In Vitro Alpha-Glucosidase and Alpha-Amylase Inhibitory Activities and Antioxidant Capacity of Helichrysum cymosum and Helichrysum pandurifolium Schrank Constituents

: Diabetes mellitus (DM) is a group of systemic metabolic disorders with a high rate of mor-bidity and mortality worldwide. Due to the detrimental side effects of the current treatment, there is a great need to develop more effective antidiabetic drugs with fewer side effects. Natural products are a well-known source for the discovery of new scaffolds for drug discovery, including new antidiabetic drugs. The genus Helichrysum has been shown to produce antidiabetic natural products. In this investigation, the methanolic extract of H. cymosum and H. pandurifolium resulted in the isolation and identiﬁcation of eleven known compounds viz 5,8-dihydroxy-7-methoxy-2-phenyl ﬂavanone ( 1 ), pinostrobin ( 2 ), dihydrobaicalein ( 3 ), glabranin ( 4 ), allopatuletin ( 5 ), pinostrobin chalcone ( 6 ), helichrysetin ( 7 ), 5-hydroxy-3,7-dimethoxyﬂavone ( 8 ), 3,5-dihydroxy-6,7,8-trimethoxyﬂavone ( 9 ), 3-O-methylquercetin ( 10 ), and 3-methylethergalangin ( 11 ). The in vitro bio-evaluation of isolated compounds against alpha-glucosidase showed that 10 , 5 , and 11 demonstrated the highest alpha-glucosidase inhibitory activity with IC 50 values of 9.24 ± 0.4, 12.94 ± 0.2, and 16.00 ± 2.4 µ M respectively, followed by 7 and 3 with IC 50 values of 18.16 ± 1.2 and 44.44 ± 0.2 µ M respectively. However, none of these compounds showed a measurable inhibitory effect on alpha-amylase under the experimental conditions used except compound 10 which showed a poor alpha-amylase inhibitory activity with an IC 50 value of 230.66 ± 15.8 µ M. Additionally, strong total antioxidant capacities were demonstrated by 10 , 5 and 7 in ferric-ion reducing antioxidant power assay (374.34 ± 69.7; 334.37 ± 1.7; 279.93 ± 0.8) µ mol AAE/mmol. This is the ﬁrst scientiﬁc report to be carried out on alpha-glucosidase inhibitory activities and antioxidant capacities of H. cymosum constituents and a ﬁrst report on the isolation and identiﬁcation of methoxyﬂavanoids from H. pandurifolium . Our ﬁndings suggest that these compounds are promising candidates to inhibit alpha-glucosidase as well as oxidative stress related to diabetes. Results from molecular docking provided insight into the observed in vitro alpha-glucosidase inhibitory activities for 5 , 7 , 10 , and 11 . It is envisaged that the isolated phytochemicals from these plants may contribute to the development of hypoglycemic lead compounds with anti-diabetic potential. ﬁnal results the concentration range was made up between 0 mg/L (blank) and 500 mg/L of L-ascorbic acid for analytical measurement, where each concentration of L-ascorbic acid was measured in triplicate. The results were reported as µ mol ascorbic acid equivalents per mmol ( µ mol AAE/mmol) of the dry weight test samples.

Traditionally, H. pandurifolium ( Figure 1B) is used to treat respiratory conditions, heart troubles, constipation, back pain, and kidney stones [14]. Biological activity of H. pandurifolium total extract using chloroform:methanol (1:1) has been previously tested for antibacterial activity and cytotoxicity [22]. Currently, there are no reports of anti-diabetic or antioxidant studies on this species or any of its isolated compounds. Although H. pandurifolium finds many uses traditionally, it is a relatively understudied species of the Helichrysum genus [23].
As a part of a study to explore Helichrysum plants for hypoglycemic agents in our lab, the methanolic extracts of H. cymosum and H. pandurifolium exhibited activity against the alpha-glucosidase enzyme. In a quest to identify the active phytochemical constituents, the phytochemical analysis of these plants was undertaken. The chemical structure identification of the isolated phytochemical constituents from these plants was performed and isolated constituents were biologically evaluated for in vitro enzyme inhibitory effects on key digestive enzymes α-glucosidase and α-amylase in addition to their evaluation of their antioxidant properties by Ferric-ion Reducing Antioxidant Power (FRAP) and Trolox Equivalent Absorbance Capacity (TEAC) antioxidant assays. Molecular docking was used to explain the observed inhibitory activities as previously done for isolated constitutes from Plectranthus ecklonii [24].
Nuclear magnetic resonance (NMR) spectra were recorded on Bruker 400 MHz instrument. The chemical shifts were reported as δ values in ppm downfield from tetramethylsilane (TMS) and deuterated residual peaks were used as internal standards. Absorbance values on 96-well plates were measured in clear 96-well microplates (Greiner, height 14.8 mm) using a UV plate reader (Molecular Devices Spectra Max i3X).

