*Mikania micrantha* **Extract Inhibits HMG-CoA Reductase and ACAT2 and Ameliorates Hypercholesterolemia and Lipid Peroxidation in High Cholesterol-Fed Rats**

**Azlinda Ibrahim 1, Nurul Husna Shafie 1,2,\*, Norhaizan Mohd Esa 1, Siti Raihanah Shafie 1, Hasnah Bahari <sup>3</sup> and Maizaton Atmadini Abdullah 4,5**


Received: 27 July 2020; Accepted: 15 September 2020; Published: 9 October 2020

**Abstract:** The present study aimed to determine the effect of an ethyl acetate extract of *Mikania micrantha* stems (EAMMS) in hypercholesterolemia-induced rats. Rats were divided into a normal group (NC) and hypercholesterolemia induced groups: hypercholesterolemia control group (PC), simvastatin group (SV) (10 mg/kg) and EAMMS extract groups at different dosages of 50, 100 and 200 mg/kg, respectively. Blood serum and tissues were collected for haematological, biochemical, histopathological, and enzyme analysis. Total cholesterol (TC), triglycerides (TG), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), aspartate aminotransferase (AST), alanine aminotransferase (ALT), urea, creatinine, malondialdehyde (MDA) level, as well as enzymes of HMG-CoA reductase (HMGCR) and acetyl-CoA acetyltransferase 2 (ACAT2), were measured. Feeding rats with high cholesterol diet for eight weeks resulted in a significantly (*p* < 0.05) increased of TC, TG, LDL-C, AST, ALT and MDA levels. Meanwhile, the administration of EAMMS extract (50, 100 and 200 mg/kg) and simvastatin (10 mg/kg) significantly reduced (*p* < 0.05) the levels of TC, TG, LDL-C and MDA compared to rats in the PC group. Furthermore, all EAMMS and SV-treated groups showed a higher HDL-C level compared to both NC and PC groups. No significant difference was found in the level of ALT, AST, urea and creatinine between the different dosages in EAMMS extracts. Treatment with EAMMS also exhibited the highest inhibition activity of enzyme HMGCR and ACAT2 as compared to the control group. From the histopathological examination, liver tissues in the PC group showed severe steatosis than those fed with EAMMS and normal diet. Treatment with EAMMS extract ameliorated and reduced the pathological changes in the liver. No morphological changes showed in the kidney structure of both control and treated groups. In conclusion, these findings demonstrated that EAMMS extract has anti-hypercholesterolemia properties and could be used as an alternative treatment for this disorder.

**Keywords:** *Mikania micrantha*; anti-hypercholesterolemia; lipid profile; steatosis; nutraceuticals

#### **1. Introduction**

Hypercholesterolemia is a metabolic disorder that mainly results in an elevated concentration of plasma low density lipoprotein (LDL) cholesterol [1]. Hypercholesterolemia has been associated with many cardiovascular diseases, including atherosclerosis, stroke, cerebral paralysis, myocardial infarction [2] and also inflammation and cancer [3]. One of the bigger challenges in modern medicine is the identification of a cure for hypercholesterolemia which does not confer side effects. In recent times, plant-sourced products have been considered to be possible novel therapeutic agents as these are considerably less toxic, cost-effective and most importantly, they produce no or relatively lesser side effects as compared to their synthetic counterparts.

*Mikania micrantha* Kunth originates from the tropical central and southern part of America and is extensively spread in the Pacific region and Southeast Asian countries. *M. micrantha* is traditionally used to treat stomach aches, jaundice, respiratory diseases, dysentery and rheumatism. This perennial creeping vine is also consumed as a juice as an alternative medicine for the treatment of diabetes, hypertension and hypercholesterolemia [4,5]. *M. micrantha* were previously reported to possess many health benefits such as antioxidant [6], anti-diabetic [7], anti-cancer [8,9], antiproliferative [10], anti-dermatophytic [11], anti-inflammatory [12] and antibacterial [5] activities. These beneficial effects are related to the richness of chemical constituents such as terpenoids, flavonoids, alkaloids and vitamins [5,6,11,13].

Despite its traditional use, scientific findings to prove the traditional claims of the anti-hypercholesterolemia properties of *M. micrantha* are limited. For that reason, this study aimed to determine the hypocholesterolemic potential of an ethyl acetate extract of *M. micrantha* stems (EAMMS) on male high-cholesterol-fed rats by determining the serum lipid profile [TC, total cholesterol; TG, triglycerides; HDL-C, high-density lipoproteins cholesterol; LDL-C, low-density lipoproteins cholesterol], lipid peroxidation, enzymatic activities and histopathological evaluation.

#### **2. Materials and Methods**

#### *2.1. Reagents and Chemicals*

Ethyl acetate (HmbG Chemical, Hamburg, Germany), ethanol (R&M Chemicals), haematoxylin (Sigma-Aldrich, St Louis, MO, USA), simvastatin (Pharmaniaga Logistics (M) Sdn. Bhd, Malaysia), 10% formalin (R&M Chemicals), xylene (R&M Chemicals) and eosin (Leica Biosystems Richmond Inc., Richmond, IL, USA) were used in this study. Pierce BCA Protein Assay Kits for protein quantification were purchased from Thermo Fisher Scientific (Rockford, IL, USA). Malondialdehyde (MDA) assay kits for lipid peroxidation assays were purchased from Elabscience Biotechnology Inc. (Houston, TX, USA). ELISA Kits (HMGCR and ACAT2) were purchased from Sunlong Biotech Co. Ltd. (Zhejiang, China). High cholesterol diet (1%) was purchased from EnvigoTeklad (Cambridgeshire, UK).

#### *2.2. Sample Preparation*

*Mikania micrantha* was collected in August 2017 from Negeri Sembilan, Malaysia (GPS: 2.695652,102.160987) and a plant sample was deposited in the Forest Research Institute Malaysia (FRIM), Kepong, Selangor, Malaysia for taxonomic identification with a voucher specimen number of SBID 051/15. Fresh stems of *M. micrantha* was selected and washed and then dried at 35 ◦C for 72 h in a ventilated drying oven. *M. micrantha* powdered stems were then extracted using ethyl acetate and evaporated at 48 ◦C [6].

#### *2.3. Animals*

Sprague Dawley rats (male, 150–200 g weight) were purchased in this study. The rats were given tap water *ad libitum* and acclimatized under standardized laboratory circumstances (temperature 22 ± 2 ◦C; humidity 60 ± 4%; 12 h light-dark cycle). All animals used had received approval from the Universiti Putra Malaysia's Institutional Animal Care and Use Committee (UPM/IACUC/AUP-R081/2017).

#### *2.4. Animal Experimental Design*

The rats were divided into six groups (*n* = 6). Group 1 as normal control (NC) was fed with a normal diet for 8 weeks. Group 2 to 6 were orally administered with high cholesterol diet (1%) throughout the study for 8 weeks. After the 4th week of the induction period, Group 2 was served as cholesterol-induced rats (PC). Group 3 was treated by orally administered via gavage with an aqueous suspension of simvastatin, SV (10 mg/kg) and groups of 4, 5 and 6 were orally administered via gavage with EAMMS at the dosage of 50, 100 and 200 mg/kg, respectively, during the treatment periods. The rats were treated for 4 weeks. Upon the administration of the last treatment dose, the animals were left to fast for 18 h. Blood samples were then acquired through the cardiac puncture approach with subjects under anaesthesia (ketamine/xylazine). Then, the blood samples were centrifuged at 3000 rpm at 4 ◦C for 10 min and the serum was stored at a temperature of −80 ◦C until the point of the assay. The liver and kidney were excised, weight and also stored at a temperature of −80 ◦C until the point of analysis.

#### *2.5. Liver, Kidney and Haematogram Profile Analysis*

Analysis of both liver and kidney profile including aspartate aminotransferase (AST), alanine aminotransferase (ALT), urea and creatinine were analyzed using a fully automated BiOLiS 24i Premium clinical analyzer (Hitachi, Bolton, UK). The total number of red blood cells (RBC), haemoglobin (Hb), packed cell volume (PCV), mean corpuscular volume (MCV), mean corpuscular haemoglobin volume (MCHC), white blood cell (WBC), icterus index, and plasma protein concentration were analysed.

#### *2.6. Lipid Profile Analysis*

The parameters that were analysed included total cholesterol (TC), triglycerides (TG), low density lipoproteins (LDL-C) and high density lipoproteins (HDL-C). The analyses were conducted using a fully automated clinical analyser (BiOLiS 24i Premium, Hitachi).

#### *2.7. Serum Lipid Peroxidation*

MDA level was quantified in the blood serum using a rat malondialdehydes (MDA) ELISA kit from Elabscience Biotechnology Inc., (Houston, TX, USA). All the procedures were conducted carefully according to the manufacturer's instructions. The absorbance was measured at 450 nm.

#### *2.8. Histopathological Examination*

Briefly, 10% formalin buffer solution was used to fix the liver and kidney. Upon fixing, the tissues were subjected to paraffin embedding and being stained with haematoxylin and eosin (H&E) dye. The tissues of interest which have been stained were viewed and analysed using an image analyser. The histology of selected tissues was evaluated qualitatively and quantitatively. The microscopic analysis of all tissue samples was evaluated as a blind study by a pathologist and other researchers. Grading was conducted on the severity of steatosis which afflicted liver tissues and this was done according to the approach as described by Brunt et al. [14].

#### *2.9. HMGCR and ACAT2 Activity Assays*

Protein from liver tissue was homogenized in phosphate buffered saline (PBS, pH 7.4) at 4 ◦C. The supernatant was then centrifuged at 3500× *g* with 4 ◦C for 10 min and quantified with a BCA protein assay kit according to kit instructions. The levels of HMG-CoA reductase (HMGCR) and acetyl-CoA acetyltransferase 2 (ACAT2) enzymes were measured according to the ELISA protocol provided by the manufacturer.

#### *2.10. Statistical Analysis*

The data were expressed as mean ± standard error of the mean (SEM) for body and organ weight, haematological parameters, liver and kidney profile, lipid profile, lipid peroxidation and enzymatic analysis. All data were analysed using one-way analysis of variance (ANOVA) and Tukey's multiple comparison tests (*p* < 0.05).

#### **3. Results**

#### *3.1. Body and Organ Weights*

The changes in body weights of rats during the experiments are presented in Figure 1. All groups showed increased body weights throughout the study. The NC group exhibited an increasing body weight throughout the experimental period. After 4 weeks of induction with high cholesterol diet (HCD), the body weight of all HCD-induced groups was increased compared to the normal group (NC) but not significantly when compared to each other (Figure 1). Besides, there was a significant increased (*p* < 0.05) in the body weight of rats fed with high cholesterol diet for 8 weeks compared to the NC group. On the other hand, after the treatment period with EAMMS at different dosages, the body weight was shown to have lower body weight but not significant (*p* > 0.05) when compared to the PC group. Among the EAMMS-treated group, 100 mg/kg showed the highest body weight increment but not significant when compared to 50 mg/kg and 200 mg/kg treatment groups. The SV-treated group was showed no significant difference (*p* > 0.05) of body weight when compared to PC and EAMMS (50, 100 and 200 mg/kg) groups.

**Figure 1.** Effects of ethyl acetate extract of *Mikania micrantha* stems (EAMMS) on body weight in hypercholesterolemia rats. Values are expressed as mean ± SEM (*n* = 6). Means with different superscripts (a or b) indicate statistically different at *p* < 0.05 using Tukey's multiple comparison test. NC—Normal control; PC—Positive control with high cholesterol diet (1%); SV—High cholesterol diet (1%) with simvastatin (10 mg/kg body weight); EAMMS50—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (50 mg/kg body weight); EAMMS100—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (100 mg/kg body weight); EAMMS200—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (200 mg/kg body weight).

Table 1 indicates the weights of the livers and kidneys of the experimental rats. The liver weight from the PC group was significantly higher (*p* < 0.05) compared to the NC group and no difference with all EAMMS-treated groups. Meanwhile, there are no significant differences (*p* > 0.05) of kidney weight (right or left) between all groups.


**Table 1.** Organ weight of rats at week 8.

Values are expressed as mean ± SEM (*n* = 6). Means with different superscripts (a or b) in a column indicate statistically different at *p* < 0.05 by Tukey's multiple comparison tests. NC—Normal control; PC—Positive control with high cholesterol diet (1%); SV—High cholesterol diet (1%) with simvastatin (10 mg/kg body weight); EAMMS50—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (50 mg/kg body weight); EAMMS100—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (100 mg/kg body weight); EAMMS200—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (200 mg/kg body weight).

#### *3.2. E*ff*ect of EAMMS on Liver, Kidney and Haematogram Parameters*

Table 2 shows the serum haematological, liver and kidney profiles of rats. Rats that were fed with a diet supplemented with 1% cholesterol for 8 weeks displayed a significant (*p* < 0.05) increase in AST, ALT and MCV levels. This is apparent when the PC group is compared against the NC group. Meanwhile, the levels of urea, creatinine, RBC, MCHC, WBC from the PC group showed slight increases but these were not significant (*p* > 0.05) when compared to the NC group. After 4 weeks of treatment, the levels of AST, ALT and MCV in the EAMMS-treated groups were not significant (*p* > 0.05) when compared to the PC group whilst the RBC, MCHC, WBC and additionally the kidney profiles (urea, creatinine), showed no significant differences between all groups.

**Table 2.** Liver, kidney and haematology values of rats after treated with EAMMS extracts (week 8).


Values are expressed as mean ± SEM (*n* = 6). Means with different superscripts (a, b or c) in the same row indicate statistically different at *p* < 0.05 by Tukey's multiple comparison test. NC—Normal control; PC—Positive control with high cholesterol diet (1%); SV—High cholesterol diet (1%) with simvastatin (10 mg/kg body weight); EAMMS50—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (50 mg/kg body weight); EAMMS100—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (100 mg/kg body weight); EAMMS200—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (200 mg/kg body weight); AST—Aspartate aminotransferase; ALT—Alanine aminotransferase; RBC—Red blood cells; Hb—Haemoglobin; PCV—Packed cell volume; MCV—Mean corpuscular volume; MCHC—Mean corpuscular haemoglobin concentration; WBC—White blood cells.

However, among the different dosages of EAMMS, 200 mg/kg resulted in a significant increase (*p* < 0.05) in the level of creatinine compared to the other dosages of EAMMS-treated and NC groups. In the SV-treated group, both AST and ALT levels showed significantly increases (*p* < 0.05) when compared to PC groups. Besides, there were no significant (*p* > 0.05) differences in Hb concentration and icterus index at all EAMMS dosages and controls. The PCV level also was not affected by the treatment of EAMMS in this study. For the leukocyte parameters, no significant differences were identified in terms of WBC count as well as differential leukocyte count.

#### *3.3. E*ff*ect of EAMMS on Lipid Profile*

Table 3 shows TC, TG, LDL-C and HDL-C levels in the serum of the experimental rats. The results show that feeding rats a diet supplemented with 1% cholesterol for 8 weeks resulted in a significant (*p* < 0.05) increase in TC, TG and LDL-C levels in the PC group compared to the NC group. After 4 weeks of the treatment (EAMMS and SV) period, there were significant (*p* < 0.05) decreases in serum TC, TG and LDL-C levels observed in the EAMMS-treated and SV groups compared to the PC group (Table 3). The levels of HDL-C in all EAMMS-treated groups and SV group showed no significant (*p* > 0.05) differences when compared to control groups (NC and PC). There were no significant differences (*p* > 0.05) in the level of serum TC, TG, LDL-C and HDL-C when compared between different dosages of EAMMS extract at week 8. Besides, the level of TC, TG, LDL-C and HDL-C in all EAMMS-treated groups were comparable and not significant (*p* > 0.05) as compared to the SV-treated group.

**Table 3.** Serum lipid profiles of hypercholesterolemia-induced rats treated with different concentrations of *Mikania micrantha* stems (EAMMS) extracts at week 8.


Values are expressed as mean ± SEM (*n* = 6). Means with different superscripts (a, b or c) in a column indicate statistically different at *p* < 0.05 by Tukey's multiple comparison tests. NC—Normal control; PC—Positive control with high cholesterol diet (1%); SV—High cholesterol diet (1%) with simvastatin (10 mg/kg body weight); EAMMS50—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (50 mg/kg body weight); EAMMS100—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (100 mg/kg body weight); EAMMS200—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (200 mg/kg body weight); TC: Total cholesterol; TG: Triglyceride; LDL-C: Low density lipoprotein cholesterol; HDL-C: High density lipoprotein cholesterol.

#### *3.4. E*ff*ect of EAMMS on Lipid Peroxidation*

The levels of malondialdehyde (MDA) in the serum of experimental rats are listed in Table 4. Rats fed with a diet supplemented with 1% cholesterol over 8 weeks had significant (*p* < 0.05) increases in MDA levels when compared between the PC group and the NC group. After 4 weeks of the treatment period, there were marked and significant decreases (*p* < 0.05) in serum MDA levels observed in the EAMMS-treated and SV groups compared to the PC group but these were not significant when compared to the NC group. The comparison of different dosages of EAMMS extract did not show any significant differences (*p* > 0.05) in the serum MDA levels at week 8 (Table 4).


**Table 4.** Effect of ethyl acetate extract of *Mikania micrantha* stems (EAMMS) on lipid peroxidation using TBARS assay in hypercholesterolemia rats at week 8.

Values are expressed as mean ± SEM (*n* = 6). Means with different superscripts (a or b) indicate statistically different at *p* < 0.05 by Tukey's multiple comparison tests. NC—Normal control; PC—Positive control with high cholesterol diet (1%); SV—High cholesterol diet (1%) with simvastatin (10 mg/kg body weight); EAMMS50—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (50 mg/kg body weight); EAMMS100—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (100 mg/kg body weight); EAMMS200—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (200 mg/kg body weight); TBARS—Thiobarbituric acid reactive substances; MDA—Malondialdehyde.

