Introduction

Diseases such as cancers [1,2,3,4,5,6,7,8] especially tumors [9,10,11,12] have caused too much trouble and cause stress/anxiety to human being [13, 14]. The way to prevent and treat have attracted great attention [15,16,17,18,19]. The non-alcoholic fatty liver disease (NAFLD), also known as metabolic associated fatty liver disease (MAFLD), is a metabolic syndrome characterized by the accumulation of massive amounts of fat in the liver. NAFLD commences with hepatic steatosis, commonly referred to as fatty liver, and may progress to non-alcoholic steatohepatitis (NASH) and, eventually, hepatic fibrosis (HF). It can be developed into hepatic sclerosis (HS) or even hepatocellular carcinoma (HCC) [20]. The prevalence of NAFLD has now reached up to 25% worldwide, posing a serious health risk as well as a considerable economic cost to society [21]. It is important to emphasize that there are no approved chemically synthesized drugs for the treatment of this condition [22]. Natural products have the advantage of treating this condition [23], but they have the limitations of unclear material bases and mechanisms of its action.

With increasing awareness of healthcare [24,25,26] and required high-standard living conditions [27,28,29], more advanced materials [30] and devices [31,32,33,34] have been designed and built [35]. Natural products such as bamboo [36,37,38] and lignin [39,40,41] have attracted more attentions due to their increasing usage in our daily lives [42,43,44,45,46]. Many value-added products such as polysaccharide have been produced from natural products [47,48,49,50,51,52]. There are many medicinal and culinary advantages to ginger, the root of the plant Zingiber officinale Roscoe. Ginger is not only a common spice, but also a traditional Chinese medicine with both medicinal and culinary uses. Ginger contains several phytochemicals, including flavonoids, phenolic acids, terpenes, etc., which have a wide range of biological activities, such as antioxidant, anti-inflammatory, antibacterial, and anti-obesity [53, 54]. As a constituent of ginger, gingerols are considered to be the most active constituent and they may be effective in treating metabolic diseases [55]. 6-gingerol is believed to be effective in treating non-alcoholic fatty liver disease [56]. Previous studies have shown that 6-gingerol may ameliorate hepatic steatosis caused by aging through multiple mechanisms [57], including the inhibition of de novo lipogenesis, enhancement of fatty acid oxidation, and reduction of oxidative stress. It is unknown whether other components of ginger can improve NAFLD on a pharmacological basis and in what ways, whereas gingerols are structurally similar and may exert similar effects. Research of this class of components can help to clarify the basis for ginger’s pharmacological effects, as well as the mechanism of its action in improving NAFLD, by providing insight into its material basis and mechanism of action.

Stearoyl-CoA desaturase 1 (SCD1), which is primarily expressed in liver and adipose tissue, is a microsomal enzyme bound to the endoplasmic reticulum. In addition to regulating fatty acid metabolism, this enzyme plays an important role in hepatic de novo lipid synthesis by catalyzing the conversion of saturated fatty acids (SFA) to monounsaturated fatty acids (MUFA) [58]. SCD1 catalyzes the generation of monounsaturated fatty acids to triglyceride (TG) by diacylglycerolacyltransferase-1 (DGAT1), which promotes hepatocyte steatosis [59]. The hepatocellular steatosis and lipid deposition characteristic of NAFLD are pathological features of the disease. Based on clinical studies, the activity of hepatic SCD1 is found to be significantly higher in obese patients with NAFLD in comparison to obese patients without NAFLD. Therefore, it seems likely that hepatic SCD1 activity is closely linked to NAFLD, and it has been demonstrated that hepatic-specific inhibition of SCD1 attenuates the progression of hepatic steatosis [60]. Studies have demonstrated that SCD1 is also involved in the oxidation of fats within cells and that silencing or inhibiting SCD1 expression increases intracellular fatty acid β-oxidation, lowers hepatocyte lipid levels, and ameliorates hepatocyte steatosis. Galmed Pharmaceuticals has independently developed a new oral drug, Aramchol (arachidyl amido cholanoic acid), which targets and inhibits SCD1. This drug has reduced liver fat significantly in people with NAFLD/NASH. The Phase III clinical trials [61] demonstrate that inhibiting SCD1 can be an effective target for the treatment of non-alcoholic fatty liver disease [62].

