD is up to 50% in people with dyslipidemia. The serum levels of TC and LDL-c presented in this model were similar to the increase of serum cholesterol in people with NAFLD, which is closely related to NAFLD [22]. However, there is no obvious difference found in the serum TG between model and control mice for all the timepoint. This result was similar to our previous experimental result, demonstrating no change in serum TG level in the mice fed with HFD for 18 weeks [23, 24]. Another study for the NAFLD model also showed mild downregulated trend of serum TG in C57BL/6, CD-1, and 129Sv mice after HFD feeding for 9 weeks [25]. Elevated serum ALT indicating liver damage was also found in the model mice. More importantly, the liver steatosis area and degree of mice gradually aggravated with longer HFD, and at the end of the 16th week, slight inflammation and ballooning degeneration of hepatocytes even appeared in the liver tissue of model mice. Taken together, after a 16-week HFD, the mice became obese with hypercholesterolemia, hepatic lipid accumulation, and liver injury, confirming the development of NAFLD, which is consistent with the previously reported studies [26, 27]. Further, after another 6 weeks, the aforementioned indices increased more seriously at the end of the 22nd week of the experiment.

Using the HFD-induced NAFLD mice, the effect of SNN was investigated. The mice with SNN administration had lower body weight, liver weight, serum LDL-c, and ALT than those of the NAFLD model. In addition, hepatosteatosis, inflammatory cell infiltration, and ballooning in the liver tissues of model mice were significantly reduced by SNN. Consistently, the intrahepatic TG content was downregulated by SNN. These results suggested that SNN can ameliorate the lipid accumulation and liver damage of the NAFLD mice, which is consistent with studies using other HFD-induced NAFLD rodent models [13, 14, 16].

SIRT1 is an NAD⁺-dependent deacetylase, which can promote the deacetylation of lysine residues of various proteins. A large number of recent studies revealed that SIRT1 regulates lipid homeostasis through multiple nutrient sensors such as SREBP-1, AMPK, PGC1α, and PPARα [28‐31]. SIRT1 can enhance insulin sensitivity, regulate liver lipid metabolism, suppress oxidative stress, and reduce the inflammatory response. Studies have shown that decreased expression and/or activity of SIRT1 plays a key role in NAFLD development [32‐34]. Moreover, SIRT1 is negatively related to NAFLD degree. Its plasma values were found lower in severe NAFLD compared to simple mild hepatosteatosis [35]. Many factors may contribute to the down-regulation of the SIRT1 level. NAFLD hepatic tissues and FFA-treated HepG2 and Huh-7 cells presented miR-122 upregulation, which was found able to suppress Sirt1 expression via binding to its 3′-untranslated region (UTR) [36]. SUV39H2 expression could be induced by pro-NASH stimuli in the hepatocyte, mice, and human livers. It was found able to bind to the SIRT1 gene promoter and suppress SIRT1 transcription [37]. On the other hand, SIRT1 could be interacted with and subsequently degraded through ubiquitination by the gene related to energy in lymphocytes (GRAIL), which was upregulated in the livers of humans and mice with hepatic steatosis [38]. In the present study, 4-week HFD resulted in a higher hepatic level of SIRT1 than control mice, however, since the 8th week of a high-fat diet, the expression of this molecule in the liver of model mice was significantly lower than that of control mice. Besides the decreased expression of SIRT1, we also found increased total Acetylated-lysine in NAFLD mice, indicating less de-acetylation activity of SIRT1. Similarly, our result also showed that SUV39H2 was up-regulated in the mice liver of the NAFLD model group. The SNN treatment increased the expression level of SIRT1, but decreased the Acetylated-lysine and SUV39H2, suggesting SNN may upregulate SIRT1 expression by inhibiting SUV39H2.

AMPK is a metabolic sensor, composed of a catalytic (α) subunit and two regulatory (β, γ) subunits, and AMPK activation requires phosphorylation of the α subunit [39, 40]. AMPK is the major negative kinase regulator of ACC through phosphorylation to inhibit ACC activity. The latter molecule plays an essential role in inhibiting fatty acid oxidation through inhibiting carnitine palmitoyltransferase (CPT) and promoting fatty acid synthesis [41]. Studies have also shown that AMPK could down-regulate the expression and maturation of SREBP-1 [42], thus reducing the expression level of FASN and stearyl coenzyme A dehydrogenase-1 (SCD-1), thereby inhibiting the synthesis of fatty acids and triglycerides. AMPK can enhance the activity of PPARα, thereby increasing the expression of some key enzymes of lipid oxidation including ACOX, which plays a key role in promoting lipid oxidation degradation [43]. AMPK activation can inhibit the synthesis of de novo lipid and promote the oxidation of fatty acids. AMPK activity decreases in NAFLD, which can serve as an important target for treatment [44]. In our study, it was identified that AMPK activation had a higher trend in the liver tissues of mice with HFD feeding from the 4th week, but decreased with NAFLD progress. And the obvious difference was displayed between the model and control at the 16th week and the end of the experiment (22 w). The SNN treatment significantly upregulated the AMPK activation.

