The Cholesterol-Modulating Effect of the New Herbal Medicinal Recipe from Yellow Vine (Coscinium fenestratum (Goetgh.)), Ginger (Zingiber officinale Roscoe.), and Safflower (Carthamus tinctorius L.) on Suppressing PCSK9 Expression to Upregulate LDLR Expression in HepG2 Cells

PCSK9 is a promising target for developing novel cholesterol-lowering drugs. We developed a recipe that combined molecular docking, GC-MS/MS, and real-time PCR to identify potential PCSK9 inhibitors for herb ratio determination. Three herbs, Carthamus tinctorius, Coscinium fenestratum, and Zingiber officinale, were used in this study. This work aimed to evaluate cholesterol-lowering through a PCSK9 inhibitory mechanism of these three herbs for defining a suitable ratio. Chemical constituents were identified using GC-MS/MS. The PCSK9 inhibitory potential of the compounds was determined using molecular docking, real-time PCR, and Oil red O staining. It has been shown that most of the active compounds of C. fenestratum and Z. officinale inhibit PCSK9 when extracted with water, and C. fenestratum has been shown to yield tetraacetyl-d-xylonic nitrile (27.92%) and inositol, 1-deoxy-(24.89%). These compounds could inhibit PCSK9 through the binding of 6 and 5 hydrogen bonds, respectively, while the active compound in Z. officinale is 2-Formyl-9-[.beta.-d-ribofuranosyl] hypoxanthine (4.37%) inhibits PCSK9 by forming 8 hydrogen bonds. These results suggest that a recipe comprising three parts C. fenestratum, two parts Z. officinale, and one part C. tinctorius is a suitable herbal ratio for reducing lipid levels in the bloodstream through a PCSK9 inhibitory mechanism.


