We investigate several hypotheses in our study:

This study used Rattus norvegicus, strain Sprague Dawley rats, weighing 180 – 200 g and aged 8 – 10 weeks old.

Plasma PAI-1 levels were determined using Rat PAI-1 ELISA kit, Finetest from Wuhan Fine Biotech Co. Ltd. Lipid profiles were determined using Cholesterol FS (10’) 5 x 25 mL/1x 3 mL Standard, Triglycerides FS (10’) 5 x 25 mL/1 x 3 mL Standard, HDL Precipitant 250 mL, LDL Precipitant 250 mL from DiaSys, Holzheim, Germany.

The male Sprague Dawley rats (n = 42) were purchased from the Center of the Food and Nutrition Studies of Gadjah Mada University, Yogyakarta, Indonesia. An acclimatization period lasting seven days was scheduled before the beginning of the experiment. After the acclimatization period, rats were divided into seven groups (n = 6 per group): healthy rats given standard feed (K₁⁺), NAFLD rats without treatment (K₂⁺); 0.9 mg simvastatin treatment group (K₁^-); 45 mg metformin treatment group (K₂^-); single clove black garlic (SCBG) 450 mg per 200 g BW (X1); 900 mg per 200 g BW (X2); and 1350 mg per 200 g BW (X3). All animals, except the K₁⁺ group, were fed a high-fat fructose diet (HFFD) containing cholesterol (1.25%) and cholic acid (0.5%) for 8 weeks.

The standard feed containing 15% crude protein, 3 – 7% crude fat, 12% moisture content, 6% crude fiber, 7% ash, 0.5% phosphorus, and 0.9 – 1.1% calcium, and vitamins. The composition of HFFD as follow was 177.5 g.kg^-1 pork oil, 172.8 g.kg^-1 fructose, 19.87 kJ.g^-1 energy, 20% protein, 35% carbohydrates, and 45% fat.

Black garlic used in this study was produced from the Biotechnology Agency for Technology Assessment and Application (BPPT), Tangerang, Indonesia, with the commercial name of Natural Black Garlic® Lanang, which has passed the test of acute toxicity by the Laboratory Toxicity and Safety Materials, School of Pharmacy Bandung Institute of Technology, Indonesia. The main raw material for Natural Black Garlic® Lanang is single clove garlic. It's processed garlic that is produced through natural aging at a controlled temperature (70 °C) and high humidity (90%) for 26 days without other additives (Figure 1).

iMark™ microplate absorbance reader, Bio-Rad. Microprocessor Controlled Spectrophotometer, OPTIMA SP-300.

PAI-1 levels were measured using the Enzyme-Linked Immunosorbent Assay (ELISA) method (Crowther, 2009).

Total cholesterol and HDL-C levels were analyzed by Cholesterol Oxidase- Phenol Amino Phenazone (CHODPAP) and triglycerides were analyzed by Glycerol Phosphate Oxidase-Phenol Amino Phenazone (GPO-PAP) (McPherson and Pincus, 2016). LDL-C levels were calculated with the Friedewald formula.

Simvastatin, Metformin and Single Clove Black Garlic preparation as follows was 0.9 mg simvastatin and 45 mg metformin per 200g BW as comparative groups and SCBG dose 450 mg, 900 mg, and 1350 mg per 200 g BW, respectively, were mashed and diluted together with 3 mL of aqua bidestilata (Hatton et al., 2015). All samples were administered by oral gavage (Atcha et al., 2010).

Sample preparation: After 4 weeks of administration, all animals overnight fasted. Blood was collected from the sinus orbital (Wolforth, 2000) and centrifuged at 1,000 rpm for 15 minutes at 2 – 8 °C, then plasma was collected and stored at -80 °C for later analysis.

Number of samples analyzed: forty-two.

Number of repeated analyses: duplo

Number of experiment replication: There is no experiment replication in our study.

Bodyweight gain was recorded weekly, while examination data of PAI-1 levels and plasma lipid profiles were taken post-test. All data were analyzed statistically using IBM SPSS Statistics for Windows version 22.0. Shapiro-Wilk test was used to test the normality of data. The analyses used were paired t-test, one-way ANOVA, post- hoc LSD, Kruskal-Wallis, and Mann- Whitney. For the correlations between PAI-1 levels and lipid profiles used correlation Spearman test.