Collection and Identification of the Plant Material
The wild-growing H. pandurifolium was collected from the Cape Point Nature Reserve in Cape Town, South Africa in December 2017 whereas the aerial part of H. cymosum species was collected in January 2018, from Hout Bay, Cape Town, South Africa. The voucher specimens (Helichrysum cymosum-UFH 2020-10-02 and Helichrysum pandurifolium-UFH 2020-10-04) were identified by Prof. Christopher Cupido and deposited at the Department of Botany, University of Fort Hare, South Africa.

Helichrysum pandurifolium
The aerial parts (leaves and flowers) of H. pandurifolium (374.5 g) were extracted with methanol as previously stated. The total extract (42.2 g) was subjected to flash column chromatography using silica gel 60 H (height 70 cm, radius 5.5 cm, 6.6 L volume) and eluted using a gradient elution of hexane, ethyl acetate, and methanol. A total of fifty fractions were collected and combined according to observations after TLC plate analysis, resulting in twelve main fractions. Main fraction III (4.0 g) was subjected to a medium-sized silica gel column (90 × 2.5 cm, 1.76 L volume) and separated using a gradient elution of hexane and ethyl acetate, resulting in twelve subfractions. The main fraction IV (5.9 g) was subjected to the same conditions and resulting in twenty-four subfractions. Compound 9 (100 mg) was obtained via crystallization using methanol. The main fraction VI (8.17 g) was subjected to the same conditions followed by Sephadex LH-20 using 5.0 aqueous methanol to yield compounds 10 (132 mg). Compound 11 (20 mg) was purified from fraction VI using preparative TLC.

Structure Elucidation and Identification of the Isolated Compounds
All the isolated compounds were elucidated by NMR spectroscopy. 1 H, 13 C, and distortionless enhancement of polarization transfer (DEPT) analysis were performed for structural elucidation. The carbon peaks in the 13 C NMR spectroscopy were further differentiated as -CH 3 , CH, or CH 2 carbons by DEPT-135 analysis. The NMR data of the compounds was compared to that available in the literature for the final identification of the compounds.

Alpha-Glucosidase Assay
The alpha-glucosidase assay of the tested compounds (1-11) was conducted according to the slightly modified standard method [25]. Into a 96-well clear microplate, 50 µL of the α-glucosidase phosphate buffer solution (pH 6.5) was added to all wells. 10 µL of the α-glucosidase enzyme solution was added to the odd-numbered wells and 10 µL of the phosphate buffer solution was added to the even-numbered wells. To the designated control wells 20 µL of methanol (HPLC grade > 99%) was added and 20 µL of the sample stock solutions (concentration range of 1 mg/mL to 0.125 mg/mL) were added in triplicate. The plate was then incubated at 37 • C for 15 min. 20 µL of the substrate solution was added to the wells containing the enzyme and 20 µL of the phosphate buffer solution was added to the remaining wells to keep the total volume constant. The plate was incubated for a further 20 min at 37.5 • C after which 50 µL of the stop solution (0.1 M Na 2 CO 3 ) was added to all the wells. The absorbance was measured at 25 • C at a wavelength of 405 nm in the plate reader (Molecular Devices Spectra Max i3X). The final results are reported as the half-maximal inhibitory (IC 50 ) concentration in µM. The results represent the mean of three independent experiments and are expressed as percentage inhibition calculated as stated below.
Inhibitory activity (%) where A is the absorbance in the presence of the test substance and B is the absorbance of control.