Introduction
Blood cholesterol levels of total cholesterol and low-density lipoprotein (LDL) cholesterol are both major risk factors for coronary heart disease (CHD). Reduced total and LDL cholesterol levels have been shown to decrease the risk of coronary heart disease.
The most given lipid-lowering drug is statins, which potently inhibit 3-hydroxy-3methylglutaryl-coenzyme A (HMG-CoA) reductase, the enzyme that decreases the biosynthesis of cholesterol [1][2][3]. This results in intracellular cholesterol depletion and subsequent upregulation of low-density lipoprotein receptors (LDLRs) expression on hepatocytes and enhanced clearance of LDL from blood circulation via the sterol regulatory elementbinding protein (SREBP) pathway. Additionally, proprotein convertase subtilisin kexin type 9 (PCSK9), a member of the subtilisin-related serine protease family, has been identified as a critical regulator of low-density lipoprotein (LDL) metabolism, and inhibitors of PCSK9 are currently being investigated for their ability to lower circulating LDL via binding to its epidermal growth factor-like repeat (EGF-A) of LDLR [4][5][6]. Secreted PCSK9, a domain found in hepatocytes, binds to LDLR and promotes its lysosomal degradation in cells [7,8].
PCSK9 deficiency leads to a more significant number of cell surface LDLRs, and enhanced hepatic LDLR expression leads to improved plasma LDL clearance, protecting against cardiovascular disease (CVD). As a result, finding a new antihyperlipidemic drug that targets PCSK9 expression is a top priority in antihyperlipidemic research. Reducing PCSK9 transcription is a potential technique for lowering LDL. Thus, we set out to find a new recipe that inhibits PCSK9 transcription to promote plasma cholesterol-reduction effects via their effect on LDLR transcription. The new herbal recipe that induced LDLR expression may be a useful technique for treating hyperlipidemia. Complementary and alternative medicine has been utilized to control cholesterol levels and improve heart health; therefore, increasing LDLR expression from herbal drugs might be a useful antihyperlipidemic method. In addition, the use of various herbs as medicinal compounds will help to improve the effectiveness of the treatment.
Ginger (Zingiber officinale Roscoe), most commonly known as ginger, is a spice and flavoring ingredient used in cuisines worldwide [16]. For thousands of years, it has been used as a spice and for medicinal purposes. Its usage is attested in ancient Sanskrit and Chinese manuscripts, as well as in Arabic, Roman, and Greek medical literature [17]. Z. officinale is regarded as a promising medication in Ayurveda due to its efficacy as a digestive stimulant, antiasthmatic, and rubefacient [18]. It is cultivated commercially in India, China, Thailand, Australia, South Africa, and Mexico. Antioxidant activity [19][20][21] has been reported in vitro for Z. officinale aqueous and organic solvent extracts. A combination of Z. officinale and garlic [17] was proven to have hypoglycaemic and hypolipidemic effects in albino rats. The previous research [22] has demonstrated that ethanolic Z. officinale extract has considerable antihypercholesterolemic action in cholesterol-fed rabbits. It should be emphasized that Z. officinale's efficacy in lowering cholesterol levels is favorable and that its usefulness should be investigated when paired with other herbs.
Safflower (Carthamus tinctorius L.) is an oil-producing crop that belongs to the Compositae or Asteraceae family. In Thailand, it is called Kamfoi, whereas, in China, it is called zang hong hua. C. tinctorius is a multifunctional crop that has been produced in Thailand and other areas of the world for generations for a variety of purposes. It is a critical plant since it provides an alternate supply of oil. C. tinctorius research and development continue to receive little attention [23]. However, it can grow in a wide variety of environmental conditions with very high yield potential and has a variety of uses for the various plant components. However, some researchers [24,25] have reported that C. tinctorius contains linoleic acid, an unsaturated fatty acid, which is widely known and helps decrease blood cholesterol levels.
In addition, all three types of herbs-C. fenestratum, Z. officinale, and C. tinctorius-contain primary metabolite and secondary metabolite. In addition, each country has a wide range of uses as shown in Table 1.  [26] oleoresin, phenolics, zingiberene, gingerols, shogaols, aromatic alcohol, and terpenoids [27] It is distributed all over the world, such as in European countries, America, China, Japan, and India [28] with the following benefits: • reducing effect on blood lipids [29] • curing heart problems, treating stomach upset, diarrhea, headaches, and cough or nausea [30] • treating digestive problems [30] • Antibacterial agent [31] • Chemopreventive effect [32] • vomiting in motion sickness [33] • Use both fresh and dried preparation of rhizome for medicinal use [30] • Steam distillation/supercritical CO 2 extraction for essential oil [31] C.fenestratum carbohy-drate, lipids, amino acids, and vitamins [34] alkaloids, tannins, saponins, flavonoids, phenolic compounds [35] It is distributed all over the world, such as in Sri Lanka, India, and Thailand with the following benefits: antidiabetic, diuretic, cholesterol lowering, anticancer, anti-inflammatory, antifungal, antihelmintic, antioxidant, and antimicrobial effects [36,37] Use stem and dried preparation with solvent extractions such as Although all three herbs have been examined for their lipid-lowering properties, none have been combined to create a lipid-lowering recipe. Therefore, in this study, new formulations from these herbs were investigated for lipid-inhibiting activity through mechanisms such as HMG-CoA, SREBP, PCSK9, and LDLR mRNA levels using molecular docking and in vitro studies. Then, the proportion of herbs in the recipe will be determined to be suitable for reducing lipid in the bloodstream.

Extraction and Isolation
Plant materials were washed and dried at 50 • C until reaching a stable weight, then ground into a powder material and prepared for extraction method.

Water Extraction
The powdered herb (400 g) was mixed with 1 L of warm deionized water. On a hot plate, the herb solution was heated to 100 • C for 15 min. Another 1000 mL of hot water was added to the solution because the herb absorbed the water. The final solution was boiled until only one-third of the solution remained. Prior to freeze-drying, the solution was filtered using Whatman No. 1 filter paper and stored at −20 • C. Freeze-dryer (Eyela FDU-2100, Bohemia, NY, USA) was used to lyophilize the frozen samples.

Ethanol Extraction
Individually, 400 g of C. fenestratum stem, C. tinctorius flower, and Z. officinale rhizomes were extracted with ethanol for three days using the maceration procedure. The filtrate was collected using Whatman No. 1 filter paper and evaporated using a rotary evaporator to obtain a viscous ethanolic extract (Heidolph Basic Hei-VAP ML, Schwabach, Germany). The maceration procedure was then performed twice more. Each herb's remaining ethanol was evaporated further in a vacuum drying chamber (Binder VD 23, Tuttlingen, Germany) until a stable weight was obtained.