#### *3.5. Histopathological Results*

Table 5 shows the results for the liver histopathological scoring analysis of rats. Liver steatosis was absent in the NC group but observed in the PC group. There was a significant difference in steatosis between the PC group and those fed with EAMMS and NC diet. Liver tissues in the PC group showed severe diffuse steatosis within the hepatocytes and loss of single cell plates. The hepatocytes showed large cytoplasmic vacuoles due to fat deposition. However, there was no lobular or portal tract inflammation observed in the PC group. In the SV group, mild portal inflammation was observed accompanying moderate steatosis. There were expanded of losses single cell plates of hepatocytes with no significant lymphocytic infiltration in the portal tract. All EAMMS-treated liver had significantly less (*p* < 0.05) liver fat deposited than the PC group, with no inflammation, fibrosis or congestion on them that similar to that of the NC group (Table 5, Figure 2).

The effects of EAMMS in kidney tissues are illustrated in Figure 3. None of the treated and control groups showed any morphological changes in kidney structure. The sections for each group showed normal appearance with regulated nuclear arrangement of uriniferous tubules and collecting tubules with normal looking glomeruli, there was absence of sclerosis, no mesangial proliferation and no inflammation in the kidney parenchyma.


**Table 5.** Liver steatosis and inflammation scores in rats of different groups.

Normal (-): No hepatocytes; Grade 1 (+): <33% of hepatocytes; Grade 2 (++): 33% to 66% of hepatocytes and Grade 3 (+++): >66% of hepatocytes involved [14].

**Figure 2.** Histopathology of rat liver tissue in different groups. (**a**) NC—Normal control, (**b**) PC—Positive control with high cholesterol diet (1%), (**c**) SV—High cholesterol diet (1%) with simvastatin (10 mg/kg body weight), (**d**) EAMMS50—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (50 mg/kg body weight), (**e**) EAMMS100 High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (100 mg/kg body weight), (**f**) EAMMS200—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (200 mg/kg body weight). Livers were stained with hematoxylin and eosin (H&E) and visualized under a light microscope at 100× *g* magnification.

**Figure 3.** Histology of rat kidney tissue in different groups. (**a**) NC—Normal control, (**b**) PC—Positive control with high cholesterol diet (1%), (**c**) SV—High cholesterol diet (1%) with simvastatin (10 mg/kg body weight), (**d**) EAMMS50—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (50 mg/kg body weight), (**e**) EAMMS100 High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (100 mg/kg body weight), (**f**) EAMMS200—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (200 mg/kg body weight). Kidneys were stained with hematoxylin and eosin (H&E) and visualized under a light microscope at 200× *g* magnification.

#### *3.6. E*ff*ect of EAMMS on HMGCR and ACAT2 Enzymes*

The levels of HMG-CoA reductase (HMGCR) of experimental rats are given in Table 6. Rats that have been fed with a diet supplemented with 1% cholesterol over 8 weeks had significant (*p* < 0.05) increases in HMGCR levels when compared to the NC group. After 4 weeks of treatment, there were significant decreases (*p* < 0.05) in the HMGCR levels observed in the EAMMS-treated and SV groups compared to the PC group. Among the three different dosages of EAMMS extract, no significant difference (*p* > 0.05) was observed in the HMGCR levels at week 8 when compared to each other (Table 6).

**Table 6.** Effect of ethyl acetate extract of *Mikania micrantha* stems (EAMMS) on HMG-CoA reductase level in experimental rats at week 8.


Values are expressed as mean ± SEM (*n* = 6). Means with different superscripts (a, b or c) indicate statistically different at *p* < 0.05 by Tukey's multiple comparison tests. NC—Normal control; PC—Positive control with high cholesterol diet (1%); SV—High cholesterol diet (1%) with simvastatin (10 mg/kg body weight); EAMMS50—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (50 mg/kg body weight); EAMMS100—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (100 mg/kg body weight); EAMMS200—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (200 mg/kg body weight).

Table 7 shows the levels of acetyl-CoA acetyltransferase 2 (ACAT2) in all experimental rats. Based on Table 7, rats that supplemented with 1% cholesterol over 8 weeks showed a significant (*p* < 0.05) increases in ACAT2 levels when compared to the NC group. After 4 weeks of treatment, rats in the EAMMS-treated and SV groups showed significantly decrease (*p* < 0.05) ACAT2 levels compared to the PC group. Among the three different dosages of EAMMS extract, the dosage of 50 mg/kg showed a slight decrease but no significant difference (*p* > 0.05) in the ACAT2 level compared to 100 mg/kg and 200 mg/kg at week 8.


**Table 7.** Effect of ethyl acetate extract of *Mikania micrantha* stems (EAMMS) on ACAT2 levels in all treated experimental rats at week 8.

Values are expressed as mean ± SEM (*n* = 6). Means with different superscripts (a, b or c) indicate statistically different at *p* < 0.05 by Tukey's multiple comparison tests. NC—Normal control; PC—Positive control with high cholesterol diet (1%); SV—High cholesterol diet (1%) with simvastatin (10 mg/kg body weight); EAMMS50—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (50 mg/kg body weight); EAMMS100—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (100 mg/kg body weight); EAMMS200—High cholesterol diet (1%) with ethyl acetate *Mikania micrantha* stem extract (200 mg/kg body weight).

#### **4. Discussion**

Feeding with a cholesterol-enriched diet is one of the most commonly used methods for the induction of hypercholesterolemia in rats [15]. Administration of a high cholesterol diet to rats produces a marked increase of serum TC, TG and LDL-C as well as body weight when compared to a normal diet. A previous study revealed that a 1% increase in cholesterol intake by rats led to hypercholesterolemia, as evidenced by significant increases in serum TC, TG and LDL-C levels [16]. These observed changes are akin to those which could be expected when there is an excessive cholesterol load reaching the liver; a load that exceeds normal physiological limits. This will cause the inability of the liver to metabolize lipids thereby resulting in a relatively higher return of cholesterol into the blood circulation [17]. The relatively high levels of LDL which were found in rats which were fed with the cholesterol-enriched diet may be associated with the downward regulation of low density lipoprotein receptors (LDLR) by saturated fatty acids and dietary cholesterol [18]. According to Fungwe et al. [19], the elevation of triglycerides is caused by the cholesterol that was present in the diet which had been shown to diminish the oxidation of fatty acids and in the process, increase triglyceride and hepatic function levels.

Treatment with EAMMS extracts at all dosages showed marked decreases in serum TC, TG, and LDL-C levels and increases of HDL-C compared to the rats in the PC group after 4 weeks of supplementation. This phenomenon was explained by Adaramoye et al. [20] who reasoned that the decreased levels of cholesterol and triglycerides upon treatment are attributable to the lowering of the biosynthesis of hepatic triglycerides and the redeployment of cholesterol molecules between the molecules of lipoproteins. The effect of EAMMS extracts is also related to the abundance of chemical constituents' presence in EAMMS extracts such as terpenoids, flavonoids, alkaloids and vitamins [5,6,11]. Terpenoids (particularly sesquiterpene lactones) are the major compounds found in the ethyl acetate extracts of the flowers, leaves and the whole part of *M. micrantha* [11]. The chemical profile of *M. micrantha* led to the identification of terpenoids such as stigmasterol, stigmasteryl-β-D-glucopyranoside, acetyl β-amyrin and lupeol [10]. The structural similarity of stigmasterol or plant sterol with cholesterol makes plant sterols some of the best substances in reducing cholesterol levels in the blood.

Several mechanisms have been postulated to clarify the process of cholesterol reduction by phytosterols or plant sterols. Plant sterols compete with biliary and also dietary cholesterol as an inhibitor to bind with mixed micelles for solubilization of micellar in the upper intestinal lumen that leads to the general reduction in the ability of the intestinal lining in absorbing cholesterol [21,22]. This hypothesis is consistent with the postulation byBrufau et al. [23] which showed that dietary plant sterols can diminish TC and LDL-C in animal and human models. Plant sterol and stanol ester increase LDLR mRNA and ex vivo LDLR protein expression in the monocytes as well as T-lymphocytes of humans and this changes correlated negatively with the changes of concentration of LDL in the blood, it may be postulated that upregulating LDLR expression leads to decreased LDL formation along the apolipoprotein B cascade [24].

One of the more important methods for diagnosing the cause of disease and the health status of rats is the assessment of haematological parameters. In this study, no significant changes were detected in all haematological values excluding MCV between EAMMS extract, simvastatin and the control groups. However, the administration of the EAMMS caused mild microcytic anemia due to the level of MCV in rat's blood. MCV represents the average volume of the red blood cells. This abnormal blood condition could be caused by the presence of chemical constituents in plant extracts such as flavonoids and alkaloids saponins. Alkaloids have been shown to cause liver cirrhosis, liver megalocytosis and nodular hyperplasia [25] while terpenoids increase membrane permeability to divalent and monovalent ions [25].

Modern toxicology often involves the utilisation of blood sera or tissues as markers to assess damage to organs and cells besides the induction, activation and inhibition of enzymes. The kidneys and liver are organs which play major roles in the detoxification of metabolic substances [26]. AST and ALT are markers which are normally used to detect and assess injuries onto hepatocytes. It is noted that both these markers are introduced into the bloodstream following incidences where cell damage or necrosis occurs [27]. From the findings of this study, PC rats exhibited significant increases in AST and ALT levels as compared to NC rats. A similar elevation of AST and ALT levels was observed in hypercholesterolemic rats by the findings of Souza et al. [28]. In the broader sense, hypercholesterolemia is often associated with toxicity due to heightened levels of liver enzymes as well as the peroxidation of lipids that produces a lot of free radicals in blood sera and tissues [29,30].

Rats which were treated with simvastatin recorded significant elevations in AST and ALT levels. It is noteworthy that prominent adverse effects associated with the administration of the statin group are asymptomatic increases in liver transaminases as well as myopathy [31]. According to Castro et al. [32], the frequency of liver enzymes increased in a small proportion of those taking statins (2.5%). On the contrary, rats which were supplemented with EAMMS showed no differences in AST and ALT levels as compared to those in the NC group. It can, therefore, be deduced that the EAMMS treatment may suppress the level of AST and ALT in the blood and possibly help in the healing of the hepatic tissue damage, suggesting that plant extract can stabilize the plasma membrane as well as cures the damage of hepatic tissues [33].

It has been determined that one of the vital mechanisms of cellular damage that are caused by free radicals such as reactive oxygen species (ROS) is lipid peroxidation. One of the products of the peroxidation of lipid is malondialdehyde (MDA) and this is used as an index to indicate oxygen-free radicals' levels. MDA constitutes the main fraction of aldehydes that are produced upon the metabolism of lipid hydroperoxides and is extensively used to quantify and for the determination of lipid peroxidation. Elevation of serum MDA and TC in the hypercholesterolemic rats without any treatment suggested that the occurrence of lipid peroxidation eventually caused hypercholesterolemia [34]. According to Yokozawa et al. [35], decreases in lipid peroxidation could result in a reduction in the probability of hypercholesterolemia. In this present study, treatment with EAMMS extract was found to confer protection against lipid peroxidation in hypercholesterolemia-induced rats. EAMMS-treated rats were used to reduce MDA concentration nearly normal level compared to the rats with high cholesterol diet, therefore suggesting that EAMMS might possess an antioxidant activity owing to the presence of its phenol, alkane hydrocarbons, flavonoids and phytosterol [6], thus suggesting that EAMMS leads to beneficial response on oxidative stress in hypercholesterolemic rats. Many previous studies have shown that plant polyphenols, flavonoids, carotenoids, vitamins can effectively lower the level of TC, TG, LDL and MDA in hyper-cholesterolemic rodents [36–39]. The present results were following the findings of Musolino et al. [40,41] that demonstrated that bergamot polyphenolic fraction (BPF) prevented the alteration of lipid profile in hypercholesterolemic rats, counteracting oxidative stress markers and also ameliorate the dysregulation of the lipoprotein metabolism; suggesting that the richness of antioxidant properties may play an important role in improving dyslipidemia.

Cholesterol-enriched diets resulted in a remarkable change of liver histology in the PC group. A large accumulation of lipid droplets within hepatocytes of the liver is known as steatosis and has been observed microscopically. The histopathological result of liver tissue in this study is consistent with the postulation by Zheng et al. [42], who found the same hepatic architecture with the presence of large fat vacuoles in high cholesterol rats. The administration of EAMMS extracts at all dosages and simvastatin (10 mg/kg) ameliorated and reduced the hepatic lipid droplets in hepatocytes of the liver. No histological changes have been observed in the kidney tissues in all the experimental rats. These facts indicated that the EAMMS were able to inhibit the accumulation of fat in the liver due to some flavonoids in the plant extract which down-regulates the enzyme HMG-CoA reductase, the key enzyme in the process of cholesterol biosynthesis. The interaction of bioactive compounds in the plant extract with enzyme-substrate complex caused the changes of the active site of the enzyme,

thus prevents the formation of cholesterol. The bioactive compounds in *M. micrantha* may help to suppress the HMGCR activity and reduce cholesterol biosynthesis in the mevalonate pathway [43]. In agreement, the previous study was done by Gliozzi et al. [44], who also reported that the flavonoids and phenolic compounds in BPF inhibit the endogenous biosynthesis of cholesterol that mediated by HMG-CoA reductase enzyme on the mechanistic action of glycosylated polyphenols (bruteridin and melitidin), which is bind to the catalytic site of HMG-CoA reductase as an endogenous HMG-CoA substrate and causing inhibition of cholesterol synthesis.

The presence of caffeic acid ester, also known as chlorogenic acid such as 3,5-di-O-caffeoylquinic acid n-butyl ester in *M. micrantha* is reported [45,46]. The present findings suggested that these active compounds also accountable for the cholesterol-lowering activity, possibly mediated by down-regulation of HMGCR and up-regulation of LDL receptor in addition to down-regulating of ACAT2. Karthikesan et al. [47] also reported that the administration of chlorogenic acid for 45 days strongly reduced the activity of HMGCR and ACAT of lipid metabolism in rats. ACAT2 has important roles in cholesterol esterification, intestinal cholesterol absorption and ApoB-containing lipoprotein release [48]. It was reported that the inhibition of ACAT2 can effectively inhibit cholesterol absorption and reduce fat levels [49]. Phytosterols such as stigmasterol and sitosterol showed significantly reduced mRNA expression of ACAT2 activity in rodents [50]. The mechanism of the hypocholesterolemic action of *M. micrantha* extracts is summarized in Figure 4.

**Figure 4.** Proposed molecular mechanisms involved in hypocholesterolemic effects of *Mikania micrantha* stem extract (EAMMS) via inhibition of hepatic lipid accumulation, lipid peroxidation, HMG-CoA reductase (HMGCR) and ACAT2; leading to a decrease of cholesterol concentration.

#### **5. Conclusions**

The present study showed that EAMMS extracts exhibited anti-hypercholesterolemia properties by improving lipid profile, enzyme inhibitory, reducing the lipid peroxidation and lipid accumulation in combating hypercholesterolemia.

**Author Contributions:** Conceptualization, A.I. and N.H.S.; methodology, A.I.; validation, A.I., N.H.S., N.M.E., S.R.S., H.B. and M.A.A.; formal analysis, A.I.; investigation, A.I.; resources, A.I. and N.H.S.; data curation, A.I.; writing—original draft preparation, A.I.; writing—review and editing, A.I., N.H.S., N.M.E., S.R.S., H.B. and M.A.A.; supervision, N.H.S., N.M.E., S.R.S. and H.B.; project administration, N.H.S.; funding acquisition, N.H.S. and N.M.E. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by Universiti Putra Malaysia (UPM). Project No. GP-IPM/2017/9524500.

**Acknowledgments:** The authors thank UPM for funding this research project and A.I. thanks UPM for Graduate Research Fellowship. The authors also thank Amirah Haziyah Ishak, Noor Syafiqa Aqila Mohd Rosmi, and Norain Mohd Tamsir; and the laboratory assistance from the Animal House and laboratory of the Department of Nutrition and Dietetics, Faculty of Medicine and Health Sciences, UPM for their help and guidance to the project.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Article* **Purified Gymnemic Acids from** *Gymnema inodorum* **Tea Inhibit 3T3-L1 Cell Di**ff**erentiation into Adipocytes**

**Papawee Saiki 1,\*, Yasuhiro Kawano 1, Takayuki Ogi 2, Prapaipat Klungsupya 3, Thanchanok Muangman 3, Wimonsri Phantanaprates 3, Papitchaya Kongchinda 3, Nantaporn Pinnak <sup>3</sup> and Koyomi Miyazaki <sup>1</sup>**


Received: 20 August 2020; Accepted: 15 September 2020; Published: 17 September 2020

**Abstract:** *Gymnema inodorum* (GI) is an indigenous medicinal plant and functional food in Thailand that has recently helped to reduce plasma glucose levels in healthy humans. It is renowned for the medicinal properties of gymnemic acid and its ability to suppress glucose absorption. However, the effects of gymnemic acids on adipogenesis that contribute to the accumulation of adipose tissues associated with obesity remain unknown. The present study aimed to determine the effects of gymnemic acids derived from GI tea on adipogenesis. We purified and identified GiA-7 and stephanosides C and B from GI tea that inhibited adipocyte differentiation in 3T3-L1 cells. These compounds also suppressed the expression of *peroxisome proliferator-activated receptor gamma* (*Ppar*γ)-dependent genes, indicating that they inhibit lipid accumulation and the early stage of 3T3-L1 preadipocyte differentiation. Only GiA-7 induced the expression of *uncoupling protein 1* (*Ucp1*) and *ppar*γ *coactivator 1 alpha* (*Pgc1*α), suggesting that GiA-7 induces mitochondrial activity and beige-like adipocytes. This is the first finding of stephanosides C and B in *Gymnema inodorum*. Our results suggested that GiA-7 and stephanosides C and B from GI tea could help to prevent obesity.