Both SIRT1 and AMPK were known to regulate each other and share many common target molecules, and the interaction between SIRT1 and AMPK could be reciprocal [45]. AMPK can increase the activity of SIRT1 by increasing the level of NAD⁺ [46]. SIRT1 activity relies on AMPK and was also shown to regulate AMPK activation in NAFLD [9, 10]. SIRT1 can activate AMPK by deacetylating its upstream regulator-LKB1 [47]. AMPK and SIRT1 have synergistic effects on energy metabolism and affect the pathogenesis of NAFLD. The suppression of SIRT1/AMPK resulted in reduced fatty acid utilization and abnormal lipid deposition in the liver [6, 45, 48]. In the present study, compared with control, the hepatic protein expression level of SREBP-1c was significantly increased, while the levels of p-ACC and ACOX1 decreased in the model group. SNN treatment downregulated SREBP-1c and FASN and upregulated p-ACC and ACOX1. There was an increasing trend of PPARα levels in the liver of mice with SNN treatment. These results in vivo suggest that SNN may affect lipid metabolism via regulating SIRT1/AMPK signaling. This was further verified by the in vitro experiments. SNN downregulated the lipogenesis gene and upregulated the gene for oxidation in AML-12 cells with FFA incubaton. However, the inhibition of SIRT1 or AMPK significantly diminished these effects and attenuated the improvement of SNN on FFA-induced steatosis of the hepatocytes. Therefore, the effect of SNN on the lipid metabolism of NAFLD is SIRT1/AMPK dependent.

A series of recent studies have shown that the effects of the active components in TCM herbs on NAFLD are associated with activating the SIRT1/AMPK pathway [49]. As one TCM extract, SNN contains a variety of compounds, some of which have been found to regulate lipid metabolism through the SIRT1/AMPK signaling pathway. In vitro and animal model studies have shown resveratrol reduced the hepatic accumulation of lipids and improves lipid and glucose metabolism. And evidence support that resveratrol activates SIRT1 via activation of AMPK, thereby inducing the deacetylation of SIRT1 targets, PGC-1a and FOXO1 [50]. Polydatin was proved able to prevent NASH via inhibition of oxidative stress and inflammation, as well as via regulation of multiple signaling pathways, including AMPK/LDLR, LKB1/AMPK, SIRT1-PGC-1α, etc [51]. One study demonstrated that emodin effectively ameliorated hepatic steatosis through the CaMKK-AMPK-mTOR-p70S6K-SREBP1 signaling pathway [52]. Another study also find that emodin was closely related to the regulation of AMPK signaling pathway which increases IR and fatty acid oxidation [53]. Accumulating evidence obtained from animal experiments proves that quercetin has beneficial effects on metabolism diseases. One study found a direct anti-lipogenic effect of quercetin exerted by inhibiting the de novo lipid (DNL) pathway by acting on the ACACA/AMPK/PP2A axis [54]. Resveratrol, polydatin, emodin and quercetin are the major components of SNN identified in mice serum and liver. Although many preclinical trials have shown the promising potential of these natural compounds against NALFD, however, corresponding human clinical studies on their effects remain scarce. But these research results suggest that the bioactive compounds such as resveratrol, polydatin, emodin, etc. in SSN may contribute to the therapeutic effects of SNN on NAFLD through the SIRT1/AMPK signaling pathway.

In summary, our findings demonstrate that the abnormality of the SIRT1/AMPK signaling pathway is involved in the dysfunction of lipid synthesis and degradation during NAFLD development. SNN administration could improve the lipid metabolism disorder and liver damage of NAFLD, and the regulation of the SIRT1/AMPK signaling pathway may contribute to the underlying mechanism. These results are beneficial to comprehending the therapeutic effect of the traditional herbal formula and further promote its clinical application.