Influence of Extraction Techniques and Solvents on the Antioxidant and Biological Potential of Different Parts of Scorzonera undulata

The genus Scorzonera has various medicinal values. Species belonging to this genus were traditionally used as drugs or in food. The current study aimed to determine the phytochemical composition, antioxidant activity, and biological properties of the tuber, leaf, and flower of Scorzonera undulata extracts, collected from the southwest of Tunisia. Phenolic compounds from the three parts were extracted using two solvents (water and ethanol) and two extraction techniques (maceration and ultrasound). The total phenolic content was measured by the Folin–Ciocalteu assay. Furthermore, the chemical composition of Scorzonera undulata extract was also investigated by the LC-ESI–MS method using phenolic acid and flavonoid standards. The variation of the extraction methods induced a variation in the real potentialities of the three parts in terms of bioactive molecules. However, the aerial part of S. undulata (leaves and flowers) showed, in general, the highest phenolic contents. Twenty-five volatile compounds have been detected by GC-MS in S. undulata extracts; among them, fourteen were identified before derivatization. The DPPH test showed that the aerial part of the plant has a higher antioxidant activity compared to the tuber (25.06% at 50 µg/mL for the leaf ethanolic extract obtained by ultrasound extraction). For most biological activities (anti-Xanthine, anti-inflammatory, and antidiabetic (alpha-amylase and alpha-glucosidase)), the aerial parts (flowers and leaves) of the plant showed the highest inhibition than tubers.


Introduction
The genus Scorzonera of the Asteraceae family includes about 200 species. It is widespread in temperate and subtropical regions of Eurasia and North Africa [1]. This genus is represented by perennial herbs with a caudex or tuber, and it is rarely represented by biennials or dwarf subshrubs. The leaves are significantly longer than they are broad and are usually whole [2]. It is characterized by the production of multiflorous, homogamous, and ligulate flower heads. Scorzonera has great economic importance; several species belonging to this genus were used in food, including Scorzonera hispanica [3] and Scorzonera austriaca [4]. They are also employed for their therapeutic potential against colds, diarrhea, and fever caused by bacterial and viral infections, pulmonary disease, parasitic diseases, wounds,

Polyphenolic Extract Preparation
The secondary metabolites were extracted from the harvested parts using two different techniques: cold maceration, a conventional extraction method, and ultrasound, an environmentally friendly method.

Maceration
Dried powdered samples of S. undulata parts (1 g) were extracted with 20 mL of two solvents with different polarities (ethanol and water) for 24 h at ambient temperature and pressure with medium agitation. The homogenate was filtered using Wattman paper, and then the solvents were evaporated by a rotary evaporator (Staufen, Germany; IKA; RV 10 auto V) under a vacuum at 35 • C.

Ultrasound
The ultrasound extraction was applied using the ultrasonic bath (clean-120hd) for 30 min at a temperature of 30 • C. the mixture was filtered using Wattman paper, then the solvents were evaporated by a rotary evaporator (Staufen, Germany; IKA; RV 10 auto V) under a vacuum at 35 • C.

Total Phenolic Content (TPC)
The polyphenolic content was evaluated by Folin-Ciocalteu methods as described by Ben Khadher et al. [16] with some modifications. In total, 100 µL of Folin-Ciocalteu reagent (0.2 N) and 20 µL of each extract (1 mg/mL) were mixed in a 96-well microplate. After 5 min at room temperature, 80 µL of Na 2 CO 3 at a concentration of 75 g/L was added, and the mixture was then re-incubated for 15 min in darkness and at room temperature. The absorbance was measured with a microplate reader at a wavelength of 765 nm. The total phenolic content (TPC) was expressed as mg of gallic acid equivalents per g of dry weight (mg GAE/g DW).