**Keywords:** *Gymnema inodorum*; adipogenesis; gymnemic acid; obesity

#### **1. Introduction**

*Gymnema sylvestre* is a species of the genus Gymnema that is popular in India for reducing glucose levels, suppressing glucose absorption and preventing type 2 diabetes [1–5]. *Gymnema inodorum* (GI) is a species of same genus that is indigenous to Thailand, particularly in the northern region, where it is widely consumed. The effects of GI on glucose absorption and blood glucose levels have recently been investigated [6–8]. We previously found that extracts of GI leaves decreased blood glucose in alloxan-induced diabetic rats [9] that comprise a popular model with which to study type 1 diabetes mellitus. Alloxan selectively destroys insulin production in beta cells, which consequently results in high blood glucose levels [10]. However, about 90% of patients with diabetes have type 2 diabetes mellitus (DM) which is induced by a lack of exercise and inappropriate eating habits [11]. However, obesity is the leading risk factor for type 2 DM, and it also greatly increases the risk of

fatty liver disease, atherosclerosis, metabolic diseases, insulin resistance and hypertension [12,13]. Obesity is characterized at the cellular level as being differentiated from preadipocytes. White adipose tissue (WAT) is specialized to store excess energy as triglycerides composed of fatty acids. Inhibiting preadipocyte differentiation can prevent the initiation and progression of obesity [14,15].

The differentiation of 3T3-L1 fibroblast-like cells into adipocyte-like cells stimulated by insulin and synthetic glucocorticoids is a popular model of adipogenesis and lipid metabolism *in vitro* [16,17]. Therefore, we applied the inhibition of 3T3-L1 cell differentiation to screen gymnemic acid extracted from GI tea. Gymnemic acid is an oleanane-type triterpene glycoside [16,17] that can exist as a single entity or as a mixture of several related compounds [18,19]. The major saponin fraction in *Gymnema sylvestre* is a gymnemic acid that comprises a complex mixture of at least nine similar glycosides and aglycone derivatives [20]. Moreover, only four gymnemic acids have been identified in GI, which renders the purification and identification of gymnemic acids difficult. Furthermore, current knowledge about these compounds purified from GI is limited. We isolated and purified GiA-7, stephanoside C and stephanoside B from GI tea that inhibited 3T3-L1 cell differentiation. We also determined the expression of the *peroxisome proliferator-activated receptor gamma (Ppar*γ*), CCAAT*/*enhancer-binding protein alpha (Cebp*α*), cluster of di*ff*erentiation 36 (Cd36), fatty acid synthase (Fasn), ppar*γ *coactivator 1 alpha (Pgc1*α*), lipin-1, adipose triglyceride lipase (Atgl), hormone-sensitive lipase (Hsl), sterol regulatory element-binding protein (Srebp)-1c, uncoupling protein 1 (Ucp1), glucose transporter type 4 (Glut4)* and *fatty acid binding protein 4 (Fabp4)* genes to explain the signaling of adipogenesis inhibition in 3T3-L1 preadipocytes.

#### **2. Materials and Methods**

#### *2.1. Extraction, Isolation and Purification*

Fresh GI leaves (Development of Herbs and Fruit Products Community Enterprise (Chiang Mai, Thailand)) were powdered, washed, dried and then steamed for 3 min. The leaves were dried at 60 ◦C for 2 h, stir-fried to complete dryness and then stored in darkness.

After extracting GI tea powder with 98% methanol for 24 h, the extract was mixed with hexane in a separatory funnel. The lower solution was collected, evaporated to dryness and then the residue was washed with chloroform and methanol (2:1) to remove fat components. The washed, evaporated residue dissolved in methanol (crude gymnemic acid) was eluted through a Sep-Pak tC18 cartridge (Waters Corporation, Milford, MA, USA) with a gradient of 10–100% methanol and ethanol. Six active compounds were purified from the 90% methanol fraction by high-performance liquid chromatography (HPLC) using a Model CCPD computer-controlled pump (Tosoh, Tokyo, Japan) equipped with a Capcell PAK C18 5 μm, 20-mm inner diameter (i.d.), 250-mm column (Osaka Soda Co., Ltd., Osaka, Japan) and isocratic 80% methanol with 0.1% formic acid at a flow rate of 2.5 mL/min. The compounds were detected at 254 nm using a UV wavelength detector (JASCO International Co., Ltd., Tokyo, Japan).

#### *2.2. Mass Spectrometry*

Dried purified compounds were dissolved and diluted in dimethyl sulfoxide (Fujifilm Wako Pure Chemical Industries Ltd., Osaka, Japan) at 100 ppm. The accurate molecular formula was determined by Liquid Chromatography equipped with Quadrupole Time Of Flight Mass Spectrometry (LC/Q-TOF MS) using an Agilent 6530 Accurate-Mass Q-TOF LC/MS system (Agilent Technologies Inc., Santa Clara, CA, USA) equipped with an electrospray ionization (ESI) interface. Compounds were separated by reversed-phase liquid chromatography using a photodiode array detector and monitored at a wavelength ranging from 210 to 600 nm at a flow rate of 0.4 mL/min using an ACQUITY UPLC BEH C18 column (50 × 2.1 mm i.d. and 1.7 μm particle size (Waters Corp.) at 40 ◦C). The mobile phase consisted of a linear gradient of 0.1% formic acid:acetonitrile (1:1) to 0.1% formic acid:acetonitrile (1:19) over 3 min. The high-resolution mass spectra (HRMS) conditions were: positive ion mode; desolvation

gas, N2; temperature 350 ◦C, pressure, 40 psig; flow rate, 8 L/min and capillary, fragmentary and skimmer voltages of 3500, 100 and 65 V, respectively [18].

#### *2.3. Nuclear Magnetic Resonance (NMR) Spectroscopy*

Dried compound 2 (10 mg) was exchanged into methanol-d4, 99.8 atom% D, containing 0.05% (*v*/*v*) Tetramethylsilane (TMS) (Cambridge Isotope Laboratories Inc., Andover, MA, USA). Dried compounds 5 and 6 (10 mg each) were exchanged into pyridine-d5, 99.5 atom% D (Cambridge Isotope Laboratories Inc.). Spectra were determined by one-dimensional (1H NMR, 13C NMR and dept-135) and two-dimensional COrrelated SpectroscopY (COSY), Heteronuclear Multiple Bond Correlation (HMBC) and Heteronuclear Multiple Quantum Coherence (HMQC) NMR using a Bruker 500 MHz NMR (Bruker Daltonics SPR, Hamburg, Germany).

#### *2.4. Cell Culture*

The mouse embryonic fibroblast cell line (3T3-L1 cell) was purchased from National Institutes of Biomedical Innovation, Health, and Nutrition (NIBIOHN), Osaka, Japan. These cells were cultured in low-glucose Dulbecco's modified Eagle's medium (DMEM) (Fujifilm Wako Pure Chemical Corp.) containing 10% heat-inactivated fetal bovine serum (FBS; Biowest, Tokyo, Japan) at 37 ◦C under a humidified 5% CO2 atmosphere.

#### *2.5. The Antiadipocyte Di*ff*erentiation Activity*

We seeded 3T3-L1 cells (1 <sup>×</sup> 105/mL in 200 <sup>μ</sup>L) cultured as described above into collagen-coated 96-well plates in high-glucose DMEM (Fujifilm Wako Pure Chemical Corp.) under standard conditions for 24 h, then induced their differentiation into adipocytes using 10 μg/mL insulin, 0.5 mM 3-isobutyl-1-methylxanthine (IBMX) and 1 μM water-soluble dexamethasone (Sigma-Aldrich Corp., St. Louis, MO, USA). After 1 h, the 3T3-L1 cells were incubated with samples for 7–10 days.

Cell proliferation was determined using CellTiter 96® AQueous One Solution Cell Proliferation Assays (Promega Corp., Madison, WI, USA), as described by the manufacturer. The absorbance of proliferating cells determined at 490 nm using an iMarkTM Microplate Reader (Bio-Rad Laboratories Inc., Hercules, CA, USA) was compared with that of untreated differentiated 3T3-L1 cells.

Intracellular lipid accumulation was determined using a Lipid Assay Kit (Cosmo Bio Co., Ltd., Tokyo, Japan), as described by the manufacturer. Differentiated 3T3-L1 cells were washed with phosphate-buffered saline (PBS), and fixed overnight with 4% formaldehyde at room temperature. The cells were then washed twice with distilled water, incubated with Oil red O at room temperature for 15 min and washed twice with distilled water. Oil red O extraction reagent was added into cells. Absorbance was read at 540 nm using the iMarkTM Microplate Reader. Absorption due to the intracellular lipid accumulation was determined and compared with that of control-differentiated 3T3-L1 cells.

#### *2.6. Quantitative Real-Time PCR*

We incubated 3T3-L1 cells (1 <sup>×</sup> 105/mL; 1 mL) seeded into collagen-coated 12-well plates in high-glucose DMEM under standard conditions for 3 days, then induced the cells to differentiate into adipocytes using 10 μg/mL insulin, 0.5 mM IBMX and 1 μM water-soluble dexamethasone for 1 h. The 3T3-L1 cells were then incubated with purified GiA-7 and stephanosides C and B from GI tea (100 μM each) for 8 days. The expression of genes associated with adipogenesis was analyzed using quantitative real-time PCR. Total RNA was extracted from the cells using RNAiso plus. Single-stranded cDNA was generated using PrimeScriptTM RT Master Mix. Quantitative real-time PCR was conducted using a SYBR® Premix Ex Taq™ II (Takara Bio. Inc., Otsu, Japan) and a LightCyclerTM (Roche Diagnostics, Mannheim, Germany). The sequences of all primers (Thermo Fisher Scientific Inc) are listed in Table 1 [19]. The PCR conditions were 95 ◦C for 10 s, followed by 45 cycles of 95 ◦C for 5 s,

58 ◦C for 10 s and at 72 ◦C for 10 s. The amount of target mRNA was normalized relative to the internal standard *36b4*.


**Table 1.** Primer sequences for real-time reverse transcription (RT)-PCR.

#### *2.7. Statistical Analysis*

Data were statistically assessed by one-way analyses of variance (ANOVAs) with Dunnett tests using EZR software version 1.52 (Jichi Medical University, Saitama, Japan), which is graphical user interface for R (The R Foundation for Statistical Computing) based on R commander [20]. Values are indicated as means ± SD. Significant differences are shown as *p*-values.

#### **3. Results and Discussion**

We measured the ability of the crude 10–100% methanol and ethanol fractions of gymnemic acid to inhibit 3T3-L1 cell differentiation. We found that the 90% methanol fraction was the most powerful inhibitor (Figure 1). We then found that compounds 2, 3, 5 and 6 among the six compounds separated by HPLC from this fraction (Figure 2) significantly inhibited 3T3-L1 cell differentiation (Figure 3). Compound 5 was the most powerful inhibitor. The inhibition of 3T3-L1 cell differentiation by compound 5 was concentration-dependent. Compound 6 also strongly inhibited 3T3-L1 cell differentiation. However, the yield of HPLC fraction 3 was very low. Therefore, HPLC fractions No. 2, 5 and 6 were further purified, and their structures were identified by NMR and mass spectrometry. The 13C NMR chemical shifts of compounds 2, 5 and 6 were compared with published 13C NMR chemical shifts of GiA-7 [6], stephanoside C and stephanoside B [21], respectively, and are shown in Tables 2–6, respectively. These findings showed that compounds 2, 5 and 6 were GiA-7, stephanoside C and stephanoside B, respectively.

**Figure 1.** Effects of Sep-PaktC18 fractions on 3T3-L1 cell differentiation. We assessed the abilities of 10%MeOH, 30%MeOH, 50%MeOH, 70%MeOH and EtOH fractions at the concentration of 10 mg/mL in ethanol and 90%MeOH and MeOH fractions at the concentration of 1 mg/mL in ethanol to inhibit 3T3-L1 cell differentiation. Values are shown as means ± SD (*n* = 4). \* *p* < 0.05, \*\* *p* < 0.01 and \*\*\* *p* < 0.001 vs. control (ANOVA with post hoc Dunnett tests).

**Figure 2.** Compounds separated by high-performance liquid chromatography (HPLC) from 90% methanol fraction.

**Figure 3.** Ability of HPLC fractions to inhibit 3T3-L1 cell differentiation. (**a**) Inhibition of adipogenesis. (**b**) Cell proliferation. Values are shown as means ± SD (*n* = 4). \* *p* < 0.05, \*\* *p* < 0.01 and \*\*\* *p* < 0.001 vs. control (ANOVA and post hoc Dunnett tests).




**Table 2.** *Cont.*

**Table 3.** The 13C NMR chemical shifts of stephanoside C and compound **5** (δ: ppm).



**Table 4.** The 13C NMR chemical shifts of sugar chains of stephanoside C and compound **5** (δ in ppm).

**Table 5.** The 13C NMR chemical shifts of stephanoside B and compound **6** (δ: ppm).



**Table 6.** The 13C NMR chemical shifts of sugar chains of stephanoside B and compound **6** (δ: ppm).

The molecular formulae of purified compounds 2, 5 and 6 were determined using Q-TOF LC/MS in the positive ion mode. The molecular formula of GiA-7 was C44H65NO12, according to the mass spectra (*m*/*z* 800.4580 (M + H)+, calcd. *m*/*z* 800.4582). Those of stephanoside C and stephanoside B were the same: C52H79NO18, calcd. 1006.5370. The accurate masses of stephanosides C and B were *m*/*z* 1006.5383 (M + H)<sup>+</sup> and *m*/*z* 1006.5488 (M + H)+, respectively. The molecular formulae of stephanosides C and B are the same, but their sugar chains are d-thevetose and d-allomethylose, respectively. Figure S1 shows the structures of compounds 2, 5 and 6. The NMR and mass spectrometry data confirmed that compounds 2, 5 and 6 are GiA-7, stephanoside C and stephanoside B, respectively.

We assessed the ability of purified 25, 50 and 100-μM GiA-7, stephanoside C and stephanoside B extracted from GI tea to inhibit 3T3-L1 cell differentiation. After 10 days, intercellular lipid accumulation and viable cells were determined. Each of GiA-7, stephanoside C and stephanoside B at 100 μM reduced intercellular lipid accumulation (Figure 4). Stephanoside C was the most effective inhibitor, which is the lowest concentration of significantly inhibited 3T3-L1 cell differentiation. Moreover, the inhibition of 3T3-L1 cell differentiation by GiA-7, stephanoside C and stephanoside B was concentration-dependent.

**Figure 4.** Effects of compounds 2 (GiA-7), 5 (stephanoside C) and 6 (stephanoside B) on 3T3-L1 cell differentiation. (**a**) Inhibition of adipogenesis. (**b**) Cell proliferation. Values are shown as means ± SD (*n* = 4). \* *p* < 0.05, \*\* *p* < 0.01 and \*\*\* *p* < 0.001 vs. control (ANOVA and post hoc Dunnett tests).

Several markers associated with adipogenesis control 3T3-L1 cell differentiation [22]. We assessed the expression of the *Ppar*γ, *Cebp*α, *Fasn*, *Pgc1*α*, Cd36* and *Fabp4* genes that are associated with differentiation into adipocytes to determine the effects of GiA-7, stephanoside C and stephanoside B on adipogenesis. Figure 5 shows that GiA-7, stephanoside C and stephanoside B at 100 μM significantly suppressed the expression of *Ppar*γ, *Cebp*α, *Fasn* and *Cd36*. Stephanoside B and GiA-7 significantly suppressed *Fabp4* expression. Stephanosides C and B also significantly suppressed *Pgc1*α expression. These results indicated that GiA-7, stephanoside C and stephanoside B inhibited the early stage of adipogenic differentiation by inhibiting of *Ppar*γ-dependent mechanisms. Both Hsl and Atgl are phosphorylates upon appropriate physiological signaling to induce triacylglycerol (TG) lipolysis in adipocytes [23,24]. Figure 5 shows that stephanosides C, B and GiA-7 suppressed *Hsl* and *Atgl* gene expressions. These findings suggested that none of these compounds activated TG lipolysis. However, *Ppar*γ directly regulates *Hsl* and *Atgl* gene expressions in adipocytes *in vitro* [25,26]. Our results suggested that these compounds downregulated *Hsl* and *Atgl* gene expressions by inhibiting *Ppar*γ gene expression. Lipin-1 functions in lipid droplet biogenesis during adipocyte differentiation and generates diacylglycerol for lipid synthesis [27]. Lipin-1 is important for the process of TG accumulation during the early stage of adipogenesis. Lipin-1 is a key factor for adipocyte maturation and maintenance by regulating *Ppar*γ and *Cebp*α [28]. *Lipin-1* expression is required to induce the transcription of adipogenic genes, including *Ppar*γ and *Cebp*α [29,30]. Figure 5 shows that stephanosides C and B and GiA-7 significantly suppressed *lipin-1* expression. These findings suggest that these compounds inhibited *Ppar*γ and *Cebp*α gene expressions by suppressing *lipin-1* gene expression. These observations confirm that GiA-7, stephanoside C and stephanoside B inhibited the early stage of adipogenesis and prevented TG accumulation. The transcriptional cofactor, *Pgc1*α, is important for mitochondrial biogenesis. The regulation of *Pgc1*α expression enhances mitochondrial biogenesis through *Srebp-1c* upregulation [31,32]. The present study found that only GiA-7 induced *Srebp-1c* and *Pgc1*α. Srebp-1c is also a key regulator of adipocytes and is involved in lipid metabolism [33,34]. These findings suggest that GiA-7 regulates mitochondrial biogenesis through the *Srebp-1c*-dependent upregulation of *Pgc1*α. GiA-7 also inhibits lipid accumulation in 3T3-L1 preadipocytes by downregulating adipogenic transcription factors and genes associated with lipid accumulation. Both *Pgc1*α and*Ucp1* are brown/beige cell-specific genes. Only GiA-7 induced the expression of *Pgc1*α and *Ucp1*. Beige adipocytes express low basal levels of *Ucp1*, whereas brown adipocytes constitutively express *Ucp1*. These findings suggest that GiA-7 inhibits the differentiation of white adipocytes and, also, induces beige-like adipocytes in 3T3-L1 mouse preadipocytes. Comprehensive profiles of gene expressions indicate that the characteristics of human brown and mouse beige adipocytes are compatible [33,34]. The activation of human brown adipocytes was recently examined as a possible novel therapeutic treatment for obesity [35]. Thus, GiA-7 might serve as a novel treatment for obesity in humans by inducing brown adipocytes.

