Chemical profile and beneficial effect of standardized extract of Stevia rebaudiana Bertoni leaves on metabolic syndrome in high fat diet streptozotocin-induced diabetic rats – Applied Biological Chemistry

Metabolic Syndrome (MS) is a constellation of risk factors that increases a person’s risk of developing cardiovascular disease. These factors include abdominal obesity, atherogenic dyslipidemia (elevated triglycerides, small LDL particles, low HDL cholesterol), raised blood pressure, insulin resistance (with or without glucose intolerance), pro inflammatory state and prothrombotic state [1, 2].It is reported that MS is a major public and clinical problem worldwide; over a billion people in the world are now affected with MS and the prevalence of obesity and its consequent metabolic abnormalities was found to be 32.2% [2]. MS usually starts with insulin resistance leading finally to metabolic disturbances such as hyperglycemia, hyperlipidemia, kidney, and liver impairments. Clinically, treatment of each is prescribed according to the patient’s state. However, acetylcholinesterase inhibitors used for hypertension such as enalapril and captopril, may increase serum creatinine level, or cause cough, headache, and skin rash [3]. Also, metformin, a drug used for type 2 diabetes, can induce gastrointestinal symptoms and lactic acidosis [4].Therefore, traditionally used herbal drugs such as Camellia sinensis, Hibiscus sabdariffa, Citrus limon, and Punica granatum can be considered as a complementary or alternative medicine for metabolic diseases [5].

Stevia rebaudiana Bertoni (Fam. Asteraceae) popularly known as stevia or sweet leaf, is native to Paraguay, where it has a long history of use as a non-calorie sweetener in beverages and foods [6]. Recently, the leaves have gained increased industrial and scientific interests as a perfect alternative to sucrose and artificial sweeteners that have many health hazards [7]. Sweetness of the leaves is imparted by the presence of a complex mixture of zero caloric sweeteners (steviol glycosides, SVGs); mainly stevioside and rebaudioside A (250–300 sweeter than sucrose) [8]. The leaves also contain other important phytochemicals, such as vitamin C and polyphenols (flavonoids and phenolics), which are mostly responsible for the antioxidant activities of its extracts [9]. Accordingly, Stevia has great economical and health values in food industry as non-alcoholic beverages, as food additive. Also, as a natural control for diabetes and to help control weight in obese persons [10]. In number of animal experiments, stevia extracts showed anti-hyperglycemic [10], antioxidant [11], antihypertensive [12], anti-inflammatory [13] and anti-obesity [14] activities. Also, inhibited α-amylase [15] and decreased fasting blood glucose, glycosylated hemoglobin, and improved insulin and glycogen levels in STZ-induced diabetic rats [16]. However, nothing was reported concerning the effect of the extract of stevia cultivated in Egypt or steviosides on MS. Therefore, it deemed of interest to assess the effect of a standardized Stevia extract on hyperglycemia, hyperlipidemia and relevant metabolic parameters associated with MS in a high fat diet (HFD)-streptozotocin (STZ)-induced diabetic rats [17]. This combination of high-fat diet with STZ generates rats with hyperglycemia associated with hypertriglyceridemia and introduces many other metabolic alterations present in human diabetes type 2 (DM2) [17, 18]. Additionally, the antioxidant effect and effect on key digestive enzymes in the hydrolysis of carbohydrates and fats such as pancreatic lipase, α-amylase, and α-glucosidase were investigated [19]. Besides, the chemical profile of the extract was identified by UPLC/MS/MS and major SVGs and flavonoids were isolated and quantified in the leaves of the Egyptian stevia cultivar using HPLC.

Experimental

Materials

Plant material

Samples of the leaves of S. rebaudiana (Bertoni) Egyptian cultivar (through crossing between Chinese and Spanish varieties) were collected from the Sugar Crops Research Institute (SCRI), Agricultural Research Centre (ARC), Giza, Egypt and identified by Dr. Ahmed Attia; Senior Researcher, SCRI, ARC (Breeding and Genetic Department). A voucher specimen (No. 2.9.2019.I) is kept at the Herbarium of the Department of Pharmacognosy, Faculty of Pharmacy, Cairo University. The leaves were collected, air-dried in shade, powdered (with mesh size of 0.2–0.636 mm) and kept in tightly closed glass containers till use.