GC-MS/MS Analysis
Scion 436 GC Bruker model performed GC-MS/MS analyses to analyze the material at a 3 mg/mL concentration. The GC-MS/MS separation of the compounds was performed with a 30-m fused silica capillary column (0.25 mm internal diameter, 0.25 µm thickness). The carrier gas was helium gas (99.999 percent) with a constant flow rate of 1 mL/min and an injection volume of 10 µL. (split ratio of 10:1). The injector was heated to 250 • C, while the ion source was heated to 280 • C. The oven temperature was kept at 110 • C for 2 min, increased to 280 • C at 5 • C/min, and then kept isothermal at 280 • C for 9-min, for a total GC run duration of 60 min. The mass analysts by ionization energy of 70 eV with 0.5 s interval scan were designed, with fragments ranging from m/z 50 to 500 Da. The intake temperature was set to 280 • C, while the source temperature was set at 250 • C. By comparing the average peak area of each component to the total areas, the relative fraction of each component was computed. MS Workstation 8 was used for handling mass spectra and chromatograms. The chemical components were identified using the NIST Version 2.0 library database of the National Institute of Standards and Technology (NIST).

Treatment of HepG2 Cells
The ATCC (Manassas, VA, USA) provided the human hepatocellular carcinoma HepG2 cell line cultured in DMEM supplemented with 10% fetal bovine serum (FBS). The cells were seeded in 96-well plates with 5 × 10 4 cells/mL in a normal serum medium for 24 h before being changed to DMEM without FBS overnight. For an additional 24 h, cells were treated with extracts of the C. fenestratum, Z. officinale, and C. tinctorius, as well as a recipe of C. fenestratum (3 parts), Z. officinale (2 parts), and C. tinctorius (1 part) extracted with water and ethanol at concentrations ranging from 10 to 400 µg/mL prior to cell viability testing, real-time PCR, and oil red O staining.

Cell Viability Analysis
An MTT assay was used to measure cell viability. Briefly, cells were treated as described above, then incubated for 4 h at 37 • C with a 1 mg/mL MTT solution [43,44]. The purple formazan crystals were dissolved in DMSO when the medium was removed. Cell viability was measured by absorbance at 550 nm of the microplate reader (Metertech M965, Taipei, Taiwan).

Quantitative Reverse Transcription PCR (RT-qPCR) Analysis
The total RNA mini kit (Geneaid, Taipei, Taiwan) was used to isolate total RNA from HepG2 cells. Using an iScript Mastermix (Bio-Rad, Hercules, CA, USA), a quantified 1 µg sample of total RNA was converted to cDNA. The primers for specific genes are listed in Table 2 using the Luna Master Mix. The level of mRNA expression was evaluated using a Quanti-Studio 3 (ThermoFisher, Waltham, MA, USA) according to the manufacturer's guidelines. To compare the groups, 2 −∆∆CT values were used, with GAPDH (glyceraldehyde 3-phosphate dehydrogenase) acting an endogenous control [45]. Table 2. List of real-time PCR primer sequences.

Gene
Forward Primer Reverse Primer

Oil Red O Staining
Ice-cold PBS rinsed the fasting-induced steatosis in HepG2 cells before being fixed by ice-cold 10% formalin for 30 min. The cells were then rinsed with distilled water and stained for 30 min at room temperature with an Oil Red O working solution to generate stain lipid droplets [46]. An optical microscope was used to study and photograph the cells (Ziess AX10, Carl Zeiss, Jena, Germany). Lipid content was also determined by dissolving Oil red O in isopropanol and measuring using a microplate reader at a wavelength of 500 nm [15].

Molecular Docking
The crystal structures of PCSK9 and HMGCR with the PDB codes 6u26 [47] and 2r4f [48] were utilized. Autodock [49] was used to optimize the protein. The missing hydrogens were inserted throughout the optimization step. The final proteins were given Kollman unified atom charges and solvation parameters. Table 3 shows the grid position and size reflecting the whole protein during the docking process. Following GC-MS/MS analysis, the 3D structures of the top 5 high yielding compounds in C. tinctorius, C. fenestratum, and Z. officinale were chosen for docking, while positive docking controls were Alirocumab [50] and Lovastatin [51] for PCSK9 and HMGCR, respectively. All 3D structures were obtained from PubChem (https://pubchem.ncbi.nlm.nih.gov, accessed on 2 October 2021). All structures were optimized before molecular docking. Open Babel was used to add hydrogen atoms to every structure and all structures were optimized by Arguslab through semi-empirical Parametric Method 3 (PM3). Molecular docking was utilized to explore protein-ligand binding. Arguslab and Autodock were used for this docking study. In the beginning, the Arguslab engine was used for docking. The scoring function was set in default parameters. The accuracy of docking was set to regular. All docking was confirmed with Autodock3 through the Lamarckian genetic technique to ensure reliable results. The following are the optimal autodocking run parameters: number of GA runs: 50; population size: 200; and all other run parameters: default [44,52].

