Comparative Study of Leaf and Rootstock Aqueous Extracts of Foeniculum vulgare on Chemical Profile and In Vitro Antioxidant and Antihyperglycemic Activities

Foeniculum vulgare is a medicinal plant used in Moroccan folk medicine to treat several diseases such as diabetes. The aim of this study was to determine the phenolic bioactive compounds and to evaluate the antioxidant and antihyperglycemic activities of Foeniculum vulgare leaf and rootstock extracts. Phenolic compounds of F. vulgare rootstock and leaf extracts were determined using HPLC-DAD-QTOFMS analysis. The antioxidant activity was evaluated using 1,1-diphenyl-2-picrylhydrazyl (DPPH) and 2,2'-azinobis-(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS•+) radicals. Moreover, the in vitro antihyperglycemic effects were tested by measuring the inhibition of α-amylase and α-glucosidase activities. HPLC-DAD-QTOFMS analysis identified thirty-two phenolic components in both leaf and rootstock extracts. Caffeic acid, quinic acid, and chlorogenic acid were the major compounds of F. vulgare leaf extract (FVLE), while the main compound of F. vulgare rootstock extracts (FVRE) was quinic acid. In the DPPH assay, F. vulgare leaf extract showed important antioxidant activity (IC50 = 12.16 ± 0.02 μg/mL) than F. vulgare rootstock extract (IC50 = 34.36 ± 0.09 μg/mL). Moreover, fennel leaf extracts revealed also the most powerful antioxidant activity (IC50 = 22.95 ± 0.4 μg/mL) in the ABTS assay. The in vitro antihyperglycemic activity showed that F. vulgare rootstock extract exhibited a remarkable inhibitory capacity (IC50 = 194.30 ± 4.8 μg/mL) of α-amylase compared with F. vulgare leaf extract (IC50 = 1026.50 ± 6.5 μg/mL). Furthermore, the inhibition of α-glucosidase was more importantly with F. vulgare rootstock (IC50 of 165.90 ± 1.2 μg/mL) than F. vulgare leaf extracts (203.80 ± 1.3 μg/mL). The funding of this study showed that F. vulgare rootstock and leaf extracts presented several phenolic compounds and showed important antioxidant and antidiabetic effects. We suggest that the identified molecules are responsible for the obtained activities. However, further studies focusing on the isolation and the determination of antioxidant and antidiabetic effects of F. vulgare rootstock and leaf main compounds are required.


Introduction
Recourse to the use of medicinal plants in the treatment of diseases has been common since antiquity [1,2]. e low risk of side effects allows them to be considered a good alternative to synthesized products [3,4]. Nowadays, the prevalence of diabetes is constantly increasing; it is a chronic disease characterized by a deficiency of endogenous insulin secretion by pancreatic β-cells and/or an altered action of this hormone [5]. Recently, the treatment of diabetes has attracted great interest whether in traditional medicine or scientific research. Indeed, several species of medicinal plants are scientifically evaluated and traditionally used for this purpose [6][7][8][9]. e antidiabetic properties of these plants are attributed to the presence of certain chemical compounds such as phenolic acids, flavonoids, and terpenoids [10]. Moreover, this activity may depend on several mechanisms aimed at stimulating the secretion of insulin by β-cells, the reduction of insulin resistance, the prevention against oxidative stress, and the inhibition of sugar digestive enzymes especially α-glucosidase and α-amylase [11][12][13].
Among the medicinal plants, Foeniculum vulgare L., Mill (Apiaceae), commonly known as fennel, a very common plant in the Mediterranean region [14,15]. It is used in traditional medicine to cure a variety of diseases, and its fruits were used as culinary spices [16]. e objective of the present study is to identify the phenolic composition of aqueous extracts prepared from F. vulgare leaves and rootstocks and to evaluate their in vitro antihyperglycemic and antioxidant properties.

Preparation of Plant Extracts. Foeniculum vulgare was collected in July 2017 from Rabat-Zaer region in Morocco.
e rootstock and leaves were cleaned and were then used for further investigation. Plant material (100 g) was extracted with water using the decoction method for 1 h. e obtained extracts were then filtered on Whatman paper, and the filtrate obtained was evaporated under reduced pressure, using a rotary evaporator.