Chemicals

Porcine pancreatic lipase enzyme, p-nitrophenylbutyrate (PNPB), α-amylase, α-glucosidase, acarbose, p-nitrophenyl-α-D-maltopentoside (PNPM), p-nitrophenyl-α-D-glucopyranoside (PNPG), Streptozotocin (STZ), 2,2-diphenyl-1-picrylhydrazyl (DPPH) and gallic acid were all purchased from (Sigma, St Louis, MO, USA). Metformin and Orlistat were kindly supplied from Eva Pharma, Egypt. Quercetin (Misr Company for Pharmaceutical Industry, Cairo, Egypt). Statin was obtained from EIPICO (Cairo, Egypt). Stevioside and rebaudioside A were isolated in our lab and chemical structures were identified (see Additional file 1).

Solvents

The solvents used in this work viz.; petroleum ether (60–80 °C), n-hexane, methylene chloride, ethyl acetate and methanol were of analytical grade and purchased from the local market. Acetonitrile and methanol used for HPLC and spectrophotometric analyses were from Sigma- Aldrich (Steinheim, Germany).

Extraction, fractionation, and isolation

One kg of stevia leaves powder was extracted with boiling distilled water (4 × 2 L) then filtrated. The filtrates were combined, concentrated, and freeze dried to give 280 g of dry residue (TAqE). Part of the residue (100 g) was suspended in water and repeatedly applied onto a column of Diaion HP-20 (40 cm L × 4 cm i.d.) from (Pharmacia, Fine Chemicals AB, Uppsala, Sweden). Gradient elution started with water (4 L) and decreasing the polarity by 25% increments of methanol till 100% methanol (4 L each) to give 5 fractions. Steviosides rich fraction (SRF, 3.46 g) was obtained from fractions eluted with 25–50% aqueous methanol, and flavonoid-rich fraction (FRF, 5 g) was obtained from fractions eluted with 75%-100% methanol. Stevioside and rebaudioside A were isolated by crystallization from SRF. Also, quercetrin was isolated from FRF. Chemical structures of the isolated compounds were identified using different spectroscopic methods (see Additional file 1).

Total phenolic content (TPC)

The total phenolic content of TAqE was determined using the Folin-Ciocalteu method described by [20], and expressed as μg gallic acid equivalents (GAE) per mg of the extract. All samples were analyzed in triplicate.

Total flavonoid content (TFC)

Total flavonoidal content of TAqE was determined by the aluminium chloride colorimetric assay described by [20], and expressed as μg quercetin equivalents (QE) per mg of the extract. All samples were analyzed in triplicates.

Determination of DPPH radical scavenging activity

The DPPH radical scavenging activity was determined using the method reported by [21]. Absorbance was measured at 492 nm and ascorbic acid concentrations (0.1575–1 mg/ml) were used as standard. A blank was set up in parallel as a control. Each sample was tested twice with triplicate measurements in each experiment. The DPPH radical scavenging activity (%) was calculated as follows:

where, Ac was the absorbance of control (DPPH solution without test sample) and As was the absorbance of sample [DPPH solution + sample (extract/standard)]. The EC₅₀ is defined as the concentration of substrate that causes 50% reduction of the DPPH color. Results are displayed in Fig. 1.

DPPH antioxidant activity of TAqE, fractions and isolated compounds

UPLC-Orbitrap HRESI-MS analysis

The chemical profile of the methanolic extract (5 mg) of S. rebaudiana leaves (Egyptian cultivar) was identified using UPLC coupled to a photodiode array detector (PAD) and an Orbitrap Elite mass spectrophotometer equipped with heated electrospray ionization (ESI) source. Analysis was performed using water (A) and acetonitrile (B) with 0.1% formic acid as mobile phases. The following binary gradient was applied: 0–1 min (isocratic 5% B/A), 1–11 min (linear gradient of B/A from 5 to 100%), 11–19 min (isocratic 100% B) and 19–30 min (isocratic 5% B). The flow rate was 150 μL/min, and the injection volume was 2 μL. The CID mass spectra (buffer gas; helium) were recorded using normalized collision energy (NCE) of 35%. The instrument was equipped with a heated electrospray ion source (negative spray voltage at 3 kV, capillary temperature of 300 °C, source heater temperature of 250 °C, FTMS resolution of 30.000) and RP-18 column (particle size 1.8 μm, pore size 100 Å, 150 × 1 mm i.d., Acquity HSS T3, Waters; column temperature of 40 °C). It was externally calibrated by the Pierce ESI negative ion calibration solution (product No. 88324) from Thermo Fisher Scientific. The data were evaluated using the software Xcalibur 2.2 SP1. Metabolites were also characterized by their UV–VIS spectra (220–600 nm) [22].