Gas Chromatography-Mass Spectrometry (GC-MS) Analysis
The volatile compounds of S. undulata extracts were identified using the procedures of Rahmani et al. [17]. For analysis, the ethanol samples were prepared in their principal solvents, wherever the water samples were prepared in methanol, at a concentration of 5 mg/mL. The analysis was done on an Agilent 6890 gas chromatograph coupled to a 5975-mass detector. A volume of 1 µL of each extract was injected by the 7683 B autosampler with the use of a DB-5 MS fused silica capillary column (Supelco, Sigma-Aldrich, St. Louis, MI, USA) (30 m × 0.25 mm inner diameter, film thickness 0.25 µm). Helium (99.99% purity) was operated as a carrier gas at a flow rate of about 0.8 mL/min. Mass spectra have been registered at 70 eV with the use of an ion source temperature held at 310 • C and a transfer line heated at 320 • C. Every record of the acquisition was made in full scan mode (50-400 amu). Compounds were identified by comparison of their mass spectra with those obtained in NIST 08.

Derivatization Method
Using a modified method described by Rahmani et al. [17]. The various extracts were prepared in acetonitrile at a concentration of 5 mg/mL, and 60 µL of BSTFA (N, O-bis(trimethylsilyl)trifluoroacetamide) was added to the extracts (340 µL, 5 mg/mL) and mixed in a 2 mL vial. Then, the mixture was aerated with bubbling nitrogen and shaken for 30 s. The reaction mixture was maintained at 40 • C for 30 min. Finally, 2 µL of each derivatized solution was injected into the GC-MS and analyzed.
The data was collected and analyzed using the software Xcalibur. NIST 8 was used to identify the extracted compounds by comparison of the mass spectra.

Liquid Chromatography-Electrospray Ionization-Tandem Mass Spectrometry (LC-ESI-MS) Analysis
The LC-MS analysis of S. undulata extracts was performed in an LC-MS-2020 quadrupole mass spectrometer installed with an electrospray ionization source (ESI) and programmed to negative ionization mode. Separation was accomplished using an Aquasil C18 guard column (10 mm × 3 mm, 3 µm) and an Aquasil C18 (150 mm × 3 mm, 3 µm) (Thermo Electron, Dreieich, Germany) (Thermo Electron). A high-speed liquid chromatography system, including a binary pump system (LC-20AD XR), a column oven (CTO-20AC), a degassing system (DGU-20A 3R), and an autosampler (SIL-20AC XR), were connected online to the mass spectrometer (Shimadzu, Kyoto, Japan). The mobile phase was made up of 0.1% formic acid (v/v) in H 2 O as solvent A and 0.1% formic acid (v/v) in methanol as solvent B. Between each run, there was a regulated equilibration interval of 5 min. The column temperature was set at 40 • C, the injection volume was 5 µL, and the mobile phase flow rate was 0.4 mL/min. Spectra were analyzed using Shimadzu LabSolutions LC-MS software while being watched in SIM (selected ion monitoring) mode. The mass spectrometer was used in negative ion mode with a block source temperature of 400 • C, a dissolving line temperature of 250 • C, a 1.5 L/min nebulizing gas flow, a 12 L/min dry gas flow rate, a capillary voltage of −3.5 V, and the whole scan spectra from 50 to 2000 Da. The phenolic compounds of S. undulata extracts were identified by corresponding the retention times and mass spectra of the molecule with those of the real standards.