Chemical Composition.
Chemical composition was determined using HPLC-DAD/TOFMS analysis of phenolic profiling according to the previous work done by Marmouzi et al. [17] with some modifications. In brief, an Agilent 1100 model series of liquid chromatography apparatus (HPLC, Agilent Technologies, Wilmington, DE, USA) was consisting of binary pump (G1312-A) and autosampler (G1330-B) and equipped to diode-array detector (G1315-B). e system was coupled to a time-of-flight (TOF) and a mass spectroscopic detector (MS) equipped with an electrospray ionizer source (ESI) (Micromass Quattro Micro, Agilent Technologies, Wilmington, DE, USA). e operational conditions were as follows: negative mode; range m/z 50-1200; nitrogen was used as desolvation and cone gas; cone gas flow rate at 30.0 L/h; desolvation gas flow rate at 350 L/h; cone voltage, 20 V; capillary voltage, 3.0 kV; extractor, 2 V; desolvation temperature 350°C; and source temperature 100°C. e phenolic separation was achieved on a C18 column (2.1 mm × 100 mm × 1.7 μm, Eclipse, XDB/Agilent Zorbax) at a temperature set at 35°C. e mobile phase composed of A (0.1% acetic acid in pure water) and B (0.1% acetic acid in acetonitrile) under a gradient elution (v/v) as follows: 0 min, 10% B; 0-18 min, 10-70% B; 18-20 min, 70-100% B; 20-23 min, 100-100% B; 20-23 min, 100-100% B; 23-25 min, 100-10% B; 25-30 min, 10-10% B at a flow rate of 0.5 mL/min, and 10 μL was injected. e identification of phenolic compounds was performed by comparing their mass spectra fragmentation and retention times with those of pure standards (Sigma-Aldrich, St. Quentin Fallavier, France), while the calibration curves were applied for their quantification. Each polyphenolic extract sample was dissolved in pure water; 10 mg of each extract was dissolved in water (10 mL) and then mixed by vortex and sonicated for 10 min; then, 1 mL was filtered at a syringe filter (PVDF, 0.2 μm) before the HPLC-DAD/TOFMS analysis.

Antioxidant Effect
2.3.1. DPPH Radical Scavenging Activity. Radical scavenging activity of the extracts was measured using the stable-free radical DPPH [6]. In brief, the solution of DPPH (0.2 mM) was prepared in methanol, and 0.5 mL of this solution was added to 2.5 mL of the extract and was allowed to stand at room temperature for 30 min. e absorbance was then read at 517 nm. Trolox was used as reference compound. e radical scavenging activity (RSA) was calculated using the following equation: where A0 is the absorbance of the control reaction and A1 is the absorbance of the sample solution. Concentration of the extract required to inhibit 50% of the free radical scavenging activity (IC 50 ) was determined.

ABTS Radical Scavenging
Activity. e ABTS radical scavenging activity of the plant extracts was estimated using the previously described method [8].
e radical cation ABTS •+ was produced through the reaction between 2 mM ABTS and 70 mM potassium persulfate in water. e mixture was stored at room temperature in the dark for 24 h prior use. e ABTS solution was then diluted with methanol to obtain an absorbance of 0.70 at 734 nm. 100 μL of appropriately diluted extracts was added to 2 mL of ABTS solution, and the absorbance was recorded at 734 nm after 1 min incubation at room temperature. Trolox was used as inhibitor standard. ABTS radical scavenging activity was calculated as described in equation (1).

α-Amylase Inhibition Assay.
e α-amylase inhibition assay was performed using the previously reported method [8]. In brief, 250 μL of various concentrations of the samples and 250 μL of α-amylase (240 U/mL, in 0.02 M sodium phosphate buffer, pH 6.9) were mixed and incubated at 37°C for 20 min. A portion (250 μL) of soluble starch (1% (w/v)) (in 0.02 M phosphate buffer, pH 6.9) was added, and the mixture was further incubated at 37°C for 15 min. Finally, 1 mL of dinitrosalicylic acid (DNS) color reagent was added and incubated in a boiling water bath for 10 min to stop the reaction. e mixture was then diluted with 2 mL of distilled water, and the absorbance was measured at 540 nm. Acarbose was used as positive control. e results were expressed as percentage inhibition and calculated using the following formula: where Ac refers to the absorbance of control (enzyme and buffer), Acb refers to the absorbance of control blank (buffer without enzyme), As refers to the absorbance of sample (enzyme and inhibitor), and Asb is the absorbance of sample blank (inhibitor without enzyme).