Ginger extract can significantly reduce SCD1 activity and ameliorated hepatic lipid deposition in hyperlipidemic hamsters [63], but the specific component responsible for these effects is unknown. This study used magnetic nanomaterial magnetite (Fe3O4) immobilized SCD1 for screening SCD1 inhibitors in ginger, and liquid chromatography with tandem/mass spectrometry (LC-MS/MS) for identifying the screened active ingredients. Molecular docking was used to examine the binding mode between the target component and SCD1. Free fatty acids (FFA)-induced normal human hepatocytes (THLE-3) cells were used to evaluate the effect of each compound on reducing lipid accumulation in hepatocytes. The underlying material basis for the pharmacodynamic effects and mechanism of action through which ginger enhances the condition of NAFLD were explored.

Materials and methods

General experimental procedures

Recombinant mouse SCD1 protein (CSB-CF319560MO) was obtained from Wuhan Huamei Bioengineering Co., Ltd. (Wuhan, China). 3-Aminopropyltriethoxysilane (APTES, 99%), oleic acid, and palmitic acid were purchased from Shanghai Aladdin Biochemical Technology Co., Ltd. (Shanghai, China). Glutaraldehyde (GA, 50%) was purchased from Macklin Biochemical Technology Co., Ltd. (Shanghai, China). The reagents required for cell culture were obtained from Gibco (Carlsbad, CA, USA). These sources provided the Cell Counting Kit-8 (CCK-8) reagents required from Solarbio (Beijing, China).

Plant material

The dried ginger slices were ordered from the company Sichuan Hongsheng Pharmaceutical Co., Ltd., originating from Sichuan, China, and bearing the lot number 20,221,201.

Extraction preparation

During the experiment described above, dried ginger slices were crushed and mixed with 70% ethanol at a material-liquid ratio of 1:20 (w/v). They were extracted three times at room temperature for 30 min each using ultrasonication (700 watts) [64]. To obtain ginger extract (GE), the extract was filtered and concentrated at 50 °C using a rotary evaporator Rotavapor R-100 (Büchi, Flawil, Switzerland) and was then freeze-dried using a vacuum freeze dryer Lab-1 C-50E (Biocool, Beijing, China) to obtain mulberry branch extract (MBE).

Synthesis of materials

The methods outlined in the preceding literature were adopted [65]. In brief, the hydrothermal method was utilized to synthesize magnetite (Fe3O4) magnetic nanoparticles using the previously reported method. The particles (0.2 g) were then aminated with APTES (1 mL) to yield Fe3O4-NH2. As a final step, glutaraldehyde was used to crosslink with the SCD1 protein to obtain Fe3O4@SCD1, which will be used for the following screening. For easy understanding, the details of the materials preparation and the subsequent property testing is shown in Scheme 1.

Scheme 1
scheme 1

The preparation of the materials and the subsequent testing steps

Characterization of materials

The synthesized materials were characterized using scanning electron microscopy (SEM), vibrating sample magnetometer (VSM), and Fourier transform infrared (FT-IR) spectroscopy [66]. The method involved infrared pressing of the samples with potassium bromide (KBr) and the magnetic properties were measured using a vibrating sample magnetometer at room temperature in the magnetic field range of ± 2 T. The functional group signals were measured in the range 4000 –500 cm− 1.

Rapid screening of active ingredients

Two aliquots of Fe3O4@SCD1 materials were introduced to a blank solvent (control) and a GE solution dissolved in phosphate buffered solution (PBS). The mixture was then incubated for two hours at room temperature. Until the magnetic separation had been achieved, the elution procedure was followed as described above to obtain the eluate. Finally, the eluent was determined by ultraperformance liquid chromatography-mass spectrometry (UPLC-MS), and a mass spectrum analysis was performed on the screened components.

UPLC-MS/MS analysis

Throughout the experiment, the Waters ACQUITY UPLCI-Class tandem with a Xevo G2-XS Q-Tof (Waters Corporation, Milford, MA, USA) was utilized for UPLC-MS/MS analysis.