Antioxidant Activity (DPPH Radical Scavenging Activity)
The DPPH scavenging activity of S. undulata extracts was measured according to Dawra et al. [18]. In brief, 20 µL of various diluted plant extracts (50 µg/mL) were added to 180 µL of DPPH methanolic solution (0.2 mM), using a 96-well microplate. Vitamin C, at a concentration of 4 µg/mL, was used as a reference. After an incubation of 25 min at 25 • C, the absorbance was measured at 524 nm using a microplate reader. The following formula was used to determine the percentage of ability to scavenge the DPPH radical: where A control is the absorbance of the control, and A sample is the absorbance of the sample.
2.7. Biological Activities (Antienzymatic Activities) 2.7.1. Anti-Xanthine Oxidase Activity The anti-xanthine activity (anti-XOD) of various extracts was assessed. All S. undulata samples (50 µg/mL) were tested in 96-well plates. 50 µL of extracts and 30 µL of the XOD solution (0.1 U/mL) were combined in a 96-well plate with 60 µL of sodium phosphate buffer (Na 2 HPO 4 , 70 mM, pH = 7.5). A volume of 60 µL of xanthine solution (150 M) was added after 15 min of pre-incubation at 25 • C, and the plate was then incubated for 10 min, and the absorbance was measured at 295 nm. Allopurinol at 1 µg/mL was used as a standard. The following formula was used to determine the inhibition percentage of XOD activity: where A control is the absorbance of the control, and A sample is the absorbance of the sample.

Anti-5-Lipoxygenase Activity
The anti-5-lipoxygenase activity of S. undulata extracts has been determined according to Znat et al. [19], with modifications. A final volume of 250 µL was prepared by separately combining 20 µL of each extract with 150 µL of sodium phosphate buffer (pH 7.4), 20 µL of 5-lipoxygenase (5-LOX), and 60 µL of linoleic acid (3.5 mM). After 10 min of incubation at 25 • C, the absorbance was determined at 234 nm. The nordihydroguaiaretic acid NDGA, at 4 µg/mL, was used as a reference. The following formula was used to determine the 5-LOX activity inhibition: where A control is the absorbance of the control, and A sample is the absorbance of the sample.

Anti-α-Glucosidase Activity
α-glucosidase inhibitory activity has been determined according to Kim et al. [20], with modifications. In this study, 50 µL of sodium phosphate buffer (0.1 M, pH 8.0), 100 µL of enzyme solution, and 50 µL of plant extract were added in a 96-well microplate reader (MultiskanTM GO Thermo Scientific), and preincubated for 10 min at 25 • C. Then 50 µL of PNPG (5 mM) was added to the mixture as a substrate, and the plate was incubated at 250 • C for 5 min. Acarbose was used as a reference, and the absorbance was measured at 405 nm. The following formula was used to determine the α-glucosidase activity inhibition: where A control is the absorbance of the control, and A sample is the absorbance of the sample.

Anti-α-Amylase Activity
The anti-α-amylase activity of S. undulata extracts was determined using DNS, a spectrophotometry method that measures the concentration of the reducing sugars at 540 nm [21]. In an Eppendorf tube, 25 µL of enzyme solution (2 mg/mL) was added to 25 µL of various extracts and to 25 µL of plant extract (1.3 mg/mL). After incubation for 10 min at 25 • C, 1% starch solution was added to the mixture, and 3 min later the reaction of the enzyme solution was stopped by adding 50 µL of DNS solution (96 mM). The reaction mixture was placed in a boiling water bath for 10 min, followed by cooling at ambient temperature. Next, various solutions were diluted with 1 mL of distilled water. Acarbose was used as a positive control. Over time, the enzyme inhibitory activity was measured at 530 nm using a microplate reader. The following formula was used to determine the α-amylase activity inhibition: where A control is the absorbance of the control, and A sample is the absorbance of the sample.

Statistical Anylasis
Graphics were created using Origin Lab software. Data was done in triplicate and expressed as mean ± SD. All results were provided for the study of variance (ANOVA), accompanied by a Duncan test to assess if there were any significant differences (p < 0.05) between the averages. All statistical analysis and heatmaps were performed using XL-STAT 2009.