α-Glucosidase Inhibition
Assay. e α-glucosidase inhibitory activity of the extracts was estimated using pnitrophenyl-α-D-glucopyranoside (p-NPG) as substrate [18]. 150 μL of extracts at different concentrations were mixed with 100 μL of α-glucosidase enzyme solution (0.1 U/ml) prepared in 0.1 M sodium phosphate buffer (pH � 6.7) and incubated at 37°C for 10 min. en, a portion (200 μL) of 1 mM substrate p-NPG was added, and the mixture was incubated at 37°C for 30 min. e reaction was then stopped by the addition of 1 mL of Na 2 CO 3 (1 M), and the absorbance was recorded at 405 nm. e percentage inhibition of α-glucosidase enzyme of different concentrations of extracts was also measured using formula (2) described above, and the IC 50 values were calculated.

Statistical Analysis.
e significance of differences between multiple averages was determined by one-way analysis of variance (ANOVA), followed by Tukey's post hoc test at p < 0.05 significance level. Analysis was performed with GraphPad Prism 6.

Chemical Composition.
e phenolic chemical compounds of F. vulgare extracts were determined by HPLC-DAD-QTOFMS analysis. e results obtained are summarized in Table 1. As listed, thirty-two phenolic components were identified and quantified. e quantification analysis revealed important variabilities between FVLE and FVRE for all measured compounds (Table 1). Caffeic acid, quinic acid, and chlorogenic acid were the major phenolic compounds of FVLE by a percentage rate of 509203 ± 254. 60 e variability between phenolic compounds in FVLE and FVRE is attributed to the difference between organs and tissue in a plant to synthesize secondary metabolites such as phenolic compounds. Indeed, the anabolism of phenolic compounds in medicinal plants is depending on tissue specific regulation. is control is mediated by epigenetic factors, in particular DNA methylation, histone modifications, and chromatin compaction [19,20]. Previous works showed that F. vulgare is rich in bioactive compounds including phenolic components [21,22]. e phenolic composition of F. vulgare is not exactly similar to those found in the literature. Indeed, Méabed et al. [21] showed that the predominant phenolic compounds in F. vulgare aqueous extracts are ferulic acid, hesperidin, and chlorogenic acid. Moreover, acetone extract of F. vulgare showed the presence of palmitic, oleic, and linoleic acid as main components [22]. De Marino et al. [23] have isolated three phenolic glycosides compounds (one benzoisofuranone and two diglucoside stilbene trimers) from F. vulgare. is difference may be due to the geographical origins of the plant, environmental stimulus, climatic conditions, and extraction methods, which affect the yield of secondary metabolites [24,25]. On the contrary, Yaldiz and Camlica [26] have studied the variation in phenolic compounds, fatty acids, and volatile compounds of fruit extracts of different fennel genotypes. ey showed a remarkable variability which explains that genotype has an important role in synthesis and secretion of secondary metabolites in medicinal plants.