Gymnemic acid extracted from the leaves of *Gymnema sylvestre* comprises a mixture of triterpene glycosides that can reduce glucose levels and inhibit glucose absorption [36–38]. The aqueous extract of *Gymnema sylvestre* induces insulin secretion in MIN6 cells [39]. One study found that GiA-7 from GI leaves inhibits glucose absorption in the isolated intestinal tract and suppresses blood glucose in rats [6]. However, we found here that GiA-7 purified from gymnemic acid extracted from GI tea inhibited 3T3-L1 cell differentiation into adipocytes. Stephanoside C and stephanoside B isolated from the stems of *Stephanotis lutchuensis var. japonica* and *Gongronema nepalense* have ant-malarial activity [21,40]. This is the first report of stephanoside C and stephanoside B isolated from *Gymnema inodorum* inhibiting 3T3-L1 cell differentiation into adipocytes. As mentioned before, obesity is characterized at the cellular level as being differentiated from preadipocytes. GiA-7, Stephanoside C and stephanoside B present in GI tea inhibited preadipocyte differentiation by suppressing the *Ppar*γ-dependent mechanisms. These findings suggest that consuming GI tea could play a role in the prevention of obesity.

**Figure 5.** Effects of stephanosides C and B and GiA-7 extracted from *Gymnema inodorum* (GI) tea on gene expressions at the initial stage of 3T3-L1 cell differentiation into adipocytes. The differentiation of 3T3-L1 cells was induced, and the cells were incubated with 100 μM GiA-7, stephanoside C and stephanoside B for 8 days; then, the gene expressions were measured. Values are shown as means ± SD (*n* = 4). \* *p* < 0.05, \*\* *p* < 0.01 and \*\*\* *p* < 0.001 vs. control (ANOVA with post hoc Dunnett tests. ND, not detected).

#### **4. Conclusions**

*Gymnema inodorum* tea has been widely applied in Thailand to control high blood glucose. Here, we screened the ability of gymnemic acids extracted from GI tea to inhibit 3T3-L1 cell differentiation into adipocytes. We isolated and purified GiA-7, stephanoside C and stephanoside B from GI tea using column chromatography and C18 HPLC, respectively, then confirmed them using NMR and mass spectrometry. All three compounds inhibited 3T3-L1 cell differentiation into adipocytes. Moreover, we determined that these compounds inhibited the early stage of adipogenesis by suppressing the *Lipin-1, Ppar*γ*, Cebp*α*, Fasn, Cd36* and *Fabp4* genes that are associated with adipogenesis. However, only GiA-7 induced *Ucp1* and *Pgc1*α, suggesting that GiA-7 enhances mitochondrial activity and beige-like adipocytes among 3T3-L1 preadipocytes. Our findings suggest that the GiA-7, stephanoside C and stephanoside B from GI tea could help to prevent obesity.

#### **5. Patents**

Papawee Saiki and Yasuhiro Kawano, the methods of inhibiting fat synthesis, fat synthesis inhibitors and food and drink for suppressing fat synthesis, JP patent 2019-218481.

**Supplementary Materials:** The following is available online at http://www.mdpi.com/2072-6643/12/9/2851/s1, Figure S1: The structural formulae of compounds 2, 5 and 6.

**Author Contributions:** P.S. and P.K. (Prapaipat Klungsupya) conceptualization; P.S., Y.K. and T.O. methodology; P.S. and Y.K. validation; P.S. formal analysis; P.S. investigation; W.P., P.K. (Papitchaya Kongchinda), N.P. and T.M. resources; P.S., Y.K. and T.O. data curation; P.S. writing—original draft preparation; P.S., Y.K., T.O., P.K. (Prapaipat Klungsupya), T.M., W.P., P.K. (Papitchaya Kongchinda), N.P. and K.M. writing—review and editing; P.S. visualization; K.M. supervision; P.S. project administration; P.K. (Prapaipat Klungsupya) funding acquisition. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Thai Royal Government, Ministry of Higher Education, Science, Research and Innovation (MHESI) through the Thailand Institute of Scientific and Technological Research, TISTR (Grant Number: TISTR6334102122).

**Acknowledgments:** We thank Katsutaka Oishi and Tomoki Abe for their helpful discussions.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Article Viburnum opulus* **L. Juice Phenolics Inhibit Mouse 3T3-L1 Cells Adipogenesis and Pancreatic Lipase Activity**

#### **Małgorzata Zakłos-Szyda \*, Nina Pietrzyk, Marcin Szustak and Anna Pods ˛edek**

Department of Biotechnology and Food Sciences, Institute of Molecular and Industrial Biotechnology, Lodz University of Technology, 90-924 Łód ´z, Poland; nina.pietrzyk@dokt.p.lodz.pl (N.P.); marcin.szustak@p.lodz.pl (M.S.); anna.podsedek@p.lodz.pl (A.P.) **\*** Correspondence: malgorzata.zaklos-szyda@p.lodz.pl

Received: 4 June 2020; Accepted: 23 June 2020; Published: 6 July 2020

**Abstract:** *Viburnum opulus* L. fruit is a rich source of phenolic compounds that may be involved in the prevention of metabolic diseases. The purpose of this study was to determine the effects of *Viburnum opulus* fresh juice (FJ) and juice purified by solid-phase extraction (PJ) on the adipogenesis process with murine 3T3-L1 preadipocyte cell line and pancreatic lipase activity in triolein emulsion, as well as their phenolic profiles by UPLC/Q-TOF-MS. Decrease of lipids and triacylglycerol accumulation in differentiated 3T3-L1 cells were in concordance with downregulation of the expression of peroxisome proliferator-activated receptor-gamma (PPARγ), CCAAT/enhancer-binding protein alpha (C/EBPβ/α), and sterol regulatory element-binding protein 1c (SREBP-1c). Furthermore, regulation of PPARγ-mediated β-lactamase expression by *V. opulus* components in reporter gene assay, as well as their binding affinity to ligand-binding domain of PPARγ, were tested. In addition, the levels of enzymes involved in lipid metabolism, like fatty acid synthase (FAS) or acetyl-CoA carboxylase (ACC), were decreased, along with inflammatory cytokines, like tumor necrosis factorα (TNFα), interleukin-6 (Il-6) and leptin. Moreover, FJ and PJ were able to inhibit pancreatic lipase, which potentially could reduce the fat absorption from the intestinal lumen and the storage of body fat in the adipose tissues. Thirty-two phenolic compounds with chlorogenic acid as the dominant compound were identified in PJ which revealed significant biological activity. These data contribute to elucidate *V. opulus* juice phenolic compounds' molecular mechanism in adipogenesis regulation in 3T3-L1 cells and dietary fat lipolysis.

**Keywords:** *Viburnum opulus*; phenolic compounds; adipogenesis; PPARγ; lipase inhibition

#### **1. Introduction**

Nutrition-related chronic diseases have become the main health problem of the world of the 21st century. According to World Health Organization data based on the latest, in the European Union countries overweight and obesity affect 30–70% and 10–30% of adults, respectively [1]. Obesity is characterized as abnormal or excessive fat accumulation, which means the increase in the number (hyperplasia) and size (hypertrophy) of differentiated adipocytes [2,3]. It is known that adipose tissue is not only a reservoir of energy in the form of triacyclglycerols (TAG), but it also secretes adipocytokines, growth factors, and hormones involved in energy homeostasis and insulin sensitivity maintenance. Thus, increased adiposity is regarded as one of the most important risk factors of insulin resistance, type 2 diabetes (T2D), and nonalcoholic fatty liver disease (NAFLD), which in turn may lead to hypertension, coronary heart disease, and stroke [4]. Among the most important cellular regulators of lipid metabolism are peroxisome proliferator-activated receptors (PPARs), belonging to nuclear receptors proteins [5]. There are three different receptor isotypes: PPARα, PPARβ/δ, and

PPARγ, among which PPARγ is involved in the regulation of adipocyte differentiation. Ligand binding to the PPARγ ligand-binding domain (LBD) leads to a conformational change and switching of nuclear receptor corepressors to coactivators [6]. Agonists activate PPARγ, which, after binding with retinoic X receptor and peroxisome proliferator responsive element (PPRE) within the promoter of target genes activates their transcription [7]. Not only PPARγ but also other transcription factors, such as CCAAT/enhancer-binding protein alpha (C/EBPβ/α) and sterol regulatory element binding protein-1c (SREBP-1c), are involved in adipocyte differentiation [8]. These factors stimulate gene expression and the expression of other proteins involved in lipid synthesis and storage, such as fatty acid synthase (FAS) or acetyl-CoA carboxylase (ACC). What is more, obese adipose tissue contributes to the elevation of inflammatory cytokines, such as tumor necrosis factor α (TNFα) or interleukin-6 (Il-6), leading to chronic inflammation state promoting insulin resistance and cancer development [9]. PPARγ agonist type of medicines, like glitazones, significantly improve glycemic control, but via promotion of free fatty acid uptake and accumulation of TAG in adipose tissue lead to weight gain, and increases in heart or renal failure [10–12]. Thus, due to the variety of induced side effects, less harmful plant-derived agents are searched as PPARγ regulators able to prevent insulin resistance without weight obesity gaining. Other mechanisms allowing the decrease of fatty acids intestinal absorption is the usage of pancreatic lipase inhibitor, which decreases the hydrolysis of diet-originated TAG into glycerol and fatty acids. Since phenolic compounds constitute an important part in the human diet, they have recently emerged as critical phytochemicals in obesity prevention and treatment [13,14]. Among fruits rich in these secondary metabolites are those of *Viburnum opulus* L. (*V. opulus* L.), known as guelder rose, or the European cranberry bush rose [15,16]. Despite its fruit bitterness it can be found in food products such as juice, jams, jellies, marmalades, sauces, herbal tea, cordials, and liqueurs, as well as fermented drinks [17]. Our previous studies showed that due to their high antioxidant potential *V. opulus* phenolics decreased chemically generated intracellular oxidative stress under *in vitro* conditions, as well as possessed different anticancer activity with apoptosis induction and inhibition of cell migration [16,18,19]. Further studies revealed *V. opulus* fruit phenolics' impact on carbohydrate metabolism as α-amylase, α-glucosidase, protein tyrosine phosphatase-1B, and dipeptidyl peptidase-4 enzyme inhibitors [20,21]. More detailed analysis showed that *V. opulus* phenolics decreased the uptake of free fatty acids and lipids accumulation in human epithelial Caco-2 cells [15]. Nevertheless, they inhibited glucose-stimulated insulin secretion in mice insulinoma MIN6 cells, as well as increased free fatty acid uptake and lipid droplets accumulation [21]. Taking into account potent phenolics' impact on the modulation of cellular metabolism, in the present study we investigated an influence of *V. opulus* fresh juice (FJ) and purified juice (PJ) on pancreatic lipase activity and adipogenesis process in mouse preadipocyte 3T3-L1 cell line [22]. In addition, phenolic components of FJ and PJ were identified by the UPLC-MS method. Studies also assessed the influence of these preparations on the expression of transcription factors (PPARγ, C/EBP, SREBP1c) and other proteins related to adipogenesis (FAS, ACC, TNFα, Il-6, leptin, adiponectin). Furthermore, the *V. opulus* components' influence on the regulation of PPARγ activity was elucidated.

#### **2. Materials and Methods**

#### *2.1. Chemicals and Reagents*

Acetonitrile (Merck, Darmstadt, Germany) and formic acid (Sigma-Aldrich, Steinheim, Germany) were hyper grades for LC-MS. Folin–Ciocalteu reagent was obtained from POCH (Gliwice, Poland). The reference compounds were obtained from Sigma-Aldrich (Steinheim, Germany) ((+)-catechin, (−)-epicatechin, rutin, gallic acid), Extrasynthese (Lyon, France) (chlorogenic acid, cyanidin 3-glucoside, quercetin 3-glucoside, isorhamnetin, and isorhamnetin 3-rutinoside) and Phytolab (Vestenbergsgreuth, Germany) (neochlorogenic acid, procyanidin B1, and procyanidin B2). Ultrapure water (Simplicity®Water Purification System, Millipore, Marlborough, MA, USA) was used to prepare

all the solutions. All cell culture reagents were obtained from Life Technologies (Carlsbad, CA, USA). Other chemicals used, if not stated otherwise, were obtained from Sigma-Aldrich (Steinheim, Germany).

#### *2.2. Preparation of V. opulus Samples, Identification and Quantitative Determination of Individual Phenolic Compounds by UPLC–PDA-Q*/*TOF-MS*

Fruits of the *V. opulus* were collected from Rogów Arboretum, Warsaw University of Life Sciences (Rogów, Poland) (account number 18162). After fruit pulp homogenization and centrifugation (5000 rpm for 10 min), fresh juice (FJ) was obtained. FJ purification by solid phase extraction with C-18 Sep-Pak cartridge (10 g capacity, Waters Corp., Milford, MA, USA; 12-Port Vacuum Manifold system) and methanolic elution processes were performed. After methanol removal under reduced pressure (T < 40 ◦C), solid residue was dissolved in water and lyophilized to purified juice (PJ). Phenolic compounds were identified using the Acquity ultraperformance liquid chromatography (UPLC) system coupled with a quadruple-time of flight mass spectrometry (Q/TOF-MS) instrument (Waters Corp., Milford, MA, USA) equipped with an electrospray ionization (ESI) source. The separation of individual phenolics was carried out using an Acquity UPLCR HSS T3 C18 column (150 × 2.1 mm, 1.8 μm; Corp., Milford, MA, USA) at 30 ◦C. The mobile phase was a mixture of 0.1% formic acid (A) and acetonitrile (B). The gradient program was as follows: initial conditions 99% (A), 12 min 65% (A), 12.5 min 100% (B), 13.5 min 99% (A). The flow rate was 0.45 mL/min and the injection volume was 5 μl. The mass spectrometer was operating in the negative mode for a mass range of 150–1500 Da, fixed source temperature at 100 ◦C, desolvation temperature 250 ◦C, desolvation gas flow of 600 L/h, cone voltage of 45 V, a capillary voltage of 2.0 kV, and a collision energy of 50 V. Leucine enkephalin was used as a lock mass. The instrument was controlled by Mass-LynxTM V 4.1 software (Waters Corp., Milford, MA, USA). The runs were monitored at the following wavelengths: flavanols at 280 nm, hydroxycinnamic acids at 320 nm, flavonols at 360 nm, and anthocyanins at 520 nm. Photodiode detector (PDA) spectra were measured over the wavelength range of 200–600 nm. Calibration curves were run for the external standards: (+)-catechin, procyanidin C1, neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, caffeic acid, and quercetin 3-rutinoside, and quercetin 3-glucoside. Phenolic compounds were identified using their UV-Vis characteristic, MS and MS2 properties using data gathered in-house and from the literature.

#### *2.3. Inhibition Assay for Pancreatic Lipase Activity*

Inhibitory activities of FJ and PJ were expressed as the IC50 values (half-maximal inhibitory concentration). Orlistat was used as a positive inhibitor control. The IC50 value was concluded from the graph of lipase inhibition (%) versus the concentration of juices or Orlistat per 1 mL of the reaction mixture under assay conditions. The pancreatic lipase activity was tested by measuring the fatty acids released from emulsified triolein (0.6 g of triolein, 25 mL of Tris-buffer, and 0.4 g of bile acids) according to the method described previously [23]. Briefly, 0.3 mL of FJ, PJ, and orlistat solutions diluted with buffer were mixed with 0.5 mL of the triolein emulsion and pre-incubated at 37 ◦C for 5 min before adding lipase supernatant (0.063 mL). Blanks with buffer instead of the lipase supernatant were prepared for background correction. The control consisted of all solutions without inhibitor. Finally, the reaction mixtures were incubated in a shaking bath (200 rpm) at 37 ◦C for 30 min. The reaction was terminated by adding 0.23 mL of HCl. Then, 3 mL of isooctane was added and vortexed for 0.5 min. The upper layer (2 mL) was collected, followed by the addition of 0.4 mL copper reagent (5% copper acetate, pH 6.1 regulated by pyridine). After vortexing for 1 min, the upper layer was centrifuged at 10,000 rpm for 10 min and its absorbance (A) was measured at 720 nm against a reagent blank. All samples were assayed in triplicate. Percent inhibition of pancreatic lipase activity was calculated using the formula:

$$\text{Lippase inhibition (\%)} = \left[ (\text{A}\_{\text{c}} - \text{A}\_{\text{cb}}) - (\text{A}\_{\text{s}} - \text{A}\_{\text{sb}}) \right] / \left( \text{A}\_{\text{c}} - \text{A}\_{\text{cb}} \right) \times 100 \tag{1}$$

where Ac is the absorbance of the control, Acb is the blank control absorbance, As is the sample absorbance, Asb is the sample blank.