Extraction Yields
Extraction is a separation technique used to selectively separate one or more compounds from a mixture based on their chemical and physical properties. It is a very important step in the isolation and identification of bioactive molecules from the plant. There are several techniques for the extraction of biomolecules from plant sources. These techniques can be either conventional, also called traditional, or alternative [22]. In this study, S. undulata samples were extracted using a conventional (maceration) and non-conventional (ultrasound-assisted) technique. Ethanol and water were used to extract the three parts of S. undulata (TB, LV, and FL). The extraction yields obtained for all extracts were illustrated in Table 1. The yield of the S. undulata extracts ranged from 0.12 to 1.16% according to the plant parts, techniques, and solvents used. The best yield for tubers was achieved for an aqueous ultrasound extract, while the lowest was achieved for maceration (aqueous). For the leaves, maceration produced the highest yields (0.57 and 0.48%), although the lowest for ultrasound ethanolic extract. Finally, the highest yield for flowers was for the ultrasound aqueous extract and the lowest for the aqueous maceration. In most cases, the yields of the water extracts were higher than those of the ethanol, which has a lower polarity compared to the water, which indicates a difference in the chemical composition of the extract and a high content of polar compounds. The low yield of extraction was noted in other studies. It has been reported by S. Palmieri et al. [23] that on coriander seeds, the extraction yield ranged from 0.57 to 2.36% using ethanol and different extraction methods. For the hemp, the data shows that the ultrasound-assisted extraction led to a low yield (0.71%), similar to our findings for the aqueous leaf ultrasound extraction [23]. Our results are in line with data from the literature for the same techniques [24].

Polyphenol Content (TPC)
The total phenolic content (TPC) of S. undulata extracts was determined using the calibration curve (y = 0.0057x + 0.0285 and R 2 = 0.9998). Figure 1 shows that different plant parts, solvents, and extraction technique combinations led to a total polyphenol content ranging from 0.66 to 26.97 mg GAE/g DW from S. undulata. The data analysis showed a significant difference between extracts (p < 0.05). The highest TPC was measured in the leaf extract obtained by ultrasound extraction (26.97 mg GAE/g DW) and flowed at 23.97 mg GAE/g DW (MC-LV-H 2 O), while low TPC was observed for all tuber extracts. The total phenolic content was found in higher concentration in the aerial part (leaves (LV) and flowers (FL)) than in tuber extract (TB), which was confirmed by [25]. For the solvent, the TPC concentration is higher in ethanol than in the aqueous extract, except for the leaf extract obtained by maceration. Various researchers confirmed the results that the chemical profiles of plant extracts were dependent on the extraction method, solvent, and plant parts [25][26][27].

Identification of Volatile Compounds by GC-MS Analysis
The different extracts of S. undulate parts were analyzed by GC-MS (Tables 2 and 3   DW: dry weight). Extracts were tested at 50 µg/mL. Means values ± SD (n = 3). Different letters on the histograms mean a significant difference between extracts (p ≤ 0.05).

Identification of Phenolic Compounds by LC-MS Analysis
The results of high-performance liquid chromatography are presented in Tabl the two extraction methods (maceration and ultrasound). A total of twenty-one pounds were identified: eleven phenolic acids (quinic acid, 1,3-dicaffeoylquinic acid acid, protocatechuic acid, caffeic acid, synergic acid, p-coumaric acid, ferulic acid, c genic acid, rosmarinic acid, cinnamic acid) and ten flavonoids (rutin, hyperoside, lu