After administration of HFFD containing cholesterol (1.25%) and cholic acid (0.5%) for 8 weeks, generally, the six groups of rats experienced an increase in body weight as well as the healthy group that was given standard feed. Weighing the rats was performed weekly starting from the acclimatization until the end administration of HFFD. Weighing the mice after acclimatization aimed to ensure that the experimental animal still within the inclusion criteria, namely, having a bodyweight of 150 – 200 g. Table 1 shows the changes in body weight of rats before and after conditioning. These changes probably due to the HFFD administration for 8 weeks during conditioning. HFFD containing high-fat (45%) and high-fructose (172.8 g.kg^-1 diet), contributing a lot of energy (19.87 kJ.g^-1) (Jung et al., 2011). Fat also contains high energy compared to carbohydrates and protein. 1 g of fat is equivalent to 9 kcal of energy, while 1 g of protein and carbohydrates each is equivalent to 4 kcal of energy (Hardinsyah and Supariasa, 2019).

In this study, the body weight of the animals significantly increased after HFFD administration in six group treatments (p = 0.001). except for the healthy group, which had an increase but not significant. The X2 group does not have significant weight differences before and after conditioning. In this study, the administration of HFFD attempted to apply the diet composition of NAFLD patients. The difference in food intake before and after HFFD administration between groups was related to the differences in the energy composition of the diets of different rats. The dietary intake of 15 g per rat per day was selected according to the nutritional needs of laboratory animals, which is considered the minimum amount needed by rats per day to ensure normal growth of them (Fakhoury-Sayegh et al., 2015).

In this study, the increases of weights were meaningfully and dose-dependently inhibited by treatment of SCBG and also by simvastatin and metformin, respectively. The X2 and X3 groups experienced a change in obesity status after the SCBG administration. The SCBG group with doses of 900 mg and 1350 mg per 200 g BW per day had a preventive effect on obesity, which had no statistical difference from the healthy group. Meanwhile, the group of SCBG 450 mg per 200 g BW per day only had a slightly reduced Lee index, so they were unable to change the obesity status of the rats to normal (Table 2).

HFFD feeding in this study aimed to induce a fatty liver in rats. The excess consumption of carbohydrates (fructose and sucrose) and fat (fatty acids and cholesterol) plays a key role in the development and advancement of obesityrelated NAFLD through the activation of the lipid metabolism pathway modulated on a high-fat diet (Jarukamjorn et al., 2016).

The obesity status of the X2 and X3 groups had changed after the SCBG administration. The treatment group of SCBG 900 mg per 200 g BW per day and 1350 mg per 200 g BW per day provided a preventive effect of obesity and had no statistical difference from the healthy group. Meanwhile, the group of rats given SCBG 450 mg per 200 g BW did not reduce the Lee index. Garlic is a rich source of polyphenol compounds and is highly potent in inhibiting lipid peroxidation. Polyphenols have the potential to prevent obesity by inhibiting enzymes related to lipid metabolism, for example, pancreatic lipase, lipoprotein lipase, and glycerophosphate dehydrogenase, or by regulating lipid homeostasis, for example, by regulating lipogenesis and fatty acid oxidation (Chen et al., 2014). In conclusion that black garlic has an antiobesity effect (Chang, Wu and Hsu, 2015; Chang et al., 2017; Chen et al., 2014; Lee et al., 2011; Kim et al., 2011).

PAI-1 level was measured once at the end of the study. The post-hoc LSD test showed significant differences in the PAI-1 levels in all groups of rats (p = 0.001). The PAI-1 levels were meaningfully and dose-dependently decrease by treatment of SCBG and also by simvastatin and metformin, respectively. It compared to the NAFLD group, without SCBG treated. However, when compared with the healthy group, the PAI-1 levels in the three treatment groups had not yet reached their PAI-1 levels (Table 3).

Significant differences were noted in PAI-1 levels of all groups of rats (p = 0.001). The PAI-1 levels in the healthy group were significantly different from those in the NAFLD, simvastatin, metformin, X1, X2, and X3 groups. The SCBG dose of 1350 mg per 200 g BW more effectively improved the hemostasis status in NAFLD rats than 450 and 900 mg per 200 g BW respectively.

The consumption of HFFD appears to have contributed to NAFLD development and increased the risk of progression to NASH (Jarukamjorn et al., 2016), promote an increase in hemostasis status, namely, PAI-1 level. In this study, X1, X2, and X3 groups had lower levels of PAI-1 than the NAFLD group (HFFD without SCBG administration). In this study, the biological substances of garlic and its processes, which are involved in the antithrombotic function, are unknown. Allicin, one of the main components in garlic, is known to have some biological activity, but it does not seem to promote a t-PA activity or reduce PAI-1 levels. Black garlic contains less allicin in the production process due to alliinase inactivation, so it is odorless. However, whether the alliin in the diet was converted to allicin after administered in rats and whether SCBG in our study had a specific antithrombotic activity are uncertain. As is known, garlic contains many physiological functions; the composition and amount of each element are responsible for various functions, which may differ between the types and methods of manufacture. It should be noted that the physiological function of garlic is not homogeneous and depends on the number of active components and their combinations in garlic processing (Fukao et al., 2007). Therefore, which components contribute to the antithrombotic activity in NAFLD rats is uncertain. In short, garlic greatly reduces the coagulation activity, which would prevent the formation of blood clots in the thrombus-modeled pathological rats. These results may explain the benefit of garlic and its preparations for blood and blood vessels.