#### *2.4. Cell Culture and Exposure Conditions*

Mouse preadipocytes 3T3-L1 were supplied by ATCC (Manassas, VA, USA). Preadipocytes were grown in Dulbecco s Modified Eagle s Medium (DMEM) medium with high glucose supplemented with 10% bovine calf serum. For adipocyte differentiation a confluent culture of 3T3-L1 cells was grown for two days in a preadipocyte medium DMEM with 10% calf serum, then the cells were stimulated with a differentiation medium with DMEM containing 10% fetal bovine serum (FBS), 1 μM dexamethasone, 0.5 mM methylisobutylxanthine (IBMX), and 1 μg/mL insulin for two days. After 48 h of incubation, the differentiation medium was replaced with DMEM containing 10% FBS and 1 μg/mL insulin [22]. Cell medium was replaced at 2-day intervals with the addition of compounds studied. Analyses were carried out 7 days after differentiation if not stated otherwise. To perform biological activity assays, a stock solution of PJ at concentration 100 mg/mL in 50% dimethyl sulfoxide (DMSO) was prepared and further dilutions were made with culture medium. The sample's concentrations used in biological studies are presented in the descriptions of the tests carried out. All cell culture experiments were performed in a humidified 5% CO2 and 95% atmosphere at 37 ◦C. Tissue culture plastics were supplied by Greiner Bio-One GmbH (Frickenhausen, Austria). All the experimental measurements were performed using the Synergy 2 BioTek Microplate Reader (BioTek, Winooski, VT, USA). Microscopic observations were performed using contrast-phase and fluorescent microscope Nikon TS100 Eclipse (Nikon, Tokyo, Japan) under 200 × magnification, if not stated otherwise.

#### *2.5. Cell Viability*

The effects of FJ and PJ on cell viability were assayed with the PrestoBlue reagent. The 3T3-L1 preadipocytes were seeded into a 96-well plate at a density of 104 cells/well overnight. Two days after confluence, cells were treated with series of extracts concentrations for 48 hours. The final concentration of DMSO did not exceed 0.005%. Then, the PrestoBlue reagent was added for 30 min and fluorescent signal at F530/590 nm was measured. For cell visualization, 2 μM calcein AM (Thermo Fisher Scientific, Waltham, MA, USA) was directly added to the cells.

#### *2.6. Detection of Intracellular Reactive Oxygen Species Generation*

The 3T3-L1 preadipocytes were seeded into a 96-well plate at a density of 104 cells/well. After the cells' treatment with extracts, the cells were washed with phosphate buffer saline (PBS) and incubated with DMEM and 10 μM of dichloro-dihydro-fluorescein diacetate (DCFH-DA) dye. Fluorescence intensity at F485/530 nm was determined after 30 min incubation. Five-hundred μM *tert*-BOOH (*t*-BOOH) was used as a positive control. The intracellular fluorescence of cells was observed after cells treatment with chemicals under a fluorescence microscope.

#### *2.7. Determination of Lipid Accumulation, Free Fatty Acid Uptake and Triglyceride Content*

The 3T3-L1 preadipocytes were seeded into a 96-well plate at a density of 104 cells/well for each of the experiment. The lipid content in the mature adipocytes was determined using the Nile red staining method. After cell incubation with FJ and PJ, cells were washed with cold PBS and fixed in 5% paraformaldehyde for 30 min. Then, the cells were stained with Nile red (1 μg/mL) for 40 min and fluorescence intensity at F485/530 nm was measured. For cell nuclei visualization, to fixed cells, 1 μg/mL 4 ,6-Diamidine-2 -phenylindole dihydrochloride (DAPI) stain was added. The measurement of fatty acid fluorescent probe TF2-C12 uptake by cells was performed with the Fatty Acid Uptake Kit (Sigma-Aldrich, Seinheim, Germany). After the cells' treatment with the preparations, the fluorescent signal at F485/530 nm was measured after 1 h incubation with fluorescent analogue.

Triglyceride content was measured using the Triglyceride Colorimetric Assay kit (Cayman Chemical, Ann Arbor, MI, USA). To perform the experiment cells were seeded into a 6-well plate at a density of 2 <sup>×</sup> 10<sup>5</sup> cells/well. Following treatment, the differentiated 3T3-L1 adipocytes were rinsed with PBS, harvested with a cell scraper, lysed with 1% Triton X-100 and the total triglyceride content was assessed according to the manufacturer's instructions with an absorbance measurement at 540 nm.

#### *2.8. Measurement of Adipolysis*

To perform, the experiment cells were seeded into a 24-well plate at a density of 4 <sup>×</sup> <sup>10</sup><sup>4</sup> cells/well. After treatment with FJ and PJ, the cells were washed with PBS and incubated with an induction solution for 1 h. Then the medium was collected, and glycerol released into the medium was measured at 570 nm using a colorimetric assay Adipolysis Assay Kit (Sigma-Aldrich, Steinheim, Germany) and following the manufacturer's instructions. As a positive control (lipolysis inducer), 10 μM isoproterenol was used.

#### *2.9. Gene Expression Analysis*

To perform, the experiment cells were seeded into a 6-well plate at a density of 2 <sup>×</sup> 105 cells/well. After cell incubation with FJ and PJ, the total RNA was extracted with the GeneMatrix Universal RNA Purification Kit (Eurex Ltd., Gdansk, Poland) according to the manufacturer's procedure. RNA samples were purified with an amplification Grade DNase I (Sigma-Aldrich, Steinheim, Germany), and reverse transcribed with the NG dART RT Kit (Eurex Ltd., Gdansk, Poland). Real-time RT-PCR was carried out using the SG qPCR Master Mix (Eurex Ltd., Gdansk, Poland) on a BioRad CFX96 qPCR System (Bio-Rad, Hercules, CA, USA). Complementary DNA representing 6 ng of total RNA per sample was subjected to 40 cycles of PCR amplification. Samples were first incubated at 95 ◦C for 40 s, then at 55 ◦C for 30 s, and finally at 72 ◦C for 30 s. To exclude non-specific products and primer-dimers, after the cycling protocol, a melting curve analysis was performed by maintaining the temperature at 52 ◦C for 2 s, followed by a gradual temperature increase to 95 ◦C. The threshold cycle (Ct) values for that gene did not change in independently performed experiments. The level of target gene expression was calculated as 2−ΔΔCt, where ΔΔCt <sup>=</sup> [Ct(target) <sup>−</sup> Ct(βactin)]sample − [Ct(target) − Ct(βactin)]. The following primer sequences were used to determine the genes' expression: CREB-binding protein (*CBP*): forward primer (F) TTACAACAGGCCAGGTTTCC, reverse primer (R) GGCTGGCGACATACAGTACA; sterol regulatory element binding transcription factor 1 (*SREBP1*): (F) TGTTGGCATCCTGCTATCTG, (R) AGGGAAAGCTTTGGGGTCTA; β*-actin*: (F) CCACAGCTGAGAGGGAAATC, (R) AAGGAAGGCTGGAAAAGAGC; *adiponectin*: (F) AGATGGCACTCCTGGAGAGAAG, (R) ACATAAGCGGCTTCTCCAGGCT; *leptin*: (F) GGATCAGGTTTTGTGGTGCT, (R) TTGTGGCCCATAAAGTCCTC; fatty acid synthase (*FAS*): (F) TTGCTGGCACTACAGAATGC, (R) AACAGCCTCAGAGCGACAAT; peroxisome proliferator activated receptor gamma (*PPAR*γ): (F) GCGGAAGAAGAGACCTGGG, (R) AGAACGTGACTTCTCAGCCC; interleukin-6 (*Il-6*): (F) GTCCTTCCTACCCCAATTTCCA, (R) TAACGCACTAGGTTTGCCGA; CCAAT/enhancer binding protein (*C*/*EBP*) (F) GTGTGCACGTCTATGCTAAACCA, (R) GCCGTTAGTGAAGAGTCTCAGTTTG; tumor necrosis factor α (*TNF*α): (F) GGGATCTGCTCCGCGGTTGT, (R) TCCGCGGCCAGGAGAACTGT; acetyl-Coenzyme A carboxylase alpha (*ACC*): (F) GGGGATCTCTGGCTTACAGG, (R) ATCGCATGCATTTCACTGCT; fatty acid translocase (*FAT*/*CD36*), (F) TGGCCTTACTTGGGATTGG, (R) CCAGTGTATATGTAGGCTCATCCA.

#### *2.10. Western Blotting*

To perform the experiment cells were seeded into a 6-well plate at a density of 2 <sup>×</sup> 105 cells/well. To prepare the total cell lysates, monolayers of 3T3-L1 adipocytes were scraped and lysed in Mammalian Protein Extraction Reagent (M-PER) containing protease and phosphatase inhibitors cocktail (Thermo Scientific, Waltham, MA, USA). Then, the lysates were centrifuged at 13,000 rpm for 5 min, and the supernatants of cell lysates were separated. The protein quantification was measured using the Protein Assay Dye Reagent Concentrate (Bio-Rad Laboratories BmbH, München, Germany). Each 20 μg of protein samples were separated by 8% or 10% sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to 0.45 μm nitrocellulose blotting membrane (GE Healthcare, Chicago, IL, USA). The membranes were blocked using 5% bovine serum albumin (BSA) in tris-buffered saline containing 0.1% Tween-20 (TBST) for 2 h at room temperature and incubated with primary antibodies overnight at 4 ◦C diluted 1:1000 in the same solution. Polyclonal rabbit antibodies targeting PPARγ (#2435), RxRα (#5388), CBP (#7389), acetyl-CoA carboxylase (#3676), phospho-acetyl-CoA carboxylase (Ser79) (#3661), phospho-AMPKα (Thr172) (#2531) and βactin (#4967) were purchased from Cell Signaling Technology (Danvers, MA, USA), SREBP-1c (14088-1-AP) from Proteintech Group (Manchester, UK), and p-IRS-1 (Ser307) from Santa Cruz Biotechnology (Dallas, TX, USA). Afterwards, the membranes were washed three times with TBST, then incubated for 1 h at room temperature with horseradish peroxidase (HRP)-conjugated secondary anti-rabbit antibody (#7074, Cell Signaling Technology, Danvers, MA, USA) diluted 1:3000 in 5% nonfat dry milk in TBST. After that membranes were rewashed three times with TBST. The proteins were visualized using an enhanced chemiluminescent SuperSignal West Pico Trial Kit (Thermo Scientific, Waltham, MA, USA). The ChemiDocTM MP Image System with Image LabTM 5.1 software (Bio-Rad Laboratories, Hercules, CA, USA) was used for acquisition and densitometric analysis of western blot images. Relative protein band intensity was normalized to β-actin and quantified with respect to control cells.

#### *2.11. Determination of Selected Proteins Levels*

To perform, the experiment cells were seeded into a 6-well plate at a density of 2 <sup>×</sup> 105 cells/well. On the last day of cells treatment the medium was collected and protein concentrations of adiponectin (Adiponectin Mouse ELISA Kit, Abcam, Cambridge, GB), leptin (Leptin Mouse ELISA Kit, Abcam, Cambridge, GB), Il6 (Mouse IL6 ELISA kit, Biorbyt Ltd., Cambridge, GB) and TNFα (Mouse TNF alpha ELISA kit, Biorbyt Ltd., Cambridge, GB), were determined using ELISA kits, following the manufacturer's instructions. The PPARγ protein level present in the nuclear fraction of the cellular lysates was determined with the PPARγ Transcription Factor Assay Kit (Abcam, Cambridge, GB).

#### *2.12. Reporter Gene Assay*

Cellular activation of the PPARγ nuclear receptor was assessed in the reporter gene assay GeneBLAzer®PPAR gamma 293H DA (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's protocol. In brief, cells stably expressing specific PPARγ ligand-binding domain fusion protein and UAS-β-lactamase reporter gene were incubated with tested extracts for 16 h, the cells were loaded with cell-permeable LiveBLAzer FRET B/G substrate, after 2 h incubation fluorescence intensities at 460 and 530 nm emission following excitation at 420 nm were measured. After subtraction of fluorescence background from cell-free wells, the ratio of fluorescence intensity at 460 versus 530 nm was calculated. As the PPARγ inhibitor 0.1 μM T0070907 was used, whereas 1 μM rosiglitazone as the activator.

#### *2.13. Molecular Modeling*

To determine the potential interaction between analyzed FJ or PJ components and the PPARγ receptor, molecular docking simulation was performed. An X-ray crystal structure for human PPARγ receptor in complex with rosiglitazone (5YCP) was obtained from the Protein Data Bank database (http://www.rcsb.org/) as a .pdb file [24]. Subsequently, the protein molecule model was prepared to docking procedure (hydrogen addition and grid box coordinates determination) with the AutoDock Tolls software. Then the protonation state of the side chains was defined with the PROPKA software (available at http://nbcr-222.ucsd.edu/pdb2pqr\_2.0.0/). Structures of rosiglitazone, chlorogenic acid, and (+)- catechin were downloaded from the ZINC database (http://zinc.docking.org/substances/ home/). The procyanidin B1 and C1 structures were built with ChemSkech software. The obtained files were converted to .pdb file with the OPENBABEL tools http://www.cheminfo.org/Chemistry/ Cheminformatics/FormatConverter/index.html). Then, all the ligand molecules were prepared for

further modeling (determine rotating bonds) with the AutoDock Tools and saved as .pdbqt file. The docking of the prepared ligand to PPARγ receptor was performed with Autodock Vina docking software. [AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization and multithreading]. The size of the grid box was set to 34 × 30 × 30 Å from the center of the binding pocket. The value of the exhaustiveness parameter was set to 150. Images of the docked ligands were shown with the use of the AutoDock Tools software.

#### *2.14. Statistical Analysis*

Unless stated otherwise, all the biological results are presented as means of 3–6 repeated experiments ± SEM. All calculations were evaluated for significance using one-way ANOVA followed by Dunnett's test with the GraphPad Prism 6.0 software (GraphPad Software, Inc., La Jolla, CA, USA). *p* ≤ 0.05 was considered statistically significant.

#### **3. Results and Discussion**

#### *3.1. Identification and Content of Phenolic Compounds in V. opulus FJ and PJ with the UPLC–PDA-Q*/*TOF-MS Method*

The results of the qualitative and quantitative analysis of *V. opulus* FJ and PJ determined by the UPLC/MS method are presented in Figure 1 and Tables 1 and 2. A total of 29 and 30 phenolic compounds were positively identified in FJ and PJ, respectively, on the basis of the complementary information provided by PDA and the ESI-MS detection, and literature data. The FJ and PJ showed the presence of different groups of phenolic compounds, such as hydroxycinnamic acids, flavanols, flavonols, and anthocyanins. Significant differences (*p* ≤ 0.05) in the content of individual phenolic compounds between FJ and PJ were noted as a result of about a 90-fold increase in the content of phenolics in PJ (878.6342 mg/g) compared to FJ (11.508 mg/g). Thus, it could be concluded that the Solid-Phase Extraction (SPE) method on a Sep-Pak C18 column is efficient for removing non-phenolic compounds from the juice. In the published studies, total phenolic determined by the Folin-Ciocalteu method content in the *V. opulus* fruit juices varied from 5.47 to 11.70 mg/g [25].

Hydroxycinnamic acids were the most predominant phenolic group found in both samples and constituted 80.46% and 77.50% of the sum of the phenolics in PJ and FJ, respectively. Chlorogenic acid (peak 9) with the negative molecular ion [M − H]- of 353 with typical fragments of quinic acid ester (m/z at 191) showed the highest content, both among phenolic compounds and hydroxycinnamic acids. Its concentrations were 645.492 mg/g and 8.039 mg/g in PJ and FJ, respectively. Chlorogenic acid, as the main phenolic compound in fresh berries of *V. opulus* and also in fresh juice, have been reported previously, where it constituted more than 96.2% of hydroxycinnamic acid derivatives [1–4]. Quantitatively, a second compound from hydroxycinnamic acids identified in FJ and PJ was caffeoylquinic acid (peak 13) with the negative molecular ion [M – H]- of 353 and fragment ions m/z at 191, 133. Additionally, two caffeoylquinic acids with the known structures, such as neochlorogenic acid (peak 1) and cryptochlorogenic acid (peak 10), were identified. Besides, five caffeoylquinic acid derivatives (peaks 3, 4, 7, 21, 22) were present in both samples. On the other hand, feruloylquinic acids (peaks 24 and 27) were only identified in PJ. Quantitatively, the second phenolic subgroup of phenolics was flavanols, which constituted 19.52 and 16.30% of FJ and PJ total phenolics, respectively (Table 2). The main flavan-3-ol isomers were (+)-the catechin (peak 6) and procyanidin dimer B1 (peak 2). Also, the procyanidin dimer B2 (peak 11) and procyanidin trimer C1 (peak 17) were identified in both samples based on reference substances. Furthermore, other procyanidin dimer and trimer, which were identified previously, were presented in our samples [26].


identificationandquantificationofphenoliccompoundsin



n.d.—not detected; dimeric adduct. Results are expressed as a mean ± standard deviation (*<sup>n</sup>* = 3). The values expressed differ significantly (one-way ANOVA and Duncan's test, *p* 0.05). 1—identification based on a comparison of retention time, UV-vis spectra (200–600 nm) and MS data for standards; 2,3,4,5—identification based on a comparison of molecular ionsand typical ion fragments with published data; 2—[6], 3—[7], 4—[8], 5—[9], 6—[10].


**Table 2.** The total content of the different phenolic groups in *V. opulus* fresh juice and purified juice according to UPLC–PDA-Q/TOF-MS analysis.