Alpha-Amylase Assay
The α-amylase inhibition assay was conducted according to the slightly modified standard method [25]. Into a 96-well clear microplate, 50 µL of the α-amylase buffer solution (maleic acid, sodium chloride, calcium chloride, and sodium azide solution adjusted to pH 6.5 using sodium hydroxide) was added to all the wells. 10 µL of the α-amylase enzyme solution was added to the odd-numbered wells and 10 µL of the buffer solution was added to the even-numbered wells. To the designated control wells, 20 µL of methanol (HPLC grade > 99%) was added and 20 µL of the sample stock solutions (concentration range between 4 mg/mL to 0.03125 mg/mL) were added in triplicate. The plate was then incubated at 37.5 • C for 15 min after which 20 µL of the starch solution was added to all the wells. The plate was then incubated for a further 20 min at 37.5 • C. 50 µL of the DNS reagent (3,5-dinitrosalicylic acid) was added to stop the reaction and react with the reducing sugars. The plate was then incubated in a covered casing and placed into a water bath at 98.5 • C for 20 min. The absorbance was then measured at a wavelength of 540 nm in the plate reader (Molecular Devices Spectra Max i3X). The final results are reported as the half-maximal inhibitory (IC 50 ) concentration in µM and represent the mean of three independent experiments and are expressed as percentage inhibition calculated by Equation (1).

Ferric-Ion Reducing Antioxidant Power (FRAP) Assay
The FRAP assay was evaluated as per previously described methods with slight modification [26]. The FRAP mechanism is based on electron transfer rather than hydrogen atom transfer and is based on the ability of phenols to reduce Fe 3+ to Fe 2+ ions by transferring an electron. The reaction is carried out at a pH of 3.6 to maintain iron solubility and to decrease the ionization potential that drives electron transfer. As a result, the redox potential increases resulting in electron transfer being the dominant mechanism [27]. The Fe 2+ ion in the presence of TPTZ (tripyridyl triazine) forms a colored complex (λ max = 593 nm) which can then be measured using a spectrophotometer to determine the extent of iron reduction by the sample.
The FRAP reagent was prepared using a combination of pre-prepared stock solutions namely the FeCl 3 solution, the TPTZ solution, and the acetate buffer solution. The FeCl 3 solution was prepared by dissolving 0.053 g FeCl 3 into 10 mL H 2 O. The TPTZ solution was prepared by dissolving 0.0186 g TPTZ into 6 mL of a 0.1 M HCl solution. The acetate buffer solution was prepared using acetic acid and the conjugate base sodium acetate such that the buffer solution maintained a pH of 3.6. The FRAP reagent was prepared by combining 3 mL of the FeCl 3 solution, 3 mL of the TPTZ solution, 30 mL of the buffer solution, and 6 mL H 2 O into a 50 mL test tube. Into a 96-well clear visible range microplate, 10 µL of the isolated compound stock solutions (8 and 11 = 1 mg/mL; 9 = 0.0625 mg/mL; 10 = 0.25 mg/mL), as well as the total extract (1 mg/mL) were added into separate microplate wells in triplicate and then mixed with 300 µL of the FRAP reagent. The microplate was incubated in dark at room temperature for 30 min, after which the plate was read in a Molecular Devices Spectra Max i3X at a wavelength of 593 nm. The standard calibration curve was obtained using L-ascorbic acid (Sigma Aldrich, Johannesburg, South Africa) as a standard. The concentration range was made up between 0 mg/L (blank) and 500 mg/L of L-ascorbic acid for analytical measurement, where each concentration of L-ascorbic acid was measured in triplicate. The results were reported as µmol ascorbic acid equivalents per mmol (µmol AAE/mmol) of the dry weight test samples. The assay was evaluated as per previously described methods [28]. The TEAC assay measures the ability of antioxidants to scavenge free radicals. In this assay, the stable free radical ABTS + (2,2'-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid)), is a blue-green chromophore (λ max = 734 nm) that in the presence of antioxidants decreases its absorption intensity. This occurs by the antioxidants' neutralization of free radicals either by direct reduction (electron transfer) or by hydrogen atom donation. The balance between these two mechanisms is usually determined by the antioxidant structure and the pH of the reaction medium [29].
The working solution was prepared by combining two pre-prepared stock solutions namely, 7 mM ABTS and 140 mM potassium-peroxodisulfate (K 2 S 2 O 8 ) (Merck, South Africa). To 5 mL of the ABTS solution, 88 µL of K 2 S 2 O 8 solution was added, mixed thoroughly, and then allowed to react in the absence of light for 24 h. Trolox (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid), a water-soluble analog of vitamin E, was used as the standard to obtain the standard calibration curve with the concentration range between 0 mg/L and 500 mg/L. Into a 96-well clear microplate, 25 µL of the isolated compound and total extract stock solutions (1 mg/mL) were added in triplicate wells with 275 µL of the working solution and allowed to react in the absence of light at room temperature for 30 min. The absorbance was measured at 25 • C at a wavelength of 734 nm in the plate reader. The final results are reported as µmol Trolox equivalents per mmol (µmol TE/mmol) of the dry weight test samples.