In vitro and in vivo studies have shown that garlic appears to have antiplatelet, fibrinolytic, and antithrombotic effects, in addition to its anti-inflammatory properties (Chang et al., 2005; Teranishi et al., 2003). Garlic also reduced PAI-1 levels in obese subjects compared to the placebo group (Szulińska et al., 2018). Another study showed that administration of Methanolic Extract of Black Garlic (MEBG) reduced levels of PAI-1 (p <0.05) in obese rats fed a high-fat diet (Chen et al., 2014). Regarding the molecular regulation of decreased levels of PAI-1 and inhibition of adipogenesis in obese mice by black garlic supplementation through increased AMPK, forkhead box protein-01(FOXO1), Sirtuin- 1(Sirt1), adipose triglyceride lipase (ATGL), hormone sensitive lipase (HSL), perilipin, acyl-CoA-1 (ACO), carnitine palmitoyltransferase (CPT-1), and uncoupling protein-1(UCP1), thereby reducing lipogenesis, increasing lipolysis and fatty acid oxidation. The expression of cluster of differentiation-36 (CD36) levels decreased, resulting in less free fatty acid uptake in the adipose tissue. Also, black garlic increases adiponectin and downregulates PAI-1, resistin, tumor necrosis factor- α (TNF-α), and glucose transporter type-4 (GLUT4), thereby increasing insulin resistance in rats fed a high-fat diet (Chen et al., 2014). Similarly, the downregulates PAI-1 mechanism of simvastatin and metformin (Arruda de Faria et al., 2019; Mazza et al., 2012; Nascimbeni et al., 2016; Rodrigues, et al., 2019).

In this study, the mean of total cholesterol (TC), highdensity lipoprotein-cholesterol (HDL-C), low-density lipoprotein-cholesterol (LDL-C), and triglycerides (TG) levels for all groups were significantly different (p = 0.001). After administration SCBG, the X1, X2, and X3 groups have a significant decrease in triglyceride levels (p = 0.001) compared to the NAFLD group (Table 3).

Consumption of high-fat diet results in the production of acetyl-CoA and malonyl CoA in the liver. Increased levels of malonyl CoA, as fuel for TG synthesis, will inhibit the expression of CPT-1, a key enzyme for lipolysis, thereby reducing fatty acid oxidation. As shown in this study, administration of HFFD and black garlic increases CPT-1 mRNA expression; thus, more TG is broken down (degradation) to supply free fatty acids as an energy source, lowering plasma cholesterol concentrations associated with decreased HMG-CoA reductase expression and ACAT. SREBP-1c levels in the liver also activate the 3-hydroxy-3-methyl-glutaryl-coenzyme A (HMG-CoA) reductase and acyl-CoA cholesterol acyltransferase (ACAT) enzymes, which regulate cholesterol synthesis and catalyze cholesterol esterification (Ha, Ying and Kim, 2015).

Calorie intake from fat and fructose promotes excess fat deposits in the adipose tissue. Excessive intake of fat from food will lead to an increase in lipogenesis activity and production of free fatty acids, resulting in the mobilization of free fatty acids from the fatty tissue to the liver and binding glycerol to form triglycerides. Giving a high-fat diet also increased levels of LDL and cholesterol in the circulation, triggering HDL to transport cholesterol from the liver to the circulation (reverse cholesterol transport). HDL is esterified into cholesterol esters, which can be taken directly to the liver then directly excreted or exchanged with triglycerides from VLDL and chylomicrons. When cholesterol ester excess, HDL-rich triglyceride (HDL low density) is breakdown by hepatic lipase resulting in lower circulating levels of HDL (Dissard et al., 2013; Wong et al., 2016).

The results showed that the mean of total cholesterol, HDL, LDL, and triglycerides levels for all groups were significantly different (p = 0.001). After SCBG administration, the X1, X2, and X3 groups had lower triglyceride levels (p = 0.001) compared to the NAFLD group. These results supported the study of Seo et al. (2009) on db/db model rats given black garlic had lower triglycerides, cholesterol, fasting blood glucose, blood glycated hemoglobin levels, and higher HDL serum compared with rats with no black garlic intervention.