Results are expressed as a mean ± standard deviation (*n* = 3). The values differ significantly (one-way ANOVA and Duncan's test, *p* ≤ 0.05).

*V. opulus* fruits are characterized by varied content of anthocyanins, with cyanidin glycosides, such as glucoside, rutinoside and sambubioside as the main pigments [15,27]. In the tested FJ and PJ, cyanidin-3-glucoside (peak 16) was the main anthocyanin with the content 13.583 mg/g for PJ and 0.139 mg/g of FJ. Content of cyanidin-3-sambubioside (peak 15) and cyanidin-3-rutinoside (peak 19) were similar in both samples. The total content of anthocyanins was 0.300 mg/g of FJ and 25.665 mg/g of PJ. Perova et al. identified ten cyanidin glycoside in *V. opulus* fruits with cyanidin-3-glucoside and cyanidin-3-xylosyl-rutinoside as the main compounds [15]. Among the phenolic compounds estimated, flavonols occurred at the lowest concentration with 2.827 mg/g of PJ and only 0.043 mg/g of FJ (Table 2). *V. opulus* PJ and FJ contained quercetin-3-vicianoside (peak 25), quercetin-3-rutinoside (peak 29), quercetin-3-rhamnoside (peak 32) and quercetin-3-galactoside (peak 28), whereas the latter has not been determined in FJ. There are reports that also isorhamnetin glycosides have been found in *V. opulus* fruit and juice [15,18].

#### *3.2. Inhibition of Pancreatic Lipase by V. opulus FJ and PJ*

Pancreatic lipase (EC 3.1.1.3; triacylglycerol acyl hydrolase) breaks down TAG into absorbable monoacylglycerols and free fatty acids. Pancreatic lipase is responsible for the hydrolysis of 50–70% of total dietary fats in the intestinal lumen [28]; thus, its inhibition could reduce the storage of body fat in the adipose tissues. In the present study, the effect of FJ and PJ on pancreatic lipase was determined in a triolein emulsion by the spectrophotometric method with a copper reagent. This lipid substrate is a triglyceride formed by esterification of the three hydroxy groups of glycerol with oleic acid. As shown in Figure 2, both *V. opulus* samples exhibited a dose-dependent inhibitory effect on pancreatic lipase activity. The concentration of PJ inhibiting lipase activity to 50% was equal to 55.26 ± 2.54 mg/mL and showed 4.7-fold higher inhibitory activity than FJ (IC50 = 261.94 ± 2.00 mg/mL). All these samples were less potent in pancreatic lipase inhibiting than Orlistat (a well-known anti-lipase agent). The IC50 value for lipase inhibition for Orlistat was 0.380 ± 0.004 μg/mL. The lower anti-lipase activity of plant extracts than Orlistat is associated with the presence of other components in the extracts that do not affect the activity of the enzyme. In addition, fruit juice contains several classes of phenolic compounds, which in combination might differently affect the lipase activity due to the different interactions between them, as well as with other fruit components.

**Figure 2.** Inhibitory effect of fresh juice (FJ) (**A**), purified juice (PJ) (**B**), and Orlistat (**C**) on pancreatic lipase. Data are means of triplicate assays ± standard deviations.

In contrast to the widely studied various fruits, literature data on the anti-lipase activity of juices and drinks are scarce. Gironés-Vilaplana et al. demonstrated a significant increase in isotonic citric acid drink and isotonic lemon juice drinkability to inhibit pancreatic lipase after enriching them in açai, blackthorn or maqui berries [29]. Magui berry exhibited the most potent inhibitory activity, and the anti-lipase properties of the drinks were positively correlated with anthocyanins amounts. A similar relationship was found by Fabroni and others [30]. Among 13 fruit extracts, juices, plant tissues, legume seeds, cereals, and vegetables, "Moro" orange juice with the highest anthocyanin content, had the strongest inhibitory effect on pancreatic lipase. The orange juice lipase inhibition IC50 value of 0.46 mg/mL was higher than the IC50 of Orlistat (0.064 mg/mL). In the present study, anthocyanins constituted only 2.61 and 2.92% of the sum of the phenolics in FJ and PJ, respectively. Chlorogenic acid, quantitatively the main phenolic compounds in both samples, has also been demonstrated as a pancreatic lipase inhibitor with an IC50 value of 286.5 μM, however it has rather weak inhibitory potential compared to orlistat with an IC50 = 1.2 μM [31]. On the other hand, Worsztynowicz et al. demonstrated that chlorogenic acids isolated from chokeberry fruit did not inhibit the pancreatic lipase [32]. Some authors suggest that the inhibition of the pancreatic lipase by fruits is attributed to proanthocyanidins [23,28,33]. This group of polyphenols constituted 19.52% and 16.30% of the sum of the phenolics in FJ and PJ, respectively. Most likely, the anti-lipase potential of fruit extracts and juices, which are characterized by a complex phenolic composition, is the result of their mutual interactions. Other components of the samples analyzed may also influence their inhibitory activity toward pancreatic lipase. In the present study, IC50 values for lipase inhibition by FJ and PJ differed almost five times, despite the fact that PJ showed almost a ninety times higher concentration of phenolic compounds than FJ.

#### *3.3. The E*ff*ects of V. opulus on Cellular Viability, Adipogenesis and Adipolysis*

To check the influence of *V. opulus* on the metabolic activity of 3T3-L1 cells were exposed to increasing concentrations of extracts (from 10 to 200 μg/mL) for 48 h after reaching confluence. As is presented in Figure 3A, fresh juice inhibited cell viability by almost 35% at the highest concentration used. Dosages of FJ not exceeding 100 μg/mL had no cytotoxic effect on cells. Purified juice (PJ) showed higher cytotoxic potential, because its 100 μg/mL concentration decreased metabolic activity by almost 65% (Figure 3B). Within the dosages studied, the IC50 cytotoxicity parameter was obtained only for PJ (IC50 ≈ 85 μg/mL). The highest non-cytotoxic concentrations (IC0) chosen for studies of adipogenesis regulation were 100 μg/mL for FJ and 25 μg/mL for PJ, respectively.

**Figure 3.** The influence of *V. opulus* FJ and PJ on 3T3L-1 cell metabolic activity determined by the PrestoBlue assay after 48 h exposure of FJ (**A**) and PJ (**B**); control cells were not exposed to any compound; values are means ± standard deviations from at least three independent experiments, *n* ≥ 12; statistical significance was calculated versus control cells (untreated), \*\*\* *p* ≤ 0.001. Morphology of 3T3L1 cells observed on the 5th day of the cell differentiation process (**C)** with 25 and 50 μg/mL of PJ, and 100 μg/mL of FJ; randomly chosen fields were photographed at × 200 phase-contrast and a fluorescent microscope (cells stained with 2 μM calcein AM).

The lack of cytotoxic effects of the *V. opulus* samples used at the IC0 concentration was also confirmed with microscopic observations of the adipocytes performed 5 days after the initiation of their differentiation (Figure 3C). Control cells appeared more rounded, had lipid droplets visible and strong cytosolic green fluorescence of calcein after staining with calcein AM ester. Simultaneously, cells incubated with PJ used in its cytotoxic concentration (50 μg/mL) had lower cytoplasmic esterase activity; thus, decreased green fluorescence of calcein was observed, and the cells were more irregular and elongated in their shape. The IC0 dosages obtained for FJ and PJ seem to be comparable with those previously observed for Caco-2 and MIN-6 cells; however, 3T3-L1 cells were more sensitive to the compounds studied—probably due to the longer incubation time [18,21].

The differentiation process of preadipocyte to mature adipocyte is associated with an increase of lipid droplets formation [22]. The cells staining with Nile red, a hydrophobic fluorescent dye that accumulates in neutral lipid droplets, allowed us to determine the *V. opulus* influence on cellular lipid droplets formation and accumulation (Figure 4).

**Figure 4.** Influence of *V. opulus* FJ and PJ preparations on the accumulation of lipid droplets in 3T3-L1 cells stained with Nile red observed on the 7th day of differentiation (**A**). The control cells were not exposed to any compound; the values are means ± standard deviations from at least three independent experiments, *n* ≥ 12; the statistical significance was calculated versus the control cells \*\* *p* ≤ 0.01, \*\*\* *p* ≤ 0.001. Cell lipid droplets with Nile red were visualized under a fluorescent microscope (200× magnification) (**B**). DAPI staining allowed for the visualization of cell nuclei with 25 μg/mL of PJ and 100 μg/mL of FJ.

As is presented in Figure 4A, the FJ at 100 μg/mL significantly reduced the formation of lipid droplets compared to the control cells, whereas PJ was able to decrease lipid accumulation by 25% at 25 μg/mL. These data were confirmed by microscopic observation (Figure 4B). As evident by Nile red staining, PJ reduced the accumulation of lipid droplets in cytoplasm, as well as the size of droplets formed was decreased. The intracellular content of TAG was also quantified. The data show that in the presence of the preparations used at IC0 dosage the level of TAG was reduced by 20–25% (Figure 5A). It can be also seen that the PJ used at concentrations higher than 50 μg/mL was the most active inhibitor of lipid accumulation. However, it needs to be emphasized that observed lipid content reduction resulted from decreased cell viability (Figure 3B).

**Figure 5.** The influence *V. opulus* extracts on the triglyceride level (**A**) and fatty acid analog TF2-C12 uptake measured in 3T3L1 cells on the 7th day of differentiation (**B**); control cells were not exposed to any compound; values are means ± standard deviations from at least three independent experiments, *n* ≥ 9; statistical significance was calculated versus control cells \* *p* ≤ 0.05, \*\*\* *p* ≤ 0.001. Cellular uptake of FFA-C12 analogue visualized under a fluorescent microscope (400 × magnification) (**C**).

Simultaneously, the influence of FJ and PJ on cellular free fatty acid (FFA) uptake was determined. As shown in Figure 5B, the level of fluorescent free fatty acid analogue TF2-C12 incorporated in the presence of PJ was decreased by almost 10%. Cells treated with PJ lacked strong fluorescence of

lipid bodies, with the fluorescent analogue present mainly in the cytoplasm (Figure 5C). Collectively, it can be concluded that the purified juice of *V. opulus* (PJ) inhibited the adipogenic differentiation of 3T3-L1 cells.

Besides inhibiting lipogenesis, it is possible that *V. opulus* could influence the lipolysis process, which leads to the stored TAG breakdown to fatty acids and glycerol [34]. Cell incubation with preparations increased the amount of glycerol released from adipocytes, with maximal stimulation by 20% observed for 25 μg/mL of PJ (Figure 6), while for the FJ preparation used in 100 μg/mL concentration this increase was only 7%. The present study provides evidence that *V. opulus* components may inhibit lipogenesis and stimulate adipolysis. The decrease in lipolysis observed in the presence of elevated dosages of PJ (50 μg/mL) resulted from its cytotoxic potential. Taking into account the obtained results, it can be concluded that the *V. opulus* influence is dose dependent according to the concentration of phenolic compounds. Regardless of the 90-fold higher concentration of phenolic compounds in PJ, its IC0 dose against 3T3-L1 cells is only 4-times lower than FJ. The most responsible for observed activity seems to be chlorogenic acid, which was identified as the main phenolic compound in probes. However, due to potential synergic activities and chemical interactions, other compounds present in juice, but lost during its purification process, such as procyanidins and proteins, are also responsible for the observed cellular effect. In this regard, the presented results are in agreement with our previous studies, where phenolic rich fraction from *V. opulus* fruit had stronger activity in Caco-2 and MIN6 cells [18,21].

**Figure 6.** Lipolytic activity of *V. opulus* FJ and PJ preparations on differentiated 3T3L1; as a positive control 10 μM isoproterenol was used; control cells were not exposed to any compound; values are means ± standard deviations, *n* = 6; statistical significance was calculated versus control cells \* *p* ≤ 0.05, \*\*\* *p* ≤ 0.001.

#### *3.4. The E*ff*ects of V. opulus on Expression of Genes Associated with Adipocyte Di*ff*erentiation*

Differentiation of preadipocyte to adipocyte is regulated by set of transcription factors, which, upon activation, induce the expression of other adipocyte-specific genes [35]. In order to clarify the molecular effects of *V. opulus* components, the analysis of selected genes' expressions, as well as protein levels, was performed. Given the fact that *V. opulus* diminished lipid accumulation in 3T3-L1 cells, we first checked the effects of FJ and PJ used at the IC0 concentration on the mRNA expression of master adipogenic regulators: peroxisome proliferator-activated receptor gamma (PPARγ), CCAAT/enhancer-binding proteins (C/EBPα) and sterol regulatory element-binding protein 1c (SREBP-1c), which have a direct impact on fat cells' development [35]. As shown in Figure 7, both samples suppressed the expression of these factors compared to untreated cells. In regard to previous studies, the purified juice rich in phenolic compounds presented a higher impact on 3T3-L1 cells than the fresh juice. Following PJ treatment, the PPARγ mRNA level decreased to 50%, while C/EBPα, CBP and SREBP-1c mRNA levels declined by 30–40%. Among the genes studied, fresh juice had no impact on the changes of SREBP-1c mRNA expression, whereas a decrease in other genes' levels did not exceed 30%. Furthermore, we investigated the effect of tested samples on these transcription factors' protein level. Immunoblot analysis also revealed that the PJ reduced the amount of PPARγ protein in adipocyte to 40% as compared to the control cells (Figure 8A). It is known that the activation of PPARγ protein by specific agonists results in its translocation to the nucleus, where heterodimerizes with the retinoid x receptor alpha (RxRα) [35]. Thus, we decided to study a subcellular distribution of PPARγ receptor. The level of protein detected in the nuclear fraction of adipocytes with ELISA technique showed that, in this regard, PJ diminished PPARγ distribution inside the nucleus to 60% (Figure 8B). As was mentioned above, PPARγ transcriptional activation is initiated after its binding with the RxRα receptor in the nucleus. That step of heterodimer formation describes both proteins as ligand-activated transcription factors, which coordinately regulate the gene expression of other crucial proteins involved in fatty cells differentiation, such as adipogenesis, lipid storage, lipogenesis and thermogenesis [5,6]. The results in Figure 8 show that FJ induced nearly 3-fold increase in RxRα protein. This may be caused by the presence of retinoids and 9-cis-retinoic acid in studied preparations. These compounds are known precursors of carotenoids and were found to increase RxRα expression and activity [17,36–38]. The solid-phase purification of the FJ eliminated these compounds; thus, cells' treatment with PJ showed a 20% decrease in RxRα protein. Therefore, one can conclude that the observed limitation of PPARγ-RxRα heterodimer formation in the cells incubated with *V. opulus* may also partially reduce adipogenesis.

**Figure 7.** The influence of *V. opulus* FJ and PJ preparations on the gene expression in differentiated 3T3-L1 cells quantified by real-time PCR and normalized using β-actin as a reference gene. Control cells were not exposed to any compound; values are means ± standard deviations, *n* ≥ 4; statistical significance was calculated versus control cells (untreated), \* *p* ≤ 0.05, \*\* *p* ≤ 0.01, \*\*\* *p* ≤ 0.001.

**Figure 8.** Effect of *V. opulus* FJ and PJ preparations on the relative protein level of crucial proteins involved in adipogenesis and lipogenesis determined by Western blot analysis in differentiated 3T3-L1 cells (**A**); the level of PPARγ determined with ELISA assay in nuclear fraction (**B**); control cells were not exposed to any compound; the values are means ± standard deviations, *n* ≥ 5; statistical significance was calculated between treatment and control cells (untreated), \* *p* ≤ 0.05, \*\* *p* ≤ 0.01, \*\*\* *p* ≤ 0.001.

Regardless of this, it was also demonstrated that CREB-binding protein (CBP) after its activation binds to the promoter of the C/EBP gene being transcriptional coactivator that associates with PPARγ. It has been demonstrated that IBMX and cyclic-AMP (cAMP) analogs promote adipocyte differentiation via CREB phosphorylation [39]. As before, *V. opulus* downregulated CBP on the transcriptional and translational levels (Figures 8 and 9); however potential CREB phosphorylation via the cAMP-dependent pathway needs to be further elucidated.

Next, we investigated the effect of *V. opulus* FJ and PJ on the expression of genes that are up-regulated during adipocyte differentiation and controlled by the abovementioned transcription factors. Fatty acid synthase (FAS) is involved in the *de novo* synthesis of long-chain fatty acids; thus, FAS inhibition has been considered as one of the major factors decreasing the amount of intracellular fatty acids and lipogenesis [40]. While we have not directly checked the *V. opulus* influence on FAS enzymatic activity, the significant downregulation of FAS expression by both preparations was detected (Figure 7). This effect was, again, slightly stronger for the PJ than for the FJ preparation (0.55 and 0.49, respectively). In addition, 3T3-L1 cells' incubation with PJ influenced the mRNA level of the FAT/CD36 gene. The protein encoded by this gene is involved in the transmembrane movement of fatty acids and studies performed on CD36 knockout mice showed decreased fatty acids uptake by adipocytes [41]. According to the data presented in Figure 7, the 3T3-L1 cells incubated with PJ showed that FAT/CD36 expression decreased by 20%, which matches that mechanism with previously observed diminished cellular FFA uptake. Furthermore, the reduction in SREBP-1c protein expression results in the inhibition of the expression of the enzyme catalyzing synthesis of the malonyl-CoA involved in fatty acid and triglyceride synthesis—acetyl-CoA carboxylase (ACC) [42]. The levels of the gene and ACC protein expression were decreased by almost 50% (Figures 7 and 8) and these results are in agreement with the observed decrease in lipid and TAG content in mature 3T3-L1 cells. The results

obtained are in line with the decrease of FFA uptake observed in Caco-2 cells [18]. However, it was also presented that *V. opulus* increased FFA uptake and lipids accumulation in insulinoma MIN6 cells, which may deleteriously effect insulin secretion [19].