Statistical Analysis
All the measurements were repeated three times and the IC 50 was calculated using GraphPad Prism 8 version 8.4.3 (Graph pad software, Inc., La Jolla, CA, USA) statistical software. The data presented are means ± SD obtained from 96 well plate readers for all in vitro experiments.

Molecular Docking and Molecular Mechanics-Generalized Born Surface Area (MM-GBSA) Calculations
The combined sdf file of compounds 1-11 were imported into Schrodinger Maestro ®® [30] and prepared with the LigPrep wizard [31]. The chiralities, tautomers, and ring conformations of the compounds were corrected and protonated at a physiological pH range of 7 ± 2. The OPLS3e force field was then used to minimize the protonated compounds.
The X-ray diffraction three-dimensional model (3D) structure of α-Glucosidase (7KRY) and amylase (1B2Y) in the protein data bank was imported into the Schrodinger Maestro ®® as the 3D protein target. The chain A part of the 3D models was isolated and the protein preparation wizard [32][33][34] was then used to ensure structural correctness, add missing hydrogen atoms, charges, and correct bond orders. The 3D protein target was energy minimized to relieve steric clashes by applying the default constraint of 0.3 Å root-meansquare-deviation (RMSD) and the OPLS3e forcefield.
Before molecular docking, the grid generation panel was used to define a grid box for the binding pocket using the centroid of the co-crystallized ligand in both 3D models. Next, the molecular docking calculation was executed with the ligand docking panel with extra precision (XP) in the Glide module [35]. The binding mode and interaction were visualized with UCSF Chimera [36].

Isolation and Chemical Characterization of Compound Isolated from H. cymosum and H. pandurifolium
Repeated silica gel and Sephadex column chromatography of the methanolic extract of H. cymosum and H. pandurifolium yielded seven (1-7) and four (8-11) known flavonoids respectively (Figure 2), the structures of these eleven compounds were confirmed by comparison of spectroscopic data to the literature data.

Alpha-Glucosidase and Alpha-Amylase Activity
The alpha-glucosidase enzyme is situated in the brush border of the small intestine epithelium. It catalyzes the breaking down of the reaction of disaccharides and starch to glucose [37]. Glucosidase inhibitors decrease the rate of carbohydrate digestion and retard carbohydrate absorption from the alimentary tract [38]. In the search for effective hypoglycemic candidates from natural sources, a great deal of research is being carried out to identify phytoconstituents having a hypoglycemic effect with emphasis on the inhibition of alpha-glucosidase and alpha-amylase. Table 1 presents the IC 50 values in µM concentrations for α-glucosidase and α-amylase inhibition by flavonoids (1-11) and in µg/mL for the total methanolic extract of H. cymosum and H. pandurifolium. Acarbose was used as positive control and was found to have an α-glucosidase inhibition of 945.46 ± 1.0 µM.
An investigation of the inhibitory activities of H. cymosum total extract and its natural products showed that 5 demonstrated the highest alpha-glucosidase inhibitory activity with an IC 50 value of 12.94 ± 0.2 µM, followed by 7 and 3 with IC 50 values of 18.16 ± 1.2 and 44.44 ± 0.2 µM respectively. However, H. cymosum total extract and its isolated compounds did not show any measurable inhibition of alpha-amylase enzyme under the experimental conditions used in this study.