Binding Site Analysis
The structure of the compounds that resulted in lower binding energy to the targeted proteins than the standard drug was taken to visualize the binding characteristics by Discovery Studio. The ligand-protein bindings were presented as 2D and 3D. To identify the structure binding protein, the binding position was compared through CavityPlus (http://www.pkumdl.cn/cavityplus, accessed on 2 November 2021).

Statistical Analysis
The tests were carried out at least three times except molecular docking, and the results are shown as the mean ± standard deviation. SPSS 12.0 (SPSS Inc., Chicago, IL, USA) was used to perform the statistical calculations. The data were evaluated using a one-way ANOVA with Dunnet's post hoc test, with a p-value < 0.05 considered statistically significant.

GC-MS/MS Analysis
The active compounds of the herbs extracted with water and ethanol were analyzed with GC-MS/MS. In this study, the five most active compounds were selected and classified into three groups: (1) the most common, which were equal to or greater than 10%; (2) the moderately common were those that were greater than 1% but less than 10%; and (3) rare compounds are substances found less than 1% of the time, which are then chosen to study binding by molecular docking. The active compounds in each herb areshown in Tables 4-6.  The water extracted from C. tinctorius contained about 17 different compounds. Benzofuran, 2,3-dihydro-, with a molecular weight of 120 and a chemical formula of C 8 H 8 O, had the most remarkable peak area percent of 23.24 among the seventeen compounds detected. The second most significant peak was found with 3-Isopropoxy-1,1,1,7,7,7-hexamethyl-3,5,5-tris(trimethylsiloxy)tetrasiloxane, with a molecular weight of 576 and a chemical formula of C 18 H 52 O 7 Si 7, with a summative peak area percent of 21.23. The following compounds of 3,4-Dihydroxyphenylglycol, 4TMS derivative; 4H-Pyran-4-one, 2,3-dihydro-3,5-dihydroxy-6-methyl-; and Cyclohexasiloxane, dodecamethyl-had moderate peak area percent. Their respective values of peak area were 8 Table 4 and Figure S1. The water-extracted C. fenestratum contained about 43 different compounds. Tetraacetyl-d-xylonic nitrile with a molecular weight of 343 and a chemical formula of C 14 H 17 NO 9 had the most significant peak area percent of 27.92 among the forty-three compounds detected. Inositol, 1-deoxy-with a molecular weight of 164 and a chemical formula of C 6 H 12 O 5 , had the second greatest peak, with a summative peak area of 24.89. The following compounds of d-Gala-l-ido-octonic amide, Thieno[2,3-b]pyridine,3-amino-2-(3,3-dimethyl-3,4-dihydroisoquinolin-1-yl)-4,6-dimethyl-, and Megastigmatrienone had moderate peak area percent. Their respective values of summative peak area were 9.94, 5.87, and 5.56. C 8 H 17 NO 8 /255, C 20 H 21 N 3 S/335, and C 13 H 18 O/190 are their chemical formulas and molecular weights. The compounds with the lowest peak area percent are presented in Table 5 and Figure S3.
The water-extracted Z. officinale contained about 42 different compounds. With a molecular weight of 194 and a chemical formula of C11H14O3, 2-Butanone, 4-(4-hydroxy-3-methoxyphenyl)-had the greatest peak area percent of 38.21 among the forty-two compounds detected. The following compounds of (1S,5S)  Table 6 and Figure S5.