Antioxidant Activity.
Recently, the interest in natural antioxidants has considerably increased worldwide. Plants produce various antioxidants compounds such as phenolic acids, flavonoids, and tannins which have a potent capacity to prevent oxidative stress caused by reactive oxygen species. ese compounds have low or no side effects for use in preventive medicine in comparison to the synthetic antioxidant agents. Considering the complexity of the oxidation process, it is necessary to combine the responses of different and complementary tests to evaluate the antioxidant activity of samples. In this study, we evaluated antioxidant capacity of the aqueous extracts of the leaves and rootstocks of F. vulgare using the DPPH and ABTS radical scavenging methods.
e DPPH test is often used for the results rapidity as it is used for the screening of molecules with antioxidant activity present in plant extracts. e DPPH is a stable-free radical with a dark purple color; when a DPPH solution is mixed with a substance that can give a hydrogen atom, the color of the reaction mixture changes from purple to yellow with decreasing absorbance at wavelength 517 nm. e DPPH radical scavenging effect of the extracts is shown in Figure 1.
e results reveal that the extracts tested have a dose-dependent activity. In fact, at the concentration of 40 μg/mL, the aqueous extracts tested reduce the DPPH radical with an important percentage of 90.94 ± 0.09% and 53.60 ± 1.44% for FVLE and FVRE, respectively. Additionally, the IC 50 is inversely proportional to the antioxidant capacity of a compound. However, the lowest value of IC 50  However, they showed a relatively lower effect than that of Trolox (IC 50 � 1.47 ± 0.02 μg/mL). is antioxidant power of the aqueous extracts is explained by the presence of phenolic compounds including flavonoids present in the two parts studied and which are known as antioxidant substances with the ability to trap radical species and reactive forms of oxygen.
On the contrary, the antioxidant activity of samples using the ABTS method is deduced from their ability to inhibit the radical cation ABTS •+ . e ABTS •+ radical is in contact with a donor of H • leads, at 734 nm, to ABTS + and to the fading of the solution [27]. e ability of the aqueous extracts to scavenge ABTS •+ radical has been illustrated in Figure 2. e percentage inhibition of both extracts increases with increasing concentrations. Indeed, at the concentration of 90 μg/mL, FVLE and FVRE showed significant inhibition of 96.27 ± 0.33% and 27.72 ± 0.80%, respectively. e results show that the FVLE also possesses the most powerful antioxidant activity (IC 50 � 22.95 ± 0.4 μg/mL) against the ABTS radical (Table 2). e difference between FVLE and FVRE in the antioxidant effect could be attributed to the variation in the chemical composition. Indeed, several types of bioactive compounds known for their antioxidant activity [28][29][30][31][32] are identified in FVLE with high levels compared to FVRE, including flavonoids (rutin, quercitrin, and kaempferol) and phenolic acids (chlorogenic acid, ferulic acid, and sinapic acid) ( Table 1). On the best of our knowledge, no study has been conducted regarding the chemical composition and antioxidant activity of F. vulgare rootstock extracts. However, several studies have investigated the antioxidant activity of other parts of this plant, especially the seeds [33][34][35]. According to a study performed in Italy, the wild F. vulgare seeds n-Hexane extracts showed an important radical scavenging activity, with an IC 50 value of 31 μg/mL [36].  [38,39]. e results of this study suggest the therapeutic interest of this plant and justify the use of its aqueous extract in traditional Moroccan medicine.