Identification of Phenolic Compounds by LC-MS Analysis
The results of high-performance liquid chromatography are presented in Table 4 for the two extraction methods (maceration and ultrasound). A total of twenty-one compounds were identified: eleven phenolic acids (quinic acid, 1,3-dicaffeoylquinic acid, gallic acid, protocatechuic acid, caffeic acid, synergic acid, p-coumaric acid, ferulic acid, chlorogenic acid, rosmarinic acid, cinnamic acid) and ten flavonoids (rutin, hyperoside, luteolin-7-O-glucoside, naringin, quercetin, apigenin, quercetin, kaempferol, naringenin, apigenin). For the maceration, the results in Table 4 reveal the presence of all twenty-one phenolic compounds distributed among the three parts (TB, LV, and FL). The tuber extracts are characterized by low levels of phenolic acids. The total concentration was estimated at 35.87 mg/100 g DM, with an abundance of quinic acid at 34.19 mg/100 g DW, while flavonoids were detected at a low concentration of 0.45 mg/100 g DW in total. Concerning the leaf extract, eight phenolic acids were detected. The highest content is observed for quinic acid at 165.55 mg/100 g DW, followed by chlorogenic acid at 131.34 mg/100 g DW. Quinic acid is a cyclohexanecarboxylic acid that can be found in the extracts of many different medicinal plant parts, such as those from Ziziphus lotus L., Haematocarpus validus, Coffea arabica, Ziziphus lotus L., Hypericum empetrifolium, and others, and it has a variety of biological properties, including antioxidant, anti-diabetic, and anti-cancer activity [31]. For the chlorogenic acid, Sarıaltın and Acıkara [32] confirmed the presence of this phenolic acid in both aerial parts and roots of Scorzonera extracts, with the highest amount in S. coriacea aerial parts (12.73 mg/g DW). Additionally, chlorogenic acid has been reported as the most vital phenolic acid for the Asteraceae family, with an anti-inflammatory and antioxidant effect [32]. For flavonoids, the leaves were characterized by ten compounds with a total concentration of 156.58 mg/100 g DW. The major flavonoids of this organ were luteolin-7-O-glucoside (245.11 mg/100 g DW), kaempferol (153.50 mg/100 g DW), apigenin (60.05 mg/100 g DW), and naringin (37.35 mg/100 g DW). The LC-MS patterns of the maceration show that the flower extracts have the highest content of phenolic acids, with a total concentration of 286.31 mg/100 g DW. Chlorogenic acid and quinic acid are still the major compounds (196.34 mg/100 g DM and 81.70 mg/100 g DM, respectively). In addition, the flower extracts contain eight flavonoids (154.95 mg/100 g DW), of which kaempferol, apigenin, luteolin-7-O-glucoside, quercitrin, and naringin are the major compounds. The methanolic extracts prepared by ultrasound extraction (Table 4) show phenolic acid contents close to those found by maceration, where the concentration was about 334.14 mg/100 g DW, 223.97 mg/100 g DW, and 43.42 mg/100 g DW, respectively, for the flowers, leaves, and roots. The major phenolic acids (quinic acid and chlorogenic acid) were generally the same for both extraction methods. The ultrasound extraction improves the flavonoid contents of the two aerial parts (322.39 and 199.50 mg/100 g DW for flowers and leaves, respectively). This may be due to a synergy between the solvent (methanol) and the effect of sound waves on flavonoid extraction. While the tuber extract still shows a low content of flavonoids (0.2 mg/100 g DW). The LC-MS pattern confirms the results of the colorimetric assay, which showed that the tuber part contained the lowest concentration of phenolic compounds. Our results confirm those found by Lendzion et al. [1] that show that both quinic and chlorogenic acids are major compounds in the majority of Scorzonera species, which explains their antioxidant powers. Other studies confirm the presence of major flavonoids in Scorsonera (quercetin, luteolin, apigenin, quercetin, and kaempferol) [25,33]. The structure of different compounds identified by LC-MS is represented in the supplementary file.

Antioxidant Activity
The antioxidant activity of S. undulata extracts was carried out using the DPPH scavenging activity, and results are presented in Figure 3. The extract concentration was tested at 50 µg/mL, and VIT C (ascorbic acid) was tested as a reference at 4 µg/mL. The antioxidant activity of the three parts (TB, LV, and FL) varied from 5.55% for the tubers ultrasound aqueous extract to 25.06% for the leaves ultrasound ethanolic extract. The extracts showed a pattern that was comparable to that of total phenolic content, where, in general, the aerial part and the ethanolic extracts showed the highest antioxidant activity. This could be explained by the positive correlation between the total polyphenolic content (TPC) and the antioxidant activity of the extracts. These results were in accordance with those reported for three Scorzonera species, indicating that the aerial parts of S. pygmaea showed a high antioxidant capacity, which was correlated with its total polyphenolic content [34]. It was noted in this study that the tubers of Scorzonera undulata have lower antioxidant activity, regardless of the solvents (water and ethanol) and the extraction methods (maceration and ultrasound) that were used. It can be explained by the fact that the tubers represent the organ reserve of S. undulata. The reserve carbohydrate in the Asteraceae family is inulin. It's a linear polysaccharide [35]. Due to its advantageous effects on gastric health, inulin is frequently used as a prebiotic, fat and sugar replacer, texture modifier, and for the creation of functional meals in order to promote health [36]. In the roots of S. hispanica, inulin represents 20% of the plant material [1]. The ANOVA test (p < 0.05) showed a statistically significant difference between the two extraction methods except for the tuber aqueous extract and the ethanolic leaf extract.