Liquid phase conditions were maintained: The column used in this study was the ACQUITY BEH-C18 UPLC column (2.1 × 100 mm, 1.7 μm) at 40 °C at an injection volume of 1 µL at a column temperature of 40 °C. The mobile phase A comprised acetonitrile, while mobile phase B consisted of a 0.1% formic acid with water, with a flow rate set at 0.4 mL/min. In terms of elution conditions, the following were followed: 0–2 min, 5% A; 2–8 min, 20–55% A; 11–15 min, 55–90% A; 15–21 min, 90% A; 21–24 min, 90%-5% A.

The parameters for mass spectrometry determination are as follows: ESI positive/negative ion modes, 3.0 kV capillary voltage (Vcap), 120 ℃ source temperature, 40 V skimmer voltage, 400 ℃ desolvation temperature, 800 L/Hr desolvation gas flow, 30 V collision energy, 100–1200 scanning range (m/z), and automated secondary mass spectrometry data collection.

Data from mass spectrometry were acquired both in positive and negative modes, with scan ranges of 100–1200 molecular weights (m/z) and LC-MS/MS acquired automatically.

The parameters used to determine the mass spectrometry were as follows. The temperature at the source, skimmer voltage, desolvation gas flow, collision energy, and desolvation temperature were 120 °C, 40 Vs, 800 L/Hr, 30 V, and 400 °C, respectively.

Molecular docking

Molecular docking has shown its capability for examining the binding mode and others [67,68,69,70,71]. It was conducted using AutoDock Vina software [72] with the crystal structure of the Stearoyl-Coenzyme A Desaturase 1 protein (SCD1, PDB ID: 4YMK) obtained from the RCSB Protein Data Bank (http://www.rcsb.org). The six amino acids that are crucial for SCD1 and stearoyl coenzyme A binding have been identified as key amino acids [73]. In this interaction model, the center coordinate t is (-29.40, -24.97, 11.45), and small molecules are docked to the active site of the SCD1 protein to determine the interaction model. Pymol was used to display pictures based on the docking scores calculated for each compound.

Cell culture

In this study, normal human hepatocytes THLE-3 were grown in 1640 complete medium containing 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin double antibody, at an ambient temperature of 37 °C, in an incubator with 5% CO2.

Cell viability assay

Cell viability was assessed by CCK-8 method. Cells were plated at 1 × 104 cells per well into 96-well plates. After overnight incubation, each group was treated with a certain concentration of FFA (0, 25, 50, 75, 100, and 150 µM) or drug (0, 2.5, 5, 10, 20, 40, and 80 µM), with six replications in each group. Following 24 h of incubation, the medium was discarded. 100 µL of 1640 and 10 µL of CCK-8 reagent were added. After incubation for 1 h, the optical density (OD) value at 450 nm was measured using the Synergy H1 microplate reader (BioTek, VT, USA), and the cell viability was calculated to evaluate the cytotoxicity of FFA or drugs.

Lipid deposition model construction

The experiment was performed utilizing FFA (ratio of 2: 1, oleic acid: palmitic acid) as a model of lipid accumulation in hepatocytes, in which THLE-3 cells were seeded at a concentration of 1 × 105 cells per well in a 6-well plate. As soon as the cells had grown to an estimated 80%, FFA was added at a certain concentration within the safe concentration range to each group, with three replicates in each group, and the intracellular TG content was obtained after 24 h.

Determination of intracellular triglyceride content

Incubation of the treated cells was conducted for 24 h, followed by a three-time wash with sterile PBS buffer, followed by discarding the original medium. Each well was incubated with 500 mL of RIPA lysate for 10 min on ice. The lysate was scraped off with a spatula and centrifuged at 5000 rpm for 5 min to obtain the supernatant. Based on the instructions in the triacylglycerol assay kit (Nanjing JianCheng Bio, China) instruction, the TG content of the solution was determined, and the TG content of the sample was calculated based on the volume of lysate added. In addition to measuring the protein concentration of the lysate with the help of the BCA protein assay kit (Nanjing JianCheng Bio, China), the relative amount of TG in the cells was also measured by dividing the amount of TG in the sample (mg) by the amount of total protein (g).

Oil red O staining

The THLE-3 cells were digested with trypsin and well dispersed, after which 1640 was added to each well of the 12-well plate. The cells were then seeded at a concentration of 1 × 104 cells per well and growing on the glass slide. FFA was compared with a specific concentration of 6-gingerol and 10-gingerol in each well after 24 h of incubation. The medium was discarded after 24 h, followed by applying 4% paraformaldehyde for fixation for 15 min, and then washing with PBS three times, followed by the addition of oil red O for staining for 15 min, followed by three more rinses of PBS, then adding hematoxylin staining for 0.5 min, followed by three more rinses of PBS, finally, observing under the microscope and taking pictures.