Determination of Maximum Dose for HepG2
The cytotoxicity of these herbs-C. fenestratum, Z. officinale, and C. tinctorius-extracted with water and ethanol from concentrations of 10-400 µg/mL were investigated in HepG2 cells by MTT assays. The findings revealed that all herbs extracted with water or ethanol at concentrations less than 50 µg/mL were harmless to HepG2 cells (cell viability >80%). In Figure 1, water extraction of the C. fenestratum, Z. officinale, and C. tinctorius at 50 µg/mL resulted in HepG2 cell survival rates of 88.16%, 90.19%, and 97.28%, respectively. Furthermore, ethanol extraction of C. fenestratum, Z. officinale, and C. tinctorius at 50 µg/mL resulted in cell survival of 103.63%, 82.75%, and 102.71%, respectively. As a result, the maximum dosage of those herbs was indicated for further research at 50 µg/mL. From the experiement, it was found that Z. officinale extracted with ethanol had the highest toxicity. Concentration values calculated using the fitting curve showed that the maximum concentration of Z. officinale extracted with ethanol that made HepG2 cells non-toxicity was 54.16 ± 3.90 µg/mL. In addition, The MTT assay was used to assess the safety of this recipe. It was revealed that a 3:2:1 ratio of C. fenestratum, Z. officinale, and C. tinctorius could be safely used at concentrations up to 100 µg/mL in this recipe. The previous study [53] on the correlation between SREBP2 and PCSK9 has indicated that inhibiting transcriptional activation of the sterol regulatory element binding protein 2 (SREBP2), which regulates PCSK9, increases LDLR expression, as seen in Figure 2. It was discovered that inhibiting SREBP2 expression enhanced LDLR activation. C. fenestratum extracted with water and ethanol has lipid-lowering activity through upregulating hepatic LDLR. Among three herbs with two types of extraction, this study found that the most effective way to upregulate LDLR expression by up to 23.12-fold was to treat with waterextracted C. fenestratum, followed by water-extracted Z. officinale, which increased the expression of LDLR mRNA by up to 9.09-fold. From LDLR mRNA, the number of LDLR expressions on the surface of hepatocytes is a significant factor [54]. Water-extracted C. fenestratum showed the most significant LDLR mRNA expression in HepG2 cells, followed by ethanol-extracted C. fenestratum, water-extracted Z. officinale, and ethanol-extracted C. tinctorius. The reduction of PCSK9 mRNA expression is the primary cause of LDLR mRNA expression, as seen in Figure 2. Although Z. officinale's potency is less effective at inhibiting PCSK9 than the C. fenestratum, Z. officinale extract was most effective at suppressing HMGR mRNA expression, as shown in Figure 2. Therefore, the presence of Z. officinale in the recipe can reduce the production of lipids from the liver, resulting in lowering blood lipids. In Thai traditional medicine, in addition considering the effectiveness of treatment with main and assistance drugs, it is also essential to add an herb that makes it more appetizing by adjusting the color. Therefore, C. tinctorius, which gives it its reddish-orange color and is used as a lipid-lowering herb [55], is used to improve its color.

Effect of Lipid Deposition in HepG2
According to the lipid staining with Oil red O examination, the total lipid in HepG2 cells following treatment with water and ethanol extraction of the C. fenestratum was 0.95 and 0.77 folds; C. tinctorius was 0.80 and 0.86 folds; Z. officinale was 0.78 and 0.73 folds, and the recipe was 0.61 and 0.48 folds, respectively. We found that treating HepG2 cells for 24 h with a recipe containing C. fenestratum, Z. officinale, and C. tinctorius had a strong synergistic effect, causing a significant reduction in lipid deposition when compared to individual herbs. Furthermore, these herbs extracted with ethanol were discovered to play an essential role in lowering the quantity of lipid accumulated in the HepG2 cell. The low lipid accumulation in HepG2 cells was due to the suppression of lipid synthesis, which resulted in a reduction in the quantity of lipid stained in the HepG2 cells.
In this experiment, Z. officinale exhibited more significant inhibition of HMGCR mRNA than lovastatin (2.5 times) [56] through 0.51-and 1.34-fold increases in HMGCR mRNA expression in ethanol and water extracts, respectively, compared to the control. In addition, when comparing the HMGCR mRNA inhibition of the extracts with statins, it was found that all herbal extracts inhibited HMGCR mRNA better than all statins. The inhibition value of herbal extracts ranged from 0.52-7.69-fold. The results also compared statins such as simvastatin, pravastatin, fluvastatin, atorvastatin, and rosuvastatin, which can induce HMGCR mRNA expression by up to 15-, 12-, 11-, 9-, and 17-fold in order [56]. The HMGCR mRNA expression found that the three herbal extracts had better properties in inhibiting lipid formation than statins.
Statins have good inhibitory properties in the production of lipids from the liver. Therefore, Z. officinale with a mechanism of action that inhibits HMGCR mRNA expression is also effective in inhibiting lipid synthesis. As a result, the lipid accumulation in HepG2 cells was lower than in other herbs, as shown in Figure 3. However, the large amount of lipid accumulation in the HepG2 cells of C. fenestratum results from most of the compounds suppressing the PCSK9 expression, which results in increased LDLR expression. However, it has little effect on the expression of HMG-CoA reductase (HMGCR). This causes more lipid to be absorbed into HepG2 cells. According to Thai traditional knowledge, the recipe composition is divided into three parts: the main drug, the assistance drug, and the servant drug. Therefore, the main drug was classified as the C. fenestratum in the highest proportion in this study. After all, it was the effect that needed to absorb lipid to the liver from the bloodstream, followed by Z. officinale as an assistance drug because it has properties to inhibit the production of lipid from the liver, and C. tinctorius as the servant drug, which helps to adjust the color of the recipe to make it more appetizing. Figure 4 and Table 10 show that PCSK9 has three pocket-binding sites: strong binding sites, medium binding sites, and low binding sites. Figure     In conclusion, extracts of C. fenestratum and Z. officinale with water effectively inhibit PCSK9 at the strong binding site, resulting in the most effective inhibition of PCSK9. It was found that the extract could bind to PCSK9 in multiple pocket-binding sites, resulting in combinational inhibition efficiency [57]. After examining the active compounds in each herb via GC-MS/MS, the constituents of the active compounds were identified. The top five compounds were studied through molecular docking to determine that compounds PCSK9 and HMGCR exhibit protein-binding activities. The molecular docking binding studies showed that the effect was consistent with real-time PCR.