Antioxidant Activity
The antioxidant activity of S. undulata extracts was carried out using the DPPH scavenging activity, and results are presented in Figure 3. The extract concentration was tested at 50 µg/mL, and VIT C (ascorbic acid) was tested as a reference at 4 µg/mL. The antioxidant activity of the three parts (TB, LV, and FL) varied from 5.55% for the tubers ultrasound aqueous extract to 25.06% for the leaves ultrasound ethanolic extract. The extracts showed a pattern that was comparable to that of total phenolic content, where, in general, the aerial part and the ethanolic extracts showed the highest antioxidant activity. This could be explained by the positive correlation between the total polyphenolic content (TPC) and the antioxidant activity of the extracts. These results were in accordance with those reported for three Scorzonera species, indicating that the aerial parts of S. pygmaea showed a high antioxidant capacity, which was correlated with its total polyphenolic content [34]. It was noted in this study that the tubers of Scorzonera undulata have lower antioxidant activity, regardless of the solvents (water and ethanol) and the extraction methods (maceration and ultrasound) that were used. It can be explained by the fact that the tubers represent the organ reserve of S. undulata. The reserve carbohydrate in the Asteraceae family is inulin. It's a linear polysaccharide [35]. Due to its advantageous effects on gastric health, inulin is frequently used as a prebiotic, fat and sugar replacer, texture modifier, and for the creation of functional meals in order to promote health [36]. In the roots of S. hispanica, inulin represents 20% of the plant material [1]. The ANOVA test (p < 0.05) showed a statistically significant difference between the two extraction methods except for the tuber aqueous extract and the ethanolic leaf extract.

Biological Activities (Antienzymatic Activities)
To the best of our knowledge, the anti-xanthine oxidase, the antidiabetic activity (anti-alpha amylase, anti-alpha glucosidase), and the anti-inflammatory activity of S. undulata parts have not been reported in previous studies. In vitro tests, originally reported by Ben Khadher et al. [16], were used to estimate the anti-enzymatic activity of extracts ( Table 5). The different extract was tested at 50 mg/L. For the xanthine oxidase activity, allopurinol (1 µg/mL) was tested as a positive control, and the result shows low activity