**Figure 9.** Effect of *V. opulus* FJ and PJ preparations on the relative levels of phosphorylated proteins involved in adipogenesis determined by Western blot analysis in differentiated 3T3-L1 cells. Control cells were not exposed to any compound; values are means ± standard deviations, *n* ≥ 4; statistical significance was calculated between treatment and control cells (untreated) with \*\*\* *p* ≤ 0.001.

Previous studies showed that, despite the activation of PPARγ, the differentiation of preadipocyte can be modulated by AMP-activated protein kinase (AMPK) involvement [43]. To clarify an influence of *V. opulus* extracts on 3T3-L1 differentiation suppression we studied whether these extracts are able to activate AMPK. One of the key factors of AMPK activation is the elevation of AMP level leading to the protein α-subunit threonine 172 phosphorylation [44]. As is shown in Figure 9, the level of total AMPK was not affected by *V. opulus* components, whereas the level of p-AMPKα was significantly decreased by the PJ preparation.

Among the AMPK substrates involved in the adipogenesis is acetyl-CoA carboxylase (ACC). The AMPK-catalyzed phosphorylation of ACC inhibits its enzymatic activity. While the mRNA and protein levels of ACC were decreased in cells treated with preparations, only the FJ elevated amount of phosphorylated ACC (Figure 9). In cells treated with the purified juice p-ACC level was notably diminished to 20% in comparison with the control cells (Figure 9), which is also in concordance with the observed decrease in p-AMPK level. In this regard it can be concluded that *V. opulus* phenolic components may inhibit activities of AMPK upstream kinases, like tumor-suppressor liver kinase B1 (LKB1), calcium-dependent calcium/calmodulin-dependent protein kinase kinase β (CaMKKβ) and transforming growth factor-β activated protein kinase-1 (TAK1), or activate mechanism independent of AMPK [45].

#### *3.5. The E*ff*ects of V. opulus on Intracellular Reactive Oxygen Species Production and Selected Adipokines and Cytokines Secretion*

During the adipogenesis and adipocyte enlargement, the intracellular reactive oxygen species (ROS) generation by the mitochondria is intensified and contributes to energy metabolism [8]. Thus, we analyzed the influence of FJ and PJ on intracellular ROS formation. As is illustrated in Figure 10A, the *V. opulus* both samples used at the IC0 dosages decreased oxidation status in adipocytes by 10–15% as compared to the control cells. Observations with the fluorescence microscopy of cells stained

with fluorogenic dichloro-dihydro-fluorescein diacetate (DCFH-DA) correspond to the presented quantitative results (Figure 10B). This cytoprotective activity may be related to the antioxidant properties of the components of the studied extracts. The main components identified in *V. opulus* fruit juice are phenolic compounds, which are involved in direct free radical quenching. Our previous studies revealed that the comparable dosages of *V. opulus* extracts were also able to decrease radicals generation and intracellular ROS level in Caco-2 and MIN6 cells [18,21]. Furthermore, *V. opulus* compounds were able to enhance activity of enzymes involved in cellular defense system, i.e., glutathione peroxidase (GPx). Nevertheless, the elevated dosage of PJ (50 μg/mL) increased intracellular oxidative stress as a result of the mitochondrial depolarization induced by sample components [21]. The observed reduction in ROS level for PJ at 100 μg/mL concentration confirms its cytotoxic ability (demonstrated previously in Figure 3), leading to a decrease in cell number and the induction of cellular death.

**Figure 10.** Influence of *V. opulus* FJ and PJ preparations on the intracellular reactive oxygen species generation analyzed by DCFH-DA assay in differentiated 3T3-L1 cells (**A**); control cells were not exposed to any compound; as a positive control for ROS generation, 500 μM t-BOOH was used; values are means ± standard deviations from at least three independent experiments, *n* ≥ 9; statistical significance was calculated versus control cells \* *p* ≤ 0.05, \*\* *p* ≤ 0.01, \*\*\* *p* ≤ 0.001. Cells visualized under phase-contrast and fluorescent microscope (200× magnification) (**B**).

Differentiated adipocytes are able to secrete proteins that are responsible for adipose tissue remodeling, as well as inflammation generation. The most abundant adipokine is adiponectin, the levels of which are decreased in subjects with diet-related diseases, such as obesity or type 2 diabetes. Activated PPARγ, C/EBPα and SREBP1c stimulate its expression in adipocyte [7]. Adiponectin targets adiponectin receptor (AdipoR) mainly regulates energy metabolism and reveals protective insulin-sensitizing and anti-inflammatory properties. The second protein, leptin, plays role in appetite suppression and the downregulation of food intake; however, its serum concentration is increased in obesity due to observed leptin resistance [46]. As it is demonstrated in Figure 11, the incubation of differentiated 3T3-L1 cells with PJ preparation increased the expression of adiponectin gene by 15%, as well as up-regulated its extracellular secretion by 25% compared to the control cells. In the same way, there was a noticeably diminished level of lipid and TAG accumulation in 3T3-L1 cells. Despite the lack of the FJ and PJ influence on leptin mRNA expression level, there was an observed relevant decrease in leptin secretion by cells treated with these preparations (15–20%). Studies performed *in vivo* demonstrated that chlorogenic acid (dominant phenolic compound in FJ and PJ) effectively reduced blood and liver lipid accumulation, as well as decreased amounts of leptin and insulin in plasma [47]. The enlargement of adipocytes induces the release of free fatty acids, which generate oxidative stress, leading to cellular structures damage, but also stimulate the secretion of inflammatory cytokines. Among cytokines, the most related to obesity and insulin resistance are tumor necrosis factor α (TNFα) and interleukin-6 (Il-6). Chlorogenic acid has an anti-inflammatory effect, reducing the cellular release of TNFα, Il-1β and Il-6 cytokine [9,48–50]. Figure 11 shows that only purified juice declined the expression of TNFα mRNA by 20%. TNFα protein level was also downregulated by PJ to 70%. It is known that TNFα regulates the synthesis of pro-inflammatory cytokines, including Il-6 [51]. In concordance, both *V. opulus* preparations decreased the secretion of Il-6 protein to 40–70%. In contrast, Il-6 mRNA level was not changed. Hence, we can conclude that the FJ and PJ components could be further checked with animal models of obesity or insulin resistance as potential preventive agents against chronic inflammatory response.

**Figure 11.** Influence of *V. opulus* FJ and PJ preparations on the mRNA expression (**A**) and protein secretion (**B**) of selected adipokines and cytokines in differentiated 3T3-L1 cells; control cells were not exposed to any compound; values are means ± standard deviations, *n* = 6; statistical significance was calculated versus control cells \* *p* ≤ 0.05, \*\* *p* ≤ 0.01, \*\*\* *p* ≤ 0.001.

To the best of our knowledge, there are no reports on *V. opulus* fruit influence on lipid metabolism regulation with PPARγ involvement. However, it is known that chlorogenic acid, the main phenolic constituent of *V. opulus*, was able to suppress mRNA level of Pparγ, Cd36 and Fabp4 in liver and white adipose tissue in mice treated with a high-fat diet [49]. The results obtained by Villalpando-Arteaga et al. are comparable and show that aqueous extract from *Hibiscus sabdari*ff*a* containing chlorogenic acid, delphinidin-3-sambubioside and cyanidin-3-sambubioside, attenuated steatosis progression and Pparγ expression in the liver of obese mice [52]. Chlorogenic acid and rutin (quercetin-3-rutinoside) were observed as effective inhibitors of the accumulation of intracellular triglyceride content in 3T3-L1 cells [53]. They were able to down-regulate the expression of adipogenic transcription factors (PPARγ and C/EBP) and adipocyte-specific proteins (leptin), as well as up-regulate adiponectin. In comparison, other study performed with 3T3-L1 differentiated cells identified chlorogenic acid as an agonist of PPARγ2, which promoted adipocyte differentiation via the elevation of PPARγ2, CEBP and SREBP-1 mRNA and protein level [54]. Nevertheless, the other phenolic compounds identified in *V. opulus* fruit were proved to decrease the adipogenesis process [55–57]. Cyanidin-3-glucoside was able to elevate adiponectin gene expression in human adipocytes, whereas, in C57Bl/6J mice diminished the expression of Fas and Srebp-1 and decreased body weight and hepatic lipid content [55,57]. Reduced levels of SREBP-1c, ACC and FAS were noted in 3T3-L1 cells after treatment with quercetin derivatives and rutin [58,59]. Additionally, rutin and catechin were capable of suppressing adipocyte differentiation via PPARγ and RxRα receptor down-regulation [59–61]. Catechin effectively suppressed

3T3-L1 cells differentiation with the inhibition of the expression and protein levels of PPARγ and FAS; however, the observed results were stronger after the cells' treatment with a combination of catechin and caffeine [62]. Procyanidin B2 and other derivatives present in FJ and PJ preparations were also confirmed as regulators of adipocyte triglyceride content with PPARγ involvement [63,64].

#### *3.6. The E*ff*ects of V. opulus on Activity of PPAR*γ

Among different transcriptional activators of adipogenesis, PPARγ, which belongs to nuclear receptors family, is considered as a master regulator of adipocytes differentiation. It interplays with other transcription factors and binds numerous proteins involved in the regulation of transcription, such as PPARγ cofactor 1α (PGC-1α) [8,65]. PPARγ activity can be regulated by ligand binding to the ligand-binding domain, inducing protein conformational changes. Among natural precursors of ligands for PPARγ are long-chained fatty acids, as well as nitriled or oxidized lipids [66]. Thus, to detect if *V. opulus* phenolic components directly affect the activity of PPARγ protein, we used the cell-based reporter gene assay. As a PPARγ activator, we used rosiglitazone at 1 μM concentration, while compound T0070967 (1μM) was used as an antagonist. As it is shown in Figure 12, after the cells' incubation with agonist PPARγ, activity was elevated almost twofold compared to the control cells, whereas the antagonist decreased the receptor activity by 40%. PPARγ activity was diminished by 25% after treatment with PJ, whereas FJ had no effect. In addition, the cells' preliminary incubation with tested preparations significantly decreased receptor activity after rosiglitazone treatment by 90% and 40% for PJ and FJ, respectively. Based on this, we can suspect that *V. opulus* juice components possess antagonist potential against PPARγ receptor.

**Figure 12.** The effect of *V. opulus* FJ (25 and 100 μg/mL) and PJ (25 μg/mL) preparations on PPARγ activation in GeneBLAzer®PPAR gamma 293H DA reporter gene assay; as agonist 1 μM rosiglitazone was used, whereas as antagonist 1 μM T0070967; the control cells were not exposed to any compound; the values are means ± standard deviations, *n* = 4; the statistical significance was calculated versus control cells with \* *p* ≤ 0.05, \*\* *p* ≤ 0.01, \*\*\* *p* ≤ 0.001; the statistical significance was calculated versus cells after rosiglitazone treatment with ## *<sup>p</sup>* <sup>≤</sup> 0.01, ### *<sup>p</sup>* <sup>≤</sup> 0.001.

PPARγ protein belongs to the superfamily of nuclear receptors and possesses six domains, among which the ligand-binding domain (LBD), via interactions with ligands, is involved in the modulation of PPARγ activity. Afterwards, the ligand binding induces a change in the receptor conformation dynamic process of corepressor dissociation, and coactivator recruitment is started [14]. To explore the potential interaction between *V. opulus* juice components and PPARγ protein, molecular docking simulation was performed with the binding pocket located behind the H3 helix of PPARγ receptor. There are many PPARγ receptor agonists that locate in this binding pocket, i.e. indol-1-yl acetic acid, nonaoic acid or amorfrutin 1 [67–69]. As an example of the PPARγ agonist in the analyzed model, we used rosiglitazone (Figure 13), which locates deeply inside the receptor-binding pocket. The simple validation of the docking method showed slight differences between rosiglitazone in crystal structure and docked molecule in the pocket; however, this effect was acceptable for performed research with the main phenolic compounds identified in *V. opulus* juice. Molecular modeling with Autodock Vina located chlorogenic acid and (+)- catechin inside the binding pocket. Among the studied compounds, chlorogenic acid, (+)- catechin and rosiglitazone showed similar binding affinities (Figure 13). Additionally, the chlorogenic acid molecule orientation in pocket also revealed some similarity to that of rosiglitazone. Despite this, neither chlorogenic acid nor (+)- catechin did not create any possible hydrogen bond to serine 289 or tyrosine 473 of PPARγ, which were essentially made by the rosiglitazone molecule. This gives some hint to predict that these compounds have some potential to activate PPARγ, but their mechanism of action could be different than rosiglitazone; they could bind to a different part of the receptor and change its conformation in a similar way to the partial agonist. In contrast to these two phenolic ligands, procyanidins B1 and C1 cannot fit into the binding pocket because of their more complex structures and larger shape. In this case, estimated high binding affinity resulted from numerous hydrogen bonds created between procyanidins and the residues present in surface of receptor. Thus, procyanidins present in *V. opulus* juice could clog the entrance to the PPARγ binding pocket and block it from a potential agonist entering, which could resemble inverse agonist behavior. As a result, the reduction of PPARγ activity may occur despite the presence of chlorogenic acid, which was demonstrated to be a PPARγ agonist. Among the limitations of performed basic molecular docking, there is a lack of prediction of the most favorable energetic conformation for research space, as well as kinetic data describing designed configuration. Thus, further studies with kinetic modeling or isothermal titration calorimetry would give more detailed information in this regard. Still, the results obtained with molecular docking are in agreement with the biological experiments showing a reduction of signal transduction controlled by PPARγ, and, finally, a decrease in 3T3-L1 cells adipogenesis. Therefore, PPARγ transcriptional potential could be modulated on different levels: by the regulation of mRNA and protein levels, as well as conformational changes made after its binding with ligands leading to protein subcellular distribution, heterodimers' formation, and, finally, binding with peroxisome proliferator hormone response elements present in promoters of PPAR-responsive genes. The studies performed recently identified chlorogenic acid as an agonist of the PPARγ2 receptor responsible for the enhanced differentiation of 3T3-L1 cells [54]. In the identified molecular mechanism downregulation of adipocyte differentiation-inhibitor gene, Pref1 was observed, which was accompanied by the upregulation of CEBP and SREBP-1 transcriptional factors. Microscopic observations demonstrated the elevation of PPARγ in the nucleus fraction, as well as in total cell lysate. Based on this data and our results, we can conclude that other *V. opulus* juice components, such as proanthocyanins, may be responsible for the observed PPARγ limitation.

To PPARγ endoligands belong unsaturated fatty acids or eicosanoids, whereas, among synthetic ligands, the most known are thiazolidinediones, such as rosiglitazone [5]. A recent study presented chemically pure phenolic compounds' abilities to regulate the adipogenesis process and matched the obtained results with the molecular docking analysis of their binding affinity to PPARγ LBD [70]. As it was shown, the obtained in silico binding affinities of quercetin, apigenin, resveratrol, ellagic acid and coumaric acid to PPARγ were in agreement with these compounds' inhibition potential of 3T3-L1 cells differentiation. However, it can still be supposed that the nuclear receptor ligand activity might be limited not only by its bioavailability, but also by its intracellular uptake.

**Figure 13.** Interaction of tested phenolic compounds with PPARγ receptor verified with molecular docking. The active site of the PPARγ receptor is located behind the E3 helix, where a small cavity is formed. An agonist of PPARy, rosiglitazone, enters the receptor-active site and locates deep inside the binding pocket. Molecular models of rosiglitazone, (+)- catechin, chlorogenic acid, procyanidin B1 and procyanidin C1 are shown on the entrance of the active pocket site. Orientation of models are presented in comparison to the crystal model of rosiglitazone. Interactions of the models with corresponding binding affinities are presented in the two last columns. Molecular docking was performed with AutoDock Vina.

#### **4. Conclusions**

The present work provides direct evidence of the *V. opulus* juice effect on the adipogenesis of 3T3-L1 cells (Figure 14). Phenolic compounds identified in juice, mainly chlorogenic acid, procyanidins and catechins, were found to suppress adipogenesis by the downregulation of major regulators of adipogenesis, such as PPARγ, C/EBPα and SREBP-1c. The regulation of PPARγ-mediated β-lactamase expression in reporter gene assay, as well molecular docking, suggested that *V. opulus* components may work as a PPARγ antagonist. As result, the levels of enzymes involved in lipid metabolism, such as FAS or ACC, were decreased, along with adipokine TNFα, Il-6 and leptin. Additionally, *V. opulus* juice was able to inhibit pancreatic lipase, which potentially could reduce the uptake of fatty acids in the digestive tract and the storage of body fat in the adipose tissue.

**Figure 14.** *V. opulus* juice phenolic compounds as modulators of lipids metabolism—the proposed mechanism of action. *V. opulus* downregulates the adipogenesis of 3T3-L1 cells, decreases the expression and activity of the PPARγ nuclear receptor as well as PPARγ-regulated proteins (i.e., fatty acid synthase (FAS), acetyl-CoA carboxylase (ACC)), diminishes the release of inflammatory cytokines and leptin, possesses cytoprotective activity against ROS generation and increases adiponectin secretion. *V. opulus* inhibits pancreatic-lipase and limits fatty acid release from the food matrix.

Our results contribute to elucidate *V. opulus* phenolic compounds' molecular mechanism in adipogenesis regulation. However, observed adipogenesis inhibitory outcome needs further molecular evaluation after *V. opulus* juice *in vitro* digestion and its incubation with gut microflora. These processes may greatly influence the composition of the studied phytocompounds, as well as the activity. Taking into account possible *V. opulus* cellular type-dependent *in vitro* activity, its potential usage as a diet component needs further *in vivo* studies performed with an animal obesity model to show its efficacy in the regulation of lipid metabolism at safety doses.