The lipid-lowering mechanism of black garlic in plasma and the liver of rats is described in Ha et al.’s study.

They have measured the expression of liver SREBP-1c mRNA transcription factors’ key lipid metabolism and the decreased expression of SREBP-1c in rats fed extract black garlic, resulting in a decrease in liver sterol regulator element-binding protein-1c (SREBP-1c), Acetyl-CoA carboxylase (ACC), fatty acid synthase (FAS), and G-6PDH mRNA levels, the target gene for SREBP-1c, indicating that the inhibition of the gene expression led to a decrease in the fat synthesis of the liver.

This is associated with a plasma TG concentration decrease. Similar results were reported: the methanolic extract Black Garlic (MEBG) administered in rats fed with a high-cholesterol diet inhibited lipid accumulation through adenosine monophosphate-activated protein kinase (AMPK) regulation and regulation of SREBP1c, ACC, and FAS (Ha, Ying and Kim, 2015; Sobenin et al., 2019). Similarly, the lipid-lowering mechanism of simvastatin and metformin (Arruda de Faria et al., 2019; Mazza et al., 2012; Nascimbeni et al., 2016; Rodrigues et al., 2019; Sobenin et al., 2019).

Black garlic and its constituent, S-allyl-cysteine, had antioxidant and hypolipidemic effects in rats with a highfat diet (Asdaq, 2015). Therefore, the high concentration of S-allyl-cysteine in black garlic extract induced blood lipid profile changes in hepatic injury rats (Tran, Dam and Tram Le, 2018). These data support the results of this study in verifying the effect of single clove black garlic improving the lipid profile in NAFLD rats.

There were strong correlations between PAI-1 levels and lipid profiles, with r value = 0.76–0.99. The correlations were positive except for HDL-C levels (p = 0.001; r = -0.968), which means when HDL-C levels decrease, then the PAI-1 levels increase (Table 4).

Decreased plasma HDL cholesterol is the most common dyslipidemia in obese subjects and this condition is caused by hypertriglyceridemia, even if the levels are low or without elevated triglycerides. Hypertriglyceridemia is accompanied by a small increase in solid LDL, is highly atherogenic. The high concentration of non-esterified fatty acid (NEFA) released will be metabolized by the liver to increase the synthesis of liver triglycerides and very-lowdensity lipoprotein (VLDL), thereby triggering insulin resistance. High circulating levels of VLDL reduce HDL concentrations and increase small solid LDL. Dyslipidemia is also accompanied by hyperactive platelets, hypercoagulability with increased factor VII, and hypofibrinolysis with increased PAI-1. High plasma triglyceride levels increase oxidative stress by increasing superoxide production. Increased NEFA from adipocytes increases tissue factor (TF) and PAI-1 levels and increases platelet aggregation; this condition causes thrombosis. Lipoprotein (a) levels in obesity are increased, especially when associated with hyperglycemia. Lipoprotein (a) has LDL-like properties and is structurally similar to plasminogen. It, therefore, inhibits the binding of plasminogen to endothelial cells and interferes with the generation of plasmin, both of which are atherogenic and thrombogenic. Prothrombin levels and levels of vitamin K-dependent coagulation factor are increased with hyperlipidemia (Darvall et al., 2007).

Significant correlations between PAI-1 levels and lipid profiles were noted in this study. Similarly, where a significant positive correlation was seen between PAI-1 and triglyceride levels; on the other hand, PAI-1 and HDL-C levels showed a negative correlation in nondiabetic obese patients and the control group (Festa et al., 1999; Somodi et al., 2018). Besides, another study showed a strong and significant positive correlation between the most atherogenic LDL subfraction and the densest concentration of the PAI-1 plasma levels was also noted. This is the initial finding to evaluate the correlation between the small dense LDL and PAI-1 plasma levels in obese nondiabetic patients (Somodi et al., 2018; Väisänen et al., 1997). These data support the association between LDL-C and PAI-1 levels in accelerating atherogenesis and increasing the cardiovascular risk of obese patients; however, our hypothesis needs to be verified by further investigation. However, the correlation between HDL and plasma PAI-1 levels are not well understood. Also, there have been few studies on the correlations of PAI-1 with the functional and structural properties of different lipoprotein fractions especially on rats modeled NAFLD.

We were not able to evaluate the long-term effects of SCBG administration on the hemostasis status and lipid profiles in rats modeled NAFLD, as our intervention lasted 4 weeks. Given these limitations and our findings, we believe that additional studies with longer intervention periods are warranted to determine the long-term benefits of SCBG consumption on the hemostasis status and lipid profiles in rats modeled NAFLD. However, further studies are needed to elucidate the underlying mechanisms.