Biological Activities (Antienzymatic Activities)
To the best of our knowledge, the anti-xanthine oxidase, the antidiabetic activity (antialpha amylase, anti-alpha glucosidase), and the anti-inflammatory activity of S. undulata parts have not been reported in previous studies. In vitro tests, originally reported by Ben Khadher et al. [16], were used to estimate the anti-enzymatic activity of extracts ( Table 5). The different extract was tested at 50 mg/L. For the xanthine oxidase activity, allopurinol (1 µg/mL) was tested as a positive control, and the result shows low activity for all the extracts (0 ≤ IP ≤ 38.26). The highest inhibition activity was detected for the leaf ethanolic extract obtained by maceration. This can be explained by the high amount of luteolin-7-O-Glucoside, which was identified as an inhibitor of Xanthine Oxidase in the Flos Chrysanthemum extract [37]. All tubers and extracts were inactive against xanthine oxidase. It can be explained by the fact that the S. undulata roots are devoid of the bioactive molecules responsible for this activity, such as flavonoids and their compounds (quercetin, rutin, naringenin, etc.), hydroxycinnamic acids (chlorogenic acid, dihydrocaffeic acid, 4-O-caffeoylquinic acid, etc.), tannins, and others [38]. The 5-lipoxygenase was used as an enzyme to evaluate the inflammatory activity. The 5-LOX is an enzyme that is considered to be a reason for many disorders, such as inflammatory disorders (rheumatoid arthritis and allergic rhinitis) and asthma [39]. The nordihydroguaiaretic acid (NDGA) was used as a reference for the anti-inflammatory activity. The activity levels of S. undulata ranged from 0% for most of the extracts to 14.05% for the ethanolic leaf extract obtained by ultrasound extraction. Our anti-inflammatory results are in harmony with those found by [40], who found that the essential oil of Lantara Camara showed low anti-inflammatory activity (IC 50 = 81.5 µg/mL). The antidiabetic activity of Scorzonera undulata extracts was tested using two enzymatic assays: anti-alpha glucosidase and anti-alpha amylase activities, and acarbose (50 µg/mL) was tested as a reference. Our result showed lower activity against alpha glucosidase; the inhibition percentage ranged from 0 to 9.77% for the aqueous flower extract obtained by ultrasound. However, for the alpha-amylase, the inhibition percentage at 50 µg/L ranged from 0 to 31.34% for the aqueous tuber extract obtained by maceration. Because this is the first study to be evaluated, comparisons cannot be made with previous results. The correlation between total polyphenol content (TPC), antioxidant (DPPH), and antienzymatic activities (antidiabetic (alpha-amylase and alpha-glucosidase), anti-inflammatory (5-LOX), and anti-xanthine oxidase (Anti-XOD)) were analyzed. According to the data represented in Figure 4, the Pearson correlation study revealed both positive and negative correlations. The correlation coefficient was 0.80 between TPC and DPPH. The correlation coefficient between polyphenol content and antioxidant activity in plant extracts has been the subject of several studies. In this regard, Maina, S. [41] noticed a significant corre-lation between TPC and DPPH (R 2 = 0.96) of Cleome gynandra L. Similarly, a study by Sarker, U. et al. [42] confirms the strong correlation between TPC and DPPH. Furthermore, we observed a moderately positive correlation between TPC and anti-5-LOX, DPPH and anti-5-LOX, anti-alpha glucosidase, and anti-5-LOX, anti-XOD and anti-5-LOX, TPC and anti-XOD, DPPH and anti-XOD, anti-alpha amylase and anti-XOD at R 2 = 0.43, 0.41, 0.32, 0.22, 0.27, 0.34, and 0.26, respectively. Conversely, strongly negative correlations were found for the anti-alpha amylase activity with anti-5-LOX, TPC, and DPPH. Our findings were in accordance with the data obtained from Wang, L. et al. [43].  [42] confirms the strong correlation between TPC and DPPH. Furthermore, we observed a moderately positive correlation between TPC and anti-5-LOX, DPPH and anti-5-LOX, anti-alpha glucosidase, and anti-5-LOX, anti-XOD and anti-5-LOX, TPC and anti-XOD, DPPH and anti-XOD, anti-alpha amylase and anti-XOD at R 2 = 0.43, 0.41, 0.32, 0.22, 0.27, 0.34, and 0.26, respectively. Conversely, strongly negative correlations were found for the anti-alpha amylase activity with anti-5-LOX, TPC, and DPPH. Our findings were in accordance with the data obtained from Wang, L. et al. [43].