Antihyperglycemic Activity.
Diabetes mellitus is characterized by high levels of glucose in the blood that can lead to several serious complications [40]. In modern medicine, the treatment of this condition is based on the use of insulin and oral antidiabetic drugs, which have side effects [41]. Medicinal plants and their active components can be a promising source for new safer antidiabetic drugs [42,43]. Many species of plants are used for this purpose [44]. In the present work, two extracts of F. vulgare are evaluated for their in vitro antidiabetic activity. e polysaccharides and disaccharides must undergo enzymatic hydrolysis to be converted into monosaccharide absorbable by the enterocytes. e enzymes α-amylase and α-glucosidase are responsible for the breakdown of carbohydrates in the digestive tract. When starch is ingested, it is hydrolysed in a proportion of 30 to 40% by α-amylase present in the saliva. e residual starch is then hydrolysed by pancreatic α-amylase into maltose which is converted into glucose by α-glucosidases present at the brush border of the intestinal cells. e inhibition of α-amylase and α-glucosidase activity is therefore an important strategy to limit the postprandial hyperglycemia [45,46].
To determine the in vitro antidiabetic activity of aqueous extracts of F. vulgare leaves and rootstocks, we tested their inhibitory properties against α-amylase and α-glucosidase. Indeed, the inhibitory activity of the extracts on α-glucosidase has been evaluated at different concentrations and the results are expressed as percentage inhibition (Figure 3). Both extracts showed α-glucosidase inhibitory effect depending on the concentration (140 to 300 μg/mL). Furthermore, the inhibitory activity of FVRE was more remarkable (IC 50 � 165.90 ± 1.2 μg/mL) than that of FVLE (IC 50 � 203.80 ± 1.3 μg/mL) (Table 3). Moreover, the extracts showed a dose-dependent inhibitory capacity on the α-amylase enzyme ( Figure 4). Also, FVRE exhibited an inhibitory effect (IC 50 � 194.30 ± 4.8 μg/mL) greater than that of FVLE (IC 50 � 1026.50 ± 6.5 μg/mL) and is significantly more potent (p < 0.05) than the reference drug acarbose (IC 50 � 311.20 ± 1.38 μg/mL) ( Table 3). e differences between FVLE and FVRE in the enzymes inhibitory action can be attributed to the variations in the percentage of inhibition with respect to the chemical composition of the plant parts as well as the enzymatic sensitivity.
According to the results found, the aqueous extracts of F. vulgare demonstrate an inhibitory capacity against two enzymes involved in sugar digestion and, consequently, a capacity to decrease the postprandial hyperglycemia and to prevent type 2 diabetes (T2DM).
is confirms that the compounds responsible for the antidiabetic activity of fennel are extractable in water, which explains the use of this herb to treat diabetes in folk medicine. In fact, as mentioned above, several phytochemicals were identified in the extracts studied with different concentrations, especially phenolic acids (quinic acid, chlorogenic acid, and caffeic acid) ( Table 1). ese compounds have been shown to possess antidiabetic activity attributed to several mechanisms. Effectively, Ooi et al. [47] have reported that 3,4-di-O-caffeoyl  Advances in Pharmacological and Pharmaceutical Sciences quinic acid has an inhibitory effect on α-glucosidase. Chlorogenic acid can inhibit the activity of α-amylase and α-glucosidase and decrease postprandial glucose [48] by reducing glucose transport in a synergistic way [49]. With regard to caffeic acid, it has been able to reduce hyperglycemia and prevent certain complications related to diabetes [50]. An in vitro study showed that caffeic acid derivatives could inhibit α-amylase, α-glucosidase, and angiotensinconverting enzyme related to T2DM [51]. Overall, phenolic compounds have an inhibitory effect on carbohydrate-hydrolyzing enzymes by their protein binding property [52].
According to Jung et al. [53], the antidiabetic power of caffeic acid has been attributed to major mechanisms such as stimulation of insulin production and glucose uptake by adipocytes, reduction of hepatic glucose level, and promotion of antioxidant potential. However, herbal medicine depends on the therapeutic effect of the combination of several compounds that act most often synergistically. It can be deduced that the bioactive compounds of fennel extracts can exert their in vitro antidiabetic and antioxidant activities in synergy. e present data are consistent with two recent studies evaluating the in vitro antidiabetic activity of F. vulgare on α-glucosidase and α-amylase [54,55]. Indeed, Abu-Zaiton et al. [54] found that the aerial parts of this plant had an inhibitory activity of 82.26% and 82.43% on α-glucosidase and α-amylase, respectively, and the same results were noted for the seeds of three different extracts of F. vulgare [55].
On the contrary, several studies have been conducted to evaluate antidiabetic effect of fennel in animal models. Indeed, the aqueous and ethanolic extracts of F. vulgare seeds showed correction of hyperglycemia, increased insulin, and improved lipid profile in streptozotocin-induced diabetic    rats [56][57][58]. e essential oils of this plant have also shown a significant hypoglycemic effect in diabetic rats [59,60]. Interestingly, extracts from other parts of this plant have also been investigated. Indeed, fruits [61], leaves [62], and aerial parts [63] were able to decrease blood glucose levels in diabetic rats and restore other parameters. All these data can explain the use of this herb to treat diabetes in folk medicine. e present study demonstrated the inhibitory effect of aqueous extracts prepared from F. vulgare leaves and rootstocks on the activity of α-amylase and α-glucosidase. To our knowledge, it should be noted that the inhibitory power of F. vulgare rootstocks was studied for the first time.

Conclusion
Both aqueous extracts from the leaf and rootstock of F. vulgare are rich in phenolic compounds, particularly phenolic acids. In vitro biological investigations showed that F. vulgare leaf and rootstock aqueous extracts exhibited important antioxidant and antihyperglycemic effects. Indeed, the rootstock extract (investigated for the first time) revealed important antihyperglycemic effects, which were mediated by the inhibition of enzymes implicated in sugar metabolism (α-amylase and α-glucosidase). On the contrary, the antioxidant activity of F. vulgare leaf and rootstock aqueous extracts can also be useful to improve the management of people with diabetes. ese finding suggest that phenolic compounds of F. vulgare are responsible for these biological effects. However, further investigations regarding the isolation of these main compounds and evaluation of their antioxidant and antidiabetic activities are required.

Data Availability
All the data associated with this work are included in the manuscript.  Advances in Pharmacological and Pharmaceutical Sciences 7