Antioxidant Activity
Oxidative stress is well known as one of the principal causes of diabetic complications due to the accumulation of free radicals, which cause mitochondrial superoxide overproduction, β-cell dysfunction, and diabetes vascular damage [39]. The in vitro antioxidant activity of the isolated compounds from H. cymosum and H. pandurifolium was investigated by evaluating their ferric ion reducing antioxidant power (FRAP) and Trolox Equivalent Absorbance Capacity (TEAC) activities. TEAC is an assay based on a hydrogen atom transfer (HAT) mechanism [40], while FRAP is based on a single electron transfer (SET) mechanism [41].
The results of the in vitro antioxidant activity study demonstrated that compounds 5 and 7 showed strong activity on FRAP (334.37 ± 1.7; 279.93 ± 0.8) µmol AAE/mmol, followed by compound 1 which showed a moderate activity (185.72 ± 1.1) µmol AAE/mmol as shown in Table 2.  Table 2 also summarises the antioxidant capacities measured for the total extract of H. pandurifolium and its four isolated flavonoids. Compound 10 was found to be the phytochemical with the greatest antioxidant capacity of the four compounds isolated from H. pandurifolium.

Molecular Docking
The molecular docking of compounds 1 to 11 and acarbose as a reference on the active sites of the 3D model of alpha-glucosidase (neutral alpha-glucosidase ab chain A) and alphaamylase showed free binding energy ranging from −5.7 kcal/mol to −13.07 kcal/mol for alpha-glucosidase and −4.0 kcal/mol to 12.06 kcal/mol for alpha-amylase (Figure 3). It was observed that some of the active compounds from the in vitro alpha-glucosidase assay (compounds 5, 10, and 11) have a stronger binding affinity (free energy of binding) for the enzymes than the other compounds. A closer look at these compounds and their interaction with the active site of the alpha-glucosidase 3D model revealed a hydrogen bond (at a minimum distance of 2.5 Å and 20° angles) with amino acid residues at the active site ( Figure 4). It was observed that some of the active compounds from the in vitro alpha-glucosidase assay (compounds 5, 10, and 11) have a stronger binding affinity (free energy of binding) for the enzymes than the other compounds. A closer look at these compounds and their interaction with the active site of the alpha-glucosidase 3D model revealed a hydrogen bond (at a minimum distance of 2.5 Å and 20 • angles) with amino acid residues at the active site (Figure 4).
Alpha glucosidases cleave dietary starch and oligosaccharides into glucose. A potential inhibitor of alpha-glucosidase should block the substrate-binding active site of the enzyme, thereby preventing the oligosaccharides from accessing the active site. Within the active site are two highly conserved and important residues that are pivotal for the binding of substrates to the enzyme: Asp564 acts as the catalytic nucleophile and Asp640 act as the acid/base catalyst [42]. Therefore, interactions with the catalytic nucleophile Asp564 and acid/base catalyst Asp640 amino acid residues might be a predictor of alphaglucosidase inhibition. Compounds 5 and 10 showed two hydrogen bond interactions with catalytic nucleophile Asp564 and one interaction with acid/base catalyst Asp640. Compound 11 showed two interactions with catalytic nucleophile Asp564 but no interaction with acid/base catalyst Asp640. This may account for the observed inhibitory activities. Compound 7 showed no interaction with either of the residues, the reported inhibitory activity from the in vitro assay might be due to its van der Waals interactions with Asp640.
It was observed that some of the active compounds from the in vitro alpha-glucosidase assay (compounds 5, 10, and 11) have a stronger binding affinity (free energy of binding) for the enzymes than the other compounds. A closer look at these compounds and their interaction with the active site of the alpha-glucosidase 3D model revealed a hydrogen bond (at a minimum distance of 2.5 Å and 20° angles) with amino acid residues at the active site ( Figure 4).