Heat Map Analysis
Based on their polyphenol content, antioxidants, and biological activities, extracts are represented in the heatmap ( Figure 5). The S. undulata extracts can be clearly separated into two principal groups (A and B). The first group A contains three extracts (US.LV.EtOH, MC.FL.EtOH, and MC.FL.EtOH), and the second group B contains the other extract. Interestingly, the three extracts from group A are characterized by the highest polyphenol content, DPPH radical scavenging activity, and anti-xanthine activity. Furthermore, the positive correlation between the TPC and DPPH was clearly represented in group A (red color). Moreover, a high positive correlation between total polyphenol content and DPPH activity of tea was found by Dobrinas, S. et al. [44]. Group B can be divided into two other groups (a and b); the first group a contained only two extracts (MC.LV.EtOH and US.FL.H2O), which were characterized by low anti-xanthine oxidase and anti-alpha amylase activities. The other seven extracts were regrouped into group B, with a blue color predominance. The other seven extracts were regrouped into group B with a blue color predominance. This cluster indicated lower polyphenolic content and antienzymatic activity in this group.

Heat Map Analysis
Based on their polyphenol content, antioxidants, and biological activities, extracts are represented in the heatmap ( Figure 5). The S. undulata extracts can be clearly separated into two principal groups (A and B). The first group A contains three extracts (US.LV.EtOH, MC.FL.EtOH, and MC.FL.EtOH), and the second group B contains the other extract. Interestingly, the three extracts from group A are characterized by the highest polyphenol content, DPPH radical scavenging activity, and anti-xanthine activity. Furthermore, the positive correlation between the TPC and DPPH was clearly represented in group A (red color). Moreover, a high positive correlation between total polyphenol content and DPPH activity of tea was found by Dobrinas, S. et al. [44]. Group B can be divided into two other groups (a and b); the first group a contained only two extracts (MC.LV.EtOH and US.FL.H 2 O), which were characterized by low anti-xanthine oxidase and anti-alpha amylase activities. The other seven extracts were regrouped into group B, with a blue color predominance. The other seven extracts were regrouped into group B with a blue color predominance. This cluster indicated lower polyphenolic content and antienzymatic activity in this group.

Conclusions
To assess the bioactive composition of the specie Scorzonera undulata collected from southern Tunisia, a chromatographic analysis (GC-MS and LC-ESI-MS) was used to determine the volatile and phenolic composition of three parts of S. undulata. Its antioxidant potential and biological properties were evaluated. In total, twenty-one phenolic molecules were quantified in different extracts: eleven phenolic acids and ten flavonoids. The predominant compounds were quinic acid, chlorogenic acid, kaempferol, apigenin, luteolin-7-O-glucoside, quercitrin, and naringin. Twenty-five volatile compounds have been detected by GC-MS. Both aerial parts showed considerable antioxidant and biological properties compared to the tuber due to their richness in phenolic compounds. This study shows the difficulties in establishing a single extraction method and solvent that is effective for all the plant parts and highlights the significant variations between the three S. undulata parts under investigation.
Supplementary Materials: The following supporting information can be downloaded at: www.mdpi.com/xxx/s1, Figure S1: Structure of compounds identified in S. undulata extract by GC-MS before (a) and after (b) derivatization; Figure

Conclusions
To assess the bioactive composition of the specie Scorzonera undulata collected from southern Tunisia, a chromatographic analysis (GC-MS and LC-ESI-MS) was used to determine the volatile and phenolic composition of three parts of S. undulata. Its antioxidant potential and biological properties were evaluated. In total, twenty-one phenolic molecules were quantified in different extracts: eleven phenolic acids and ten flavonoids. The predominant compounds were quinic acid, chlorogenic acid, kaempferol, apigenin, luteolin-7-O-glucoside, quercitrin, and naringin. Twenty-five volatile compounds have been detected by GC-MS. Both aerial parts showed considerable antioxidant and biological properties compared to the tuber due to their richness in phenolic compounds. This study shows the difficulties in establishing a single extraction method and solvent that is effective for all the plant parts and highlights the significant variations between the three S. undulata parts under investigation.
Supplementary Materials: The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/life13040904/s1, Figure S1: Structure of compounds identified in S. undulata extract by GC-MS before (a) and after (b) derivatization; Figure S2  Data Availability Statement: The study did not report any data.

Conflicts of Interest:
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.