Molecular Docking for the Top 5 Highest Amounts of the Compound from Each Herb
In Table 11, the binding between the active ingredients in the herbal aqueous extract and PCSK9 via Arguslab and Autodock showed that approximately 64.24% C. fenestratum including Tetraacetyl-d-xylonic nitrile, Inositol, 1-deoxy-, Thieno[2,3-b]pyridine, 3-amino-2-(3,3-dimethyl-3,4-dihydroisoquinolin-1-yl)-4, 6-dimethyl-, Megastigma-trienone binds the most to PCSK9 as it was able to bind to PCSK9 at a lower binding energy than the Alirocumab (standard drug). In Figures 5-7, the highest number of compounds found in C. fenestratum are 1) Tetraacetyl-d-xylonic nitrile (27.92%). It strongly binds to PCSK9, forming up to six hydrogen bonds with the amino acids HIS643, VAL644, ARG495, and TRP566. 2) Inositol, 1-deoxy-(24.89%) can bind the PCSK9 with different amino acids compared to Tetraacetyl-d-xylonic nitrile-PCSK9 binding. It can form up to five hydrogen bonds with the amino acids TRP461, ALA649, VAL435, and ASN439.  hydrogen bond with  the amino acids TRP461, LEU436, ASP360, ARG458, ALA649, ASP651, and THR469. The compound number of C. fenestratum extracts that can bind to PCSK9 is larger than the compound number of Z. officinale extracts, resulting in the water extract of C. fenestratum having a better inhibition effect than Z. officinale. In comparison, C. tinctorius 's active compounds have poor binding to PCSK9 because it contains only two compounds: 3,4-Dihydroxyphenylglycol, 4TMS derivative (8.94%), and Cyclohexasiloxane, dodecamethyl-(6.96%), which were found to total just 15.9%, resulting in poor inhibition of PCSK9. These compounds formed very few hydrogen bonds with PCSK9 binding compared to the two herbs mentioned above. Therefore, the preparation of the traditional recipe [58] suggested that the main drug with an excellent inhibitory effect in the highest proportion is C. fenestratum (3 parts), the assisting drug (2 parts) is Z. officinale, and the flavorful herb is C. tinctorius (1 part). Table 11. Energy binding and phytochemical inhibition constants of herbal extracts with water at the binding sites of PCSK9 from ArgusLab and Autodock analysis and quantification of each compound through GC-MS/MS analysis.