Data analysis

Statistical analysis and organization of the experimental data were carried out using SPSS 26.0 software. The results were expressed as “mean ± SEM”. Statistical differences in data between groups were analyzed using a one-way ANOVA and Duncan’s multiple range test (P < 0.05).

Results

Characterization of materials

Due to its excellent magnetic properties [74,75,76,77], Fe3O4 is widely used as a substrate for immobilized enzymes. It can be seen in Fig. 1A that SEM can accurately reflect the morphological characteristics of the sample and that the synthesized Fe3O4-NH2 has a spherical shape with a diameter of about 200 nm. In general, VSM is considered an important tool for determining the magnetic properties of magnetic materials. As can be seen from the VSM data of Fe3O4@SCD1 (Fig. 1B), the material exhibits good paramagnetic properties with a saturation magnetization of 71.4 emu/g. A similar result is shown in Fig. 1C, which shows that Fe3O4@SCD1 is dispersed in solution and is capable of being enriched at the magnet end to be separated from the solution by applying a magnetic field. Based on the results of the FT-IR pattern (Fig. 1D), the characteristic signal of the Fe-O bond at 581 cm− 1 [78], the C = O bond stretching vibration at 1656 cm− 1 (signal of the protein amide bond) [79], and the peak of the stretching vibration of the -OH at 3201 cm− 1 indicates that the material has been functionalized successfully.

Fig. 1
figure 1

Characterization of Magnetic Nanomaterials: A. SEM image of Fe3O4-NH2; B. VSM image of Fe3O4@SCD1; C. Dispersion of Fe3O4@SCD1; D. FT-IR spectrum of Fe3O4@SCD1

Component analysis of SCD1 potential inhibitors in ginger extract

The components identified through screening were subjected to UPLC-MS/MS analysis, and their total ion current (TIC) plots were observed in the positive ion mode (Fig. 2). There was a significant concentration of compounds 1, 2, and 3 in the eluent, with retention times of 11.55, 12.58, and 13.65 min, respectively. It is noteworthy that compound 3 demonstrated the greatest response, suggesting that it may have a greater affinity for SCD1.

The mass spectral data and the cleavage pattern of compound 3 (Fig. 3C) indicated that [M + H]+ and [2 M + H]+ signals are located at m/z 351.2632 and 701.4868, respectively, while fragments (fragment ions) at m/z 333.2555 and 315.2455 correspond to [M + H-H2O]+ and [M + H-2H2O]+. The mass spectrometry analysis showed that the loss of water molecules by the precursor ion resulted in fragments of m/z 179.0848, 177.1049, and 137.0753. The strongest signal was observed for [M + H-H2O]+, where the major ion m/z 179.0848 was shown to be the fragmentation signal generated by the precursor ion after it lost the neutral alkyl component [CH3(CH2)9CH, 154 Da]. Additional fragmentation of CHCHO (42 Da) was observed at m/z 137.0753 as a consequence of further loss of CHCHO.

The cleavage patterns of compounds 1 and 2 were shown in Fig. 3A and B to be similar to those of compound 3 as indicated by their similarity in structure, which indicates that they may belong to compounds with similar structures to compound 3. Additionally, their cleavage patterns were consistent with those in other papers that have already been published in the literature. Consequently, the analytical results confirm compounds 1, 2, and 3 to be 6-gingerol, 8-gingerol, and 10-gingerol, respectively.

It is evident from the total ion chromatograms (TIC) of the eluate in negative ion mode (Fig. 4B) that compound 3 and compound 2 were the primary components in the eluate, with compound 1 having a weak signal. In addition, the strong signal at the retention time of 15.49 min was also observed in the control (Fig. 4C), indicating that this compound was not specifically specific for the binding of SCD1 to ginger. In addition, the mass spectra and cleavage patterns of compounds 2 and 3 in negative ion mode (Fig. 5) confirmed that these compounds are respectively 8-gingerol and 10-gingerol.