HPLC quantitation of steviosides

The chromatographic analysis was performed on Agilent Technologies 1100 series HPLC system Agilent Technologies, Palo Alto, CA), equipped with a quaternary pump, degasser G1322A and UV detector. Agilent Chemstation software was used for data acquisition and processing. Lichrospher RP-C18 column (250 mm L × 4.6 mm ID, 5 µm, Merck, Germany), preceded by a C18 guard column (10 mm L × 4 mm ID, 5 µm) was used. The mobile phase was composed of acetonitrile “solvent A” and 0.3% H₃PO₄ in H₂O “solvent B” applying gradient elution: 20% A/B to 33.7% A/B in 7 min, then to 34% A/B in another 7 min and to 50% A/B in 1 min then to 100% A in 2 min, then to 20% A/B in 3 min. The flow rate was 1 ml/min, injection volume was 20 µL, and detection (UV) was performed at 210 nm.

Sample preparation

Sample (500 mg) of powdered S. rebaudiana leaves (mesh size of 0.2–0.636 mm) was extracted with distilled water (10 × 10 mL) by frequent sonication (for 3 min) and heating on water bath (80ºC for 2 min). The extract was filtered using Whatmann filter paper, and the volume was adjusted to the mark (100 mL) with water. An aliquot (20 µL) of the extract was used for HPLC analysis.

Construction of standard curves for stevioside and rebaudioside A

A standard stock solution of stevioside in water (4 mg/5 mL) was prepared and diluted with water to yield 4 concentrations (25, 64, 96 and 128 µg/mL). An aliquot (20 µL) of each dilution was injected in triplicates and corresponding peak area recorded. The standard calibration curve of stevioside was constructed (r² = 0.998) by plotting mean peak areas versus corresponding concentrations.

Similarly, a stock solution of rebaudioside A in water (2 mg/5 ml) was prepared and diluted to yield 4 concentrations (32, 64,120 and 160 µg/ml). As mentioned above, standard calibration curve of rebaudioside A was constructed (r² = 0.9889).

Pancreatic lipase inhibitory assay

The lipase inhibition activity was determined by a method in [23]. In this method, the enzyme was dissolved in phosphate buffer (pH 6.8) at a concentration of (100 µg/mL) and then centrifuged at 2000 rpm for 5 min to remove insoluble matter. The PNPB solution (substrate) was dissolved in acetonitrile (20 mM) and the stevia extract (TAqE) and fractions (SRF and FRF) were prepared in DMSO at different concentration (1000–7.81 μg ̸mL). The enzyme (20 μL) was incubated first with 20 μL of the sample solution and 20 μL of the phosphate buffer at 30 °C for 5 min in a 96 well plate. Subsequently, 20 μL of PNPB solution was added and the mixture was incubated at 37 °C for 60 min. The absorbance was measured at 405 nm. Orlistat was used at the same concentrations as a standard. Enzyme inhibitory activity was calculated as follows:

Results are displayed in Fig. 5.

alpha-Amylase inhibitory assay

According to the method described in [24] the enzyme α-glucosidase (from Saccharomyces cervisiae) was dissolved in phosphate buffer (pH 6.8) in concentration of 4 U/mL and then centrifuged at 2000 rpm for 5 min to remove insoluble matter. PNPM solution (substrate) was dissolved in buffer at concentration of 1.25 mM and the stevia extract (TAqE) and fractions (SRF and FRF) were prepared in methanol in varying concentrations from 1000 to 7.81 μg ̸mL. Then, 20 μL phosphate buffer (50 mM, pH = 6.8) with 20 μL of enzyme, 20 μL of the sample solution and 20 μL PNPG were incubated at 37 °C for 10 min in a 96 well plate. The absorbance of the released p-nitrophenol was measured at 405 nm using multiplate reader. Acarbose (Sigma-Aldrich, Bangalore) was used as a standard at concentrations of 0.1575–1 mg/ml. Blank was set up in parallel as a control. Enzyme inhibitory activity was calculated as follows:

Results are displayed in Fig. 6.

alpha-Glucosidase inhibitory assay

According to [25] the enzyme (α-glucosidase from Saccharomyces cerevisiae) was dissolved in phosphate buffer (pH 6.8) in concentration of 1 U/ml and then centrifuged at 2000 rpm for 5 min to remove insoluble matter. The p-nitro-phenyl-α–D-glucopyranoside (p-NPG) substrate (Hi-media) PNP solution (substrate) was prepared by dissolving it in phosphate buffer in concentration of 5 mM. The stevia extract (TAqE) and fractions (SRF and FRF) were prepared in varying concentrations from 1000 to 7.81 μg ̸mL in DMSO. Then, 20 μL phosphate buffer (50 mM, pH = 6.8) was incubated with 20 μL of enzyme, 20 μL of the sample solution and 20 μL PNPG at 37 °C for 20 min, in a 96 well plate. The absorbance of the released p-nitrophenol was measured at 405 nm using multiplate reader. Acarbose at various concentrations (0.1575–1 mg/ml) was used as a standard. A blank was set up in parallel as a control. Enzyme inhibitory activity was calculated as follows:

Results are displayed in Fig. 7.

Acute oral toxicity test

Median lethal dose (LD₅₀) was determined for evaluating the safety of TAqE of S. rebaudiana leaves as described in [26]. Forty-eight male Westar rats (200 g) were divided into eight groups (6 animals each). They were orally administered single doses of the extract (ranging from 1 to 5 g/kg b.wt., the maximum soluble dose).

In vivo antihyperlipidemic and antihyperglycemic activities

According to [17] all animals except normal control were fed a high-fat diet HFD (total energy 25.07 kJ/g including fat 60%, protein 20% and carbohydrate 20%) for 4 successive weeks. After overnight fasting, STZ (40 mg/kg) was freshly prepared in a 0.05 M citrate buffer (pH 4.5) and injected i.p. Blood glucose level was monitored after 2 days using an Accu-check blood glucose meter (Roche Diagnostics, Basel, Switzerland). Animals having blood glucose levels ≥ 200 mg/dl were included in the experiment.

• Group I: Normal-control group received saline and normal diet orally.

• Group II: HFD/STZ was kept as positive control.

• Group III: HFD/STZ induced diabetic group was administered an oral dose of TAqE (300 mg/kg).

• Group IV: HFD/STZ induced diabetic group was administered TAqE with an oral dose of (500 mg/kg).

• Group V: HFD/STZ induced diabetic group was administered an oral dose of metformin (200 mg/kg) [27] as a standard anti-hyperglycemic drug.

• Group VI: HFD/STZ induced diabetic group was administered an oral dose of statin (1 mg/kg) [28] as anti-hyperlipidemic standard.

Body weight of the rats was measured weekly and fasting blood glucose was determined every 2 weeks. At the end of the experiment, the animals were fasted overnight, and blood samples were then collected by cardiac puncture. After standing for at least 30 min, the blood samples were centrifuged in centrifuge machine (Labcent 5000, Biosan England) at 3000 rpm for 15 min and the sera were stored at − 20 °C until use. Blood insulin level was determined by Rat Insulin ELISA kit (Alpco, UK). Serum lipid profile including total cholesterol and triglycerides was measured using Biodiagnostic colorimetric kits (Biodiagnostic, Cairo, Egypt). Alanine aminotransferase (ALT) and aspartate aminotransferase (AST) were determined by kits provided by Randox Laboratories Co. (Crumlin, Antrim, UK).

Statistical analysis

Data are presented as the mean ± SD and were analyzed by one-way analysis of variance (ANOVA) followed by Tukey’s test for multiple comparison using SPSS Software (Chicago, USA) and a trial version of Graph Pad Prism. Differences were significant at p˂ 0.05.