No.
Herb Compound Name     In Table 12, the binding of active compounds in herbs extracted with ethanol and PCSK9 studied via Arguslab and Autodock showed that compounds of Z. officinale had a 71.62% inhibitor to PCSK9 as compared to C. fenestratum containing a total active inhibitor of 47.04%, thus resulting in better inhibition to PCSK9 of Z. officinale than C. fenestratum when extracted with ethanol. The results are consistent with the real-time PCR results. It was concluded that the most effective inhibitor of PCSK9 was herbal extracts in water because in water extracts, it was found that the active compounds in C. fenestratum and Z. officinale extracts are 64.24% and 61.3%, respectively. By comparison, the herb extracts in ethanol provide active C. fenestratum and Z. officinale compounds at 47.04% and 71.62%, respectively. Therefore, when combining the active compounds for PCSK9 inhibition, C. fenestratum and Z. officinale suggest the best extraction in the water extract. In addition, studies on the inhibition of HMGCR through Arguslab and Autodock showed that no herbal extract was more effective at inhibiting HMGCR than lovastatin (positive control). The study in Tables 13 and 14 found that most of the compounds in Z. officinale had good efficacy in inhibiting HMGCR compared to extracts of C. fenestratum and C. tinctorius. The results are consistent with the effect of real-time PCR. Therefore, the mechanism of HMGCR affecting lipid formation can be best suppressed with Z. officinale extract and is classified as an assistance drug in this recipe.      Interestingly, the aqueous extract of C. fenestratum contained only one substance, megastigmatrienone. The binding of SREBP2 was lower than −10 kcal/mol, but the inhibition efficiency was higher in the ethanol extraction. This is because there are two active substances that caFn inhibit SREBP2 using energy below −10 kcal/mol: Cyclopropanetetradecanoic acid, 2-octyl-, methyl ester and Megastigmatrienone. The results are also consistent with RT-PCR regarding the expression of SREBP2.

GC-MS/MS
In conclusion, the extracts with the best SREBP2 inhibition were ranked from highest to lowest efficiency. In the following order, Z. officinale, C. fenestratum, and C. tinctorius extracts were extracted, respectively, and it was found that the ethanol extract had a better inhibitory effect than the aqueous extract.