Prediction of the binding mode of potential active ingredients

The molecular docking technique analysis was conducted to investigate the docking sites of the target components concerning SCD1 (as shown in Fig. 6). It has been demonstrated that the binding energies of 6-gingerol (peak 1), 8-gingerol (peak 2), and 10-gingerol (peak 3) were − 9 kcal/mol, -8.8 kcal/mol, and − 9.1 kcal/mol, respectively, which further confirms that 10-gingerol has the strongest binding ability to SCD1. Since these three compounds have structural similarities, their binding sites with SCD1 are also similar. A notable difference between 10-gingerol and the other two compounds is the presence of more binding sites on Asn144 and Asp152 in the active pocket of SCD1 [73]. This suggests that 6-gingerol, 8-gingerol, and 10-gingerol have binding sites located in the catalytic pocket of SCD1 that are capable of directly binding to it.

Fig. 2
figure 2

Total ion chromatograms of ginger extract (A), eluent (B) in positive ion mode, and control (C)

Fig. 3
figure 3

Mass spectra and fragmentation patterns of compound 2 (A) and compound 3 (B) in positive ion mode

Fig. 4
figure 4

Total ion chromatograms of ginger extract (A), eluent (B), and control (C)in negative ion mode

Fig. 5
figure 5

Mass spectra and fragmentation patterns of compound 2 (A) and compound 3 (B) in negative ion mode

Fig. 6
figure 6

Predicted binding modes of 6-gingerol (A), 8-gingerol (B), and 10-gingerol (C) with SCD1

Target component ameliorates lipid accumulation in hepatocytes

Based on Fig. 7A, it has been determined that the optimal modeling concentration of FFA lies between 25 and 150 µM. Therefore, concentrations below 150 µM can be chosen for further development of the model. As shown in Fig. 7B, intracellular triglyceride levels have been determined after the exposure to varying concentrations of FFA for 24 h. The results indicate that intracellular TG content increases as the concentration of FFA increases from 12.5 to 50 µM. In particular, 100 µM of FFA resulted in a substantial rise in intracellular TG levels (P < 0.01). In light of the fact that the results of the 150 µM treatment group were not significantly different from those of the 100 µM treatment group, it was decided that 100 µM would be the optimal modeling concentration. It appears that the TG content of cells was consistent with those measured by oil red O staining (Fig. 7C), and the number of intracellular lipid droplets (red particles) was significantly increased in response to treatment with 100 µM and 150 µM of FFA compared with controls.

Fig. 7
figure 7

Determining the Optimal Molding Concentration: A. the following levels of cytotoxicity were observed in cells treated with different concentrations of FFA; B. the following levels of intracellular TG content were observed in cells treated with different concentrations of FFA; C. the following levels of oil red O staining were observed in cells treated with different concentrations of FFA (200×). Results are presented as means ± SD. ** P < 0.01

Initially, the cytotoxicity of each compound was determined before evaluating its ability to improve lipid accumulation in hepatocytes (Fig. 8A) shows that 6-gingerol (6-G) and 8-gingerol (8-G) did not significantly affect THLE-3 cell survival when applied in the range of 80 µM. Among them, 10-gingerol (10-G) had no significant effect on cell survival in the range of 40 µM but significantly affected cell survival at 80 µM (P < 0.01). Upon finding a tendency with 8-G and 10-G for cell survival to decrease with concentrations exceeding 40 µM, concentrations less than 40 µM were subsequently chosen for viability assay.

Fig. 8B illustrates the effects of each compound on intracellular TG content. All compounds lowered intracellular TG content, with the best effect at 10 µM concentration. When at 10 µM concentration, 6-G and 10-G reduced TG by percentages of 18.8% and 21.4%, respectively, which were more effective than 8-G (8.3%). Hepatocytes stained with oil red O (Fig. 8C) showed significant reductions in lipid accumulation by both 6-G and 10-G at 10 µM.