**Author Contributions:** M.Z.-S. conceptualization, supervision, methodology, lab work with cell cultures and results, writing—original draft preparation, review, and editing; N.P. lab work and results, obtaining and characterization of *V. opulus* phenolics; M.S. molecular modeling; A.P. involved in study conceptualization, lipase inhibition studies, *V. opulus* phenolics characterization; funding acquisition. All authors have read and agreed to the published version of the manuscript.

**Funding:** The research was supported by grant number 2016/23/B/NZ9/03629 from The National Science Centre, Poland.

**Acknowledgments:** Małgorzata Zakłos-Szyda gratefully acknowledges Maria Koziołkiewicz for her encouragement for opening new research possibilities and critical comments.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*

### **Acute Epigallocatechin-3-Gallate Supplementation Alters Postprandial Lipids after a Fast-Food Meal in Healthy Young Women: A Randomized, Double-Blind, Placebo-Controlled Crossover Study**

### **Alcides C. de Morais Junior 1, Raquel M. Schincaglia 1, Marisa Passarelli 2,3, Gustavo D. Pimentel <sup>1</sup> and João F. Mota 1,\***


Received: 21 July 2020; Accepted: 18 August 2020; Published: 21 August 2020

**Abstract:** A high-fat fast-food meal negatively impacts postprandial metabolism even in healthy young people. In experimental studies, epigallocatechin-3-gallate (EGCG), a bioactive compound present in green tea, has been described as a potent natural inhibitor of fatty acid synthase. Thus, we sought to evaluate the effects of acute EGCG supplementation on postprandial lipid profile, glucose, and insulin levels following a high-fat fast-food meal. Fourteen healthy young women 21 <sup>±</sup> 1 years and body mass index 21.4 <sup>±</sup> 0.41 kg/m<sup>2</sup> were enrolled in a randomized, double-blind, placebo-controlled crossover study. Participants ingested capsules containing 800 mg EGCG or placebo immediately before a typical fast-food meal rich in saturated fatty acids. Blood samples were collected at baseline and then at 90 and 120 min after the meal. The EGCG treatment attenuated postprandial triglycerides (*p* = 0.029) and decreased high-density lipoprotein cholesterol (HDL-c) (*p* = 0.016) at 120 min. No treatment × time interaction was found for total cholesterol, low-density lipoprotein (LDL-c), and glucose or insulin levels. The incremental area under the curve (iAUC) for glucose was decreased by EGCG treatment (*p* < 0.05). No difference was observed in the iAUC for triglycerides and HDL-c. In healthy young women, acute EGCG supplementation attenuated postprandial triglycerides and glucose but negatively impacted HDL-c following a fast-food meal.

**Keywords:** green tea; epigallocatechin; lipid profile; high-fat diet; fast food

#### **1. Introduction**

In general, fast foods are rich in refined carbohydrates, sodium, and saturated and trans-fatty acids, and they are poor sources of vitamins, minerals, and dietary fibers [1]. The frequent consumption of fast food has been associated with overweight and obesity, cardiovascular disease, type 2 diabetes, and other metabolic disorders [2]. The acute consumption of a high-fat, energy-dense, fast-food meal promotes postprandial impairments of the lipid profile and induces oxidative stress in subjects with metabolic syndrome [3]. These acute metabolic damages are not only found in individuals with chronic diseases. In a crossover study, different types of fast-food meals negatively impacted postprandial lipids, insulin, and flow-mediated endothelium-dependent dilatation in healthy volunteers [4].

Plant polyphenols have antioxidant and anti-inflammatory properties and may be responsible for numerous beneficial effects on human health [5]. Epigallocatechin-3-gallate (EGCG) is the main

catechin present in green tea, and it is among the best-studied polyphenols [6]. It has been used in the prevention and treatment of non-communicable chronic diseases [7,8]. Treatment with EGCG for 12 weeks decreased bodyweight, total cholesterol, and low-density lipoprotein (LDL-c) in women with central obesity [8], while acute EGCG supplementation was able to delay gastric emptying and increase adiponectin levels in healthy women [9]. In addition, EGCG was observed to decrease lipid absorption from the gastrointestinal tract, which was associated with improvements in insulin resistance and liver triglyceride concentrations [10]. It is important to mention that a cup of green tea (100 mL), one of the main sources for EGCG, contains approximately 144–150 mg of EGCG [11]; nevertheless, EGCG supplements containing different amounts of this catechin are commercialized and easily found for purchase.

We hypothesized that the negative impact on the lipid profile promoted by fast-food meals, rich in saturated fatty acids, could be improved by the acute administration of EGCG. This study sought to evaluate the effects of acute EGCG supplementation on the postprandial lipid profile, glucose, and insulin levels following a fast-food meal among young healthy women. Even though chronic administration of EGCG can impact lipid profile and other metabolic markers [12], we were not aware of studies evaluating the EGCG acute impact on lipid profile after a high-fat fast-food meal in young and healthy individuals. Therefore, we believe this study may help in increasing knowledge regarding EGCG's possible effects on the human lipid profile.

#### **2. Materials and Methods**

#### *2.1. Study Design and Participants*

This was an acute, randomized, double-blind, crossover, placebo-controlled trial with healthy young women, aged 18 to 25 years old. The recruitment of the participants was carried out through social media and fliers placed on notice boards at the university campus. Exclusion criteria included smoking, diagnosis of underweight or overweight as assessed by body mass index (BMI), hypertension, diabetes mellitus, dyslipidemia, heart disease, peripheral vascular disease, a history of liver or kidney disease, the use of anti-inflammatory medications and corticosteroids, dietary restrictions related to the foods used in the study (french fries, bacon, and parmesan cheese), and pregnancy and/or breastfeeding. This study was conducted according to the Declaration of Helsinki and was approved by the Ethics Committee of the Federal University of Goiás (# 2.011.261). Written informed consent was obtained from all participants. This study was registered at http://www.ensaiosclinicos.gov.br/ as RBR-2b8p4n. All participants underwent a baseline medical history screening to determine that they were healthy.

#### *2.2. Experimental Protocol*

A fast-food meal was provided to the participants after a 12 h overnight fast. All participants were instructed to abstain from fresh fruit and vegetables, alcohol, herbal teas, fruit juices, and physical exercise for at least 24 h before the trials, but otherwise were to maintain their regular diet before the two visits, with one week of washout between the sessions [6,13]. Participants were reminded 36 h before each visit about the instructions and were questioned about compliance upon arrival at the laboratory. Right before the high-fat fast-food meal, participants ingested 800 mg, split into two capsules of 400 mg each, of either EGCG or corn starch (placebo). The capsules for the placebo and the active product were similar in appearance, smell, and taste. As presented in a review of the literature performed by our team, studies have used a very wide range regarding amounts of EGCG used in clinical trials [12]. The decision to use a high EGCG dose was based on the positive effects previously found by our research group [9] and the fact that the chronic use of this amount has not promoted harmful side effects [14]. Blood samples were collected at baseline and then 90 and 120 min after the end of the fast-food meal (Figure 1). The times at 90 and 120 min for blood collection were chosen

taking into consideration the fact that green tea polyphenols reach peak serum concentrations in the range of 1.3 to 1.6 h [6].

**Figure 1.** Study design.

The fast-food meal consisted of 80 g of McDonald's® french fries, 130 g of bacon cooked for 5 min without the addition of oil, and 60 g of grated Parmesan cheese. The meal contained 20.57 g of carbohydrate, 47.41 g of protein, and 83.90 g of fat, of which 37.17 g was saturated fat, adding to a total of 1027 kcal. This meal was chosen because of its popularity in pubs and fast-food restaurants in Brazil and other countries. Habitual dietary intake was gathered from all involved in the study using a 24 h recall including three non-consecutive days: two weekdays and one weekend day.

#### *2.3. Anthropometric and Body Composition Assessments*

Body weight, height, and waist circumference were measured according to the procedures described by Lohman (1988) [15], using a digital anthropometric scale and a stadiometer (Filizola®, São Paulo, Brazil). All of the measurements were done in duplicate, and the mean between the two findings was used. These values were used to calculate the body mass index (BMI) (kg/m2). The estimations of their lean mass values and body fat percentage were assessed by dual-energy X-ray absorptiometry (DXA, Lunar DPX NT, GE Healthcare®) with the enCORE 2011 software (version 13.60).

#### *2.4. Analyses of Samples*

Blood samples were collected from the antecubital vein and immediately centrifuged at 4000 rpm for 10 min at 4 ◦C in a refrigerated centrifuge (Hitachi CF16RN, Hitachinaka, Ibaraki, Japan) to obtain the serum. The serum was immediately frozen and stored at −80 ◦C until analysis. Glucose concentrations were determined by the enzymatic colorimetric method. Total cholesterol (TC), high-density lipoprotein cholesterol (HDL-c), triglycerides (TG), and insulin were determined by immunoturbidimetry methods (Architect Plus®, Naperville, IL, USA). The concentration of low-density lipoprotein (LDL-c) was calculated using Friedewald's equation, and VLDL-c dividing TG by five [16].

#### *2.5. Statistical Analyses*

Sample size calculation was performed using G\*Power software version 3.1.9.2, taking into consideration the effect on TG [17]. With an effect size of 1.81, a level of significance of 5%, and a statistical power of 95%, the minimum required sample was seven individuals in each treatment. Software R and RStudio were used for the statistical analyses. Values are presented as mean and standard error. The analyses were conducted considering the variations at the timepoints of 90 and 120 min after supplementation and the ingestion of the fast-food meal concerning the baseline measures of the study. The analysis of normality of the residues was performed by the Lilliefors test for the study variables. The carryover effect was analyzed as explained by Rosner (2011) [18], with no significant

values. Effect sizes (ES) were calculated using Cohen's formula and classified as small (d = 0.2), medium (d = 0.5), or large (d = 0.8). Mean tests were performed by analysis of variance (factorial ANOVA) and the differentiation test of Tukey. Postprandial responses between treatments were also compared with the incremental area under the curve (iAUC) using the trapezoidal method [19]. The significance level of 5% was adopted for all tests.

#### **3. Results**

Twenty-seven participants were assessed for eligibility; however, only 25 were randomized into the treatments due to the use of anti-inflammatory drugs or BMI. During the study, nine participants dropped out due to personal reasons (*n* = 7), blood collecting difficulties (*n* = 1) and sickness (*n* = 1), and two participants were excluded from the study due to hypertriglyceridemia (Figure 2). The mean age and BMI were 21 <sup>±</sup> 1 years of age and 21.4 <sup>±</sup> 0.41 kg/m2, respectively. The baseline characteristics of the participants are described in Table 1.

**Figure 2.** Flow diagram.


**Table 1.** Baseline characteristics of participants.

HOMA-IR: homeostatic model assessment for insulin resistance.

A significant treatment × time interaction was found for TG, VLDL-c, and HDL-c (*p* < 0.05, Table 2). The EGCG group attenuated the increase in TG compared to the placebo at 120 min (+36.1 ± 6.25 vs. +45.9 ± 6.46 mg/dL, respectively, *p* = 0.029, ES = 0.33). VLDL-c concentrations followed the same pattern. Participants from the EGCG group also had greater reductions of HDL-c concentrations (EGCG: −6.5 ± 0.72 vs. placebo −.50 ± 0.44 mg/dL, *p* = 0.016, ES = 0.80) compared to the placebo group at 120 min (Table 2). There was no significant difference in LDL-c. There was a marginal significance in attenuation of postprandial glucose at 90 min (*p* = 0.061, ES = 0.71) and 120 min (*p* = 0.094, ES = 0.24) in the EGCG group compared to the control. The iAUC for glucose (*p* = 0.047) was lower in the EGCG group compared to the placebo (Table 2). No unintended effects were observed throughout the experiment that were related to the ingestion of the fast-food meal or the EGCG supplementation.


**Table 2.** The response of the epigallocatechin-3-gallate supplementation after a high-fat fast-food meal.

Values expressed as means and standard error of means ± SEM. EGCG: epigallocatechin-3-gallate. iAUC: the incremental area under the curve. † *p* = 0.05 to < 0.10 (marginal significance) and \* *p* < 0.05 between treatments analyzed by factorial ANOVA test.

#### **4. Discussion**

To our knowledge, this is the first trial that evaluated the acute impact of EGCG on lipid profile after a high-fat fast-food meal in young and healthy individuals. In the present study, acute administration of EGCG attenuated postprandial TG; however, it also promoted a larger decrease in HDL-c after the high-fat fast-food meal. Considering that this type of food is frequently consumed by young people [20] and that exaggerated postprandial hypertriglyceridemia is a risk factor for cardiovascular disease [21,22], the results of this study might have important applicability in clinical practice.

Corroborating with our results, a similar finding was observed when male adults with borderline and mild hypertriglyceridemia (range 122.12–220.35 mg/dL) submitted to acute ingestion of different amounts of catechin extract from green tea (moderate dose = 224 mg and high dose = 674 mg) [23]. After a light meal, the postprandial TG curves decreased by 15.1% and 28.7%, respectively, compared to the control group [23]. Here we did not find an effect on the iAUC for triglycerides, which may be related to the fact that our study was dealing with younger and healthier individuals. This suggests that EGCG administration can be especially important for populations at high risk of coronary heart disease, considering that postprandial lipidemia is an independent risk factor for cardiovascular diseases [24] and might be predictive of an elevated risk of myocardial infarction [25].

Chronic interventions using EGCG have also shown postprandial TG decreases in mice fed a high-fat diet [26] and in obese patients [27,28]. In a randomized, double-blind, placebo-controlled study, the consumption of a green tea extract containing 208 mg of EGCG associated with minerals decreased BMI, waist circumference, TC, and LDL-c after three months in obese patients [27]. These results for the lipid profile corroborate findings obtained in our previous study with overweight women taking

1 g/day of green tea extract (560 mg polyphenols, ~224 mg of EGCG) for three months [28]. In the present study, TC and LDL-c were not acutely affected by the one-time EGCG intervention. It has been reported that a postprandial increase in TG was associated with a decrease in LDL-c among men newly diagnosed with metabolic syndrome [29], which was also observed in the current study. Thus, the effects of EGCG on LDL-c seem to occur in a chronic rather than an acute manner.

Huang et al., 2018 [30] suggested that the impact of EGCG on lipid absorption is due to a decrease in bile acid reabsorption, which is necessary for the digestion and metabolism of lipids [31]. A decrease in postprandial HDL-c may also occur when TG is reduced, and its plasma concentration should drop rapidly, considering that HDL-c concentration is dependent on the metabolism of TG-rich lipoproteins [32]. If less fat is absorbed after EGCG administration, this polyphenol may be an important tool to decrease the postprandial deleterious effects of high-fat meals.

In an experimental study conducted with male C57BL/6J mice, the benefits of EGCG for alleviating insulin resistance and liver TG concentration have been attributed to decreased lipid absorption and reduced inflammatory cytokine concentrations [10]. In our study, the iAUC for glucose decreased in the EGCG group, which was not observed for postprandial insulin concentrations. This might be due to the acute design of the study or because the participants were healthy. It has been reported that individuals with poor fasting blood glucose control show a greater decrease in glucose and insulin concentrations when submitted to dietary interventions [33]. Thus, we might speculate that the intervention attenuates negative effects of a high-fat meal, which should be confirmed in future clinical trials. In a counter-balanced crossover design with 12 healthy men, the administration of green tea extract (containing ~366 mg of EGCG) 24 h before the oral glucose tolerance test increased insulin sensitivity by 13% [34]. In addition, the consumption of 1.5 g of green tea diluted in 150 mL of hot water decreased glucose levels at 30 and 120 min after the oral glucose tolerance test in healthy humans compared to the control group [35]. The authors did not measure the EGCG amount; however, based on previous studies [36,37], it is expected to be around 663 mg of EGCG. Hence, additional studies should be conducted with different populations and EGCG doses. The results obtained in the current study may contribute to the investigation of potential acute benefits for the use of EGCG.

Concerning the limitations of this study, the lack of biomarkers of oxidative stress, measurement of lipid absorption, and postprandial lipemia up to 4 h after the meal should be mentioned. Nevertheless, Kriketos et al., 2003 [17] and Unno et al., 2005 [23] showed the largest increase in postprandial TG occurring 2 h after a high-fat meal, and no increase was found after 3 h. It is expected that the impact of EGCG may continue in the following hours since this catechin would remain at fairly high levels in the bloodstream for 4 h after consumption of 800 mg of EGCG [14]. Besides, we only tested one type of meal, and the effects of EGCG after low-fat or standard meals may differ considerably.

#### **5. Conclusions**

We found in healthy young women that EGCG attenuated postprandial TG and glucose concentrations but negatively impacted HDL-c following a high-fat fast-food meal. Our results reinforce that even when administered acutely, EGCG may have an impact on the lipid profile. However, further studies are required to elucidate the effects of EGCG after fast-food meals on cardiovascular risk markers.

**Author Contributions:** The authors' responsibilities were as follows—A.C.d.M.J., R.M.S., M.P., and J.F.M.: designed the research; R.M.S. and A.C.d.M.J.: performed the statistical analyses; A.C.d.M.J., R.M.S., G.D.P., and J.F.M.: analyzed and interpreted the data; A.C.d.M.J.: wrote the first draft of the manuscript and had primary responsibility for the final content; J.F.M., M.P., and G.D.P.: reviewed the manuscript, contributed to the discussion, and provided essential reagents or essential materials; and all authors were involved in editing the manuscript. All authors have read and agreed to the published version of the manuscript.

**Funding:** A.C.d.M.J. and R.M.S. received scholarships from the Coordination for the Improvement of Higher Education Personnel (CAPES). This research received no external funding. J.F.M. has received support from the National Council for Scientific and Technological Development (CNPq, no. 305082/2019-1).

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*