Discussion
High levels of cholesterol are a significant risk factor for atherosclerosis and cardiovascular disease. Reducing the blood lipid profile may aid in the treatment of high levels of cholesterol-related diseases and disorders, including metabolic syndrome. Statins are medications that can lower cholesterol in a blood vessel and should be taken by most individuals. However, even after taking statins, the lipids in the blood in some individuals remained high [59]. Statins merely enhance the LDLR expression. LDLR destruction stays high if PCSK9 expression is still high [7]. Even though PCSK9 inhibition is beneficial for lipid reduction, the striking benefit achieved with only statin treatments in patients with a wide range of cholesterol levels cannot be attributed to their cholesterol-lowering effect. Therefore, inhibiting PCSK9 expression is crucial for improving lipid reduction.
In this study, the lowering cholesterol activity of three plants, C. tinctorius, C. fenestratum, and Z. officinale, as well as the potential molecular mechanisms involved in their lowering cholesterol activity, were investigated in the human liver cell line HepG2 by using molecular docking and RT-qPCR. Furthermore, we proved that combining these plants by making three parts C. fenestratum (primary herb), two parts Z. officinale (support herb), and one part C. tinctorius (coloring herb) significantly reduced lipid accumulation in hepatocytes by investigating Oil red O staining.
According to these findings, water-extracted C. fenestratum was the most effective at downregulating PCSK9 mRNA in HepG2 cells, followed by ethanol-extracted Z. officinale, water-extracted ginger, and water-extracted C. tinctorius. PCSK9 expression was reduced, which increased LDLR expression. Water-extracted C. fenestratum exhibited the most significant induction of LDLR expression, followed by water-extracted Z. officinale and water-extracted C. tinctorius. Further GC-MS/MS analysis of active compounds for these herbs revealed that excellent inhibition of lipid deposition depended on the efficacy of binding to target proteins and the number of chemical compounds present in the herb. Studies have shown that the highest number of compounds found in the C. fenestratum are the following: (1) Tetraacetyl-d-xylonic nitrile (27.92%). It binds strongly to PCSK9, forming up to six hydrogen bonds with the amino acids HIS643, VAL644, ARG495, and TRP566. (2) Inositol, 1-deoxy-(24.89%) can bind the PCSK9 with different amino acids compared to Tetraacetyl-d-xylonic nitrile-PCSK9 binding. It can form up to five hydrogen bonds with the amino acids TRP461, ALA649, VAL435, and ASN439. Z. officinale contains only 1 compound, 2-Formyl-9-[.beta.-d-ribofuranosyl]hypoxanthine, which contained only 4.37% of Z. officinale extract to form a high 8-position hydrogen bond with the amino acids TRP461, LEU436, ASP360, ARG458, ALA649, ASP651, and THR469. Finally, C. tinctorius. C. tinctorius's active compounds have poor binding to PCSK9 because it contains only two compounds: 3,4-Dihydroxyphenylglycol, 4TMS derivative (8.94%), and Cyclohexasiloxane, dodecamethyl-(6.96%), which were found to total just 15.9%, resulting in poor inhibition of PCSK9. These compounds formed very few hydrogen bonds with PCSK9 binding. C. fenestratum is the best PCSK9 inhibitor because of its high binding to the target protein and its high active compounds, followed by Z. officinale, which has a better PCSK9 inhibitor than the C. fenestratum. However, the low content of active compounds resulted in less efficacy of Z. officinale in inhibiting PCSK9. Finally, C. tinctorius was the least effective in inhibiting PCSK9 because of its fewer active compounds and poorer binding capacity than the herbs, as mentioned earlier. From the study results, an herbal recipe for reducing lipid has been designed by using the knowledge of Thai traditional medicine [58] to set the drug recipe as the main drug, which is the drug that has the highest efficiency in inhibiting lipid with the highest ratio. This recipe is three parts C. fenestratum. An assistance drug is a drug that will increase the efficiency of the main drug to reduce lipid with a lesser ratio. This recipe is two parts Z. officinale, and a colorant drug is used for adding applicability to the recipe with the lowest ratio. One part of C. tinctorius was added to this recipe. This recipe was tested for lipid reduction efficacy using HepG2 cells. It was found that this recipe could reduce lipid accumulation better than using the herb alone. Therefore, this is the world's first herbal recipe that helps reduce lipid through PCSK9 inhibition.
To clarify the substance structure and biological activity, the study found that the main inhibitors of PCSK9 were tetraacetyl-d-xylonic nitrile (27.92 percent) from C. fenestratum, and 2-Formyl-9-[.beta.-d-ribofuranosyl]hypoxanthine (4.37%) from Z. officinale. The study of Structure-Activity Relationship (SAR) is available through the website: http://way2 drug.com/PassOnline/predict.php. The structure of a substance with a Pa value greater than 0.7 indicates that the substance can be developed as a drug for the treatment of such diseases [60]. The composition analysis of C. fenestratum showed that tetraacetyl-d- hypoxanthine has slightly better water solubility than tetraacetyl-d-xylonic nitrile. As a result, both compounds with PCSK9 inhibitory activity were better extracted with water than ethanol, consistent with the results of the GC-MS/MS study that found tetraacetyld-xylonic nitrile in 27.92% water extraction while extracting only 9.47% with ethanol. Moreover, 2-Formyl-9-[.beta.-d-ribofuranosyl] hypoxanthine was extracted with a 4.37% yield in water, while there are no compounds found in ethanol extraction.

Conclusions
In conclusion, for screening PCSK9 inhibitors from three plants, C. tinctorius, C. fenestratum, and Z. officinale, an efficient technique incorporating molecular docking, RT-qPCR test, in vitro cytotoxicity, and Oil red O staining assay was devised. Two chemicals had a high yield from C. fenestratum based on GC-MS/MS detection: tetraacetyl-d-xylonic nitrile (27.92 percent) and Inositol, 1-deoxy-(24.89 percent). These compounds could inhibit PCSK9 strongly through the binding of 6 and 5 hydrogen bonds, respectively, while the active compound in Z. officinale is 2-Formyl-9-[.beta.-d-ribofuranosyl] hypoxanthine (4.37%), which inhibits PCSK9 by forming 8 hydrogen bonds. According to our findings, we may utilize a formula consisting of three parts C. fenestratum (primary herb), two parts Z. officinale (assistance herb), and one part C. tinctorius (servant herb) to define a reasonable herbal ratio for the intervention and prevention of PCSK9-related disorders in the future. Furthermore, because of targeted screening and precise analysis, this technique is expected to be used for a broader range of applications, such as fast screening of active components from herbs, and improving herb ratios in alternative medicine.

Data Availability Statement:
The datasets used and/or analyzed during the current study are available from the corresponding author Komgrit Eawsakul (komgrit.ea@wu.ac.th) on reasonable request.