Fig. 8
figure 8

The effect of various compounds on hepatocyte lipid accumulation: A. The following levels of cellular activity were observed in cells treated with different compounds at different concentrations; B. the following levels of intracellular TG content were observed in cells treated with different concentrations of compounds; C. the following levels of oil red O staining were observed in cells treated with different compounds (400×). Results are presented as means ± SD. * P < 0.05, ** P < 0.01, “ns” means P > 0.05

Discussion

Rapid screening of enzyme inhibitors in mixtures is often performed by immobilizing enzymes with magnetic nanomaterials. This method does not require complex processes such as extraction and separation of mixtures and structural identification but uses liquid chromatography-mass spectrometry (LC-MS) directly to analyze the screened components, which is a faster and more efficient method than traditional methods [80, 81]. This strategy is applied in the present study, and the compounds found in ginger extracts that bind directly to SCD1 are mainly gingerols. Based on the mass spectrometry analysis, it was found that they had been cleaved at the MS (Fig. 4). There may be a reason for this because gingerols themselves do not have good thermal stability [82]. It yields an oxyketone product under conditions of elevated temperature and/or acidic environment, as a result of possessing an instability of the β-hydroxyl ketone functional group. It is susceptible to dehydration, consequently forming α, β-unsaturated ketone [83, 84]. Due to the high desolvation temperature of 400 ⁰C at the mass spectrometry inlet, the target components in the samples were susceptible to thermal dehydration during the mass spectrometry analysis. In the mass spectra, significant [M + H-H2O] and [M + H-2H2O] signals were observed. As a result of the rearrangement of these signals, smaller fragments are produced. Thus, this class of compositions should be prepared in a manner that prevents changes in structural components caused by environmental factors.

A variety of mechanisms are responsible for the improvement of NAFLD through gingerols, and 6-gingerol is one of the most effective constituents [85]. For instance, HNF4α/miR-467b-3p/GPAT1 Cascade [56] has an association with regulating lipogenesis, fatty acid oxidation, oxidative stress, and mitochondrial dysfunction [57]. In contrast, very few reports suggest that other components of ginger contribute to the improvement of NAFLD. The present study has demonstrated that ginger contains a new compound (10-gingerol) that increases hepatic lipid accumulation through the direct action of SCD1. The present study provides an additional level of insight into the mechanism, by which ginger improves NAFLD and also contributes to the discovery of new active ingredients that can improve NAFLD. It is necessary to undertake further studies on the exact mechanism, by which 10 gingerol improves NAFLD in animal experiments to determine its exact effect.

The abnormal changes in SCD1 have been linked to a wide range of diseases in several studies. There is an increase in SCD1 expression in a wide range of tumor cells, and altered expression is associated with increased proliferation, invasion, sternness, and chemoresistance [86]. The inhibition of SCD1 results in a reduction of lung tumor growth, indicating that gingerol analogs may also have antitumor properties [87]. Several recent studies have demonstrated that the combination of 6-gingerol and cisplatin induces apoptosis in ovarian cancer cells and inhibits angiogenesis more effectively than either drug alone [88]. There is evidence that 10-gingerol has anti-tumor activity through several pathways [89, 90], including PI3K/Akt and MAPK. Despite this, gingerols do not appear to have a direct effect on SCD1, which can also be the mechanism of anti-tumor action of this class of compounds.

There is no consensus as to whether suppression of SCD1 is beneficial in all cases. In mice that were treated with systemic SCD1 knockouts, ocular and dermal adverse events were observed, including strabismus and hair loss owing to ocular and dermal abnormalities [91], inhibitors of SCD1 that are systemically distributed cause adverse effects on the eyes and skin on a mechanistic basis [92]. When mice are deficient in SCD1, chronic inflammation is exacerbated and atherosclerosis occurs [93]. It is therefore possible to further investigate in vivo the uptake and distribution of SCD1 inhibitors and create structural modifications or alterations that will enhance their liver targeting and improve their inhibitory activity against SCD1 [94].

Conclusions

In this study, by constructing Fe3O4@SCD1 and analyzing and detecting these components in combination with UPLC-MS/MS, three potential inhibitory components of SCD1 were identified and screened from ginger extracts in this study. According to the molecular docking results, all three compounds were capable of binding to SCD1, with 10-gingerol having a stronger affinity for SCD1 than 6-gingerol and 8-gingerol. The activity assay showed that all three compounds ameliorated lipid accumulation in hepatocytes, with 10-gingerol having the strongest effect. This is first reported as ameliorating fatty liver. Gingerols are shown to be the main compounds in ginger to inhibit SCD1. This has important implications for understanding ginger’s mechanism of action in alleviating NAFLD and its use in the treatment of this disease. Further investigation is necessary to determine the precise efficacy of 10-gingerol in the treatment of NAFLD.