Discussion
The structures of the isolated compounds were confirmed by comparing their spectroscopic data to the reported data in the literature. Compound 1 has been reported to be isolated from Echiochilon fruticosum [43] and it showed antioxidant, anti-inflammatory, and hepatoprotective properties [44]. Compound 2, isolated from H. gymnocomum and Echiochilon fruticosum [20,43,45], showed antibacterial, anti-inflammatory, antitumor, antidiarrhea, anti-HIV, antioxidant, and antipyretic activities [46]. Isolation of compound 3 was reported to be from Scutellaria baicalensis [47] and it showed a wide range of biological activities including anticancer [48]. Compound 4 was isolated previously from H. cymosum and H. teretifolium [19,21] and it showed antibacterial and antioxidant activities [49]. Compound 5 was isolated from Xanthium strumarium [50], which also belongs to the same family Asteraceae and it possesses antibacterial and antifungal activities [51]. Compound 6 was isolated from Helichrysum trilineatum and extracts of Piper claussenianum and demonstrated antidiabetic and antibacterial activities [52,53]. Compound 7 was isolated from Helichrysum odoratissimum and was reported to possess a wide range of pharmaceutical applications including antiproliferative, anticancer, antiplatelet, and antioxidant [54].
The alpha-glucosidase inhibitory activity of the isolated compounds (1-11) from both H. cymosum and H. pandurifolium revealed the compounds 10, 5, 11, and 7 to be the most active with compound 10 exhibiting the highest activity followed by compounds 5, 11, and 7. Among the compounds isolated from H. cymosum, 5 and 7 showed the highest alpha glucosidase inhibition whereas compounds 10 and 11 showed the highest inhibition among compounds isolated from H. pandurifolium. None of the isolated compounds showed any assessable inhibitory effect on alpha-amylase under the experimental conditions used except compound 10 which showed a poor alpha-amylase inhibitory activity.
Compound 7, Helichrysetin, has not been reported to possess hypoglycemic activity. However, phloretin, which belongs to the same class of compounds as 7, has been reported to exhibit strong reversible alpha-glucosidase inhibitory activity in a mixedtype manner with an IC 50 value of 31.26 µg/L [59]. Its glucoside (phlorizin) inhibits the sodium-dependent glucose transporters (SGLT1 and SGLT2) resulting in the reduction of the absorption and reabsorption of glucose from the small intestine and kidneys by SGLT1 and SGLT2 respectively [60]. Synthetic derivatives of phlorizin are clinically used in the treatment of diabetes [61]. Compound 7 has also been reported to exert a hypoglycemic effect in streptozotocin-induced diabetic rats and reduces insulin resistance [62]. On the other hand, pinostrobin chalcone (6) has been reported to be responsible for the reduction of blood glucose levels in rats with streptozotocin-induced diabetes [53]. Unfortunately, in this study, this compound showed moderate activity when tested for the alpha-glucosidase enzyme. Remarkably, the SAR studies demonstrated that the more polar compounds among the tested (1-7) exhibited the highest activities.
The enzymatic inhibition results of compounds (8)(9)(10)(11) isolated H. pandurifolium were in accordance with the SAR studies as previously described by Proença et al. for both α-glucosidase and α-amylase [11,12]. The greater inhibitory effect of compound 10 on alphaglucosidase and alpha-amylase enzymes can be attributed to the catechol group on ring-B, hydroxy-functional groups at 5-, 7-, or 8-position of ring-A, and the C2=C3 double bond in ring-C, all of which are crucial for significant flavonoid α-glucosidase and α-amylase activity. In literature, the flavonoids have been reported to possess significant antidiabetic activity due to the hydroxyl group allocated at C-3 on the A-ring and additional hydroxyl groups attached to the C-ring, which play an important role in the inhibition process of the alpha-glucosidase enzyme as well as hypoglycemia, insulin activation, glucose uptake activation [63]. Additionally, flavonoids are commonly used in antidiabetic diets and many of them include catechins, fisetin, kaempferol, luteolin, naringenin, quercetin, rutin, morin, silymarin, chrysin, baicalein, icariin, isoliquiritigenin, diosmin, isoangustone A and genistein were reported to be active for their antidiabetic properties [64].
It is proposed that to further establish these compounds as potential hypoglycemic agents, their inhibitory effect on other glucose regulating enzymes and the enzymes for lipid digestion and absorption should be investigated. This would further pave the way for in vivo studies of the promising hypoglycemic candidates in these investigations. It is envisaged that the results of in vivo studies would give us an indication of the possible doses appropriate for use in humans.
In antioxidant study experiments, compounds 5 and 7 from H. cymosum and compound 10 from H. pandurifolium displayed the highest activities. Compounds 5 and 7 have already been reported to be good candidates to serve as safe antioxidants [65]. The essential oil of H. cymosum has been reported to possess radical scavenging activity with an SC 50 (concentration that causes 50% free radical scavenging) value of 6.3 mg/mL [66]. Phenolic compounds have been reported to be responsible for the antioxidant activity of numerous plant species by stabilization of free radicals by donating electrons or by metal ion complexation among other mechanisms. Nevertheless, other aspects can be considered, for example, the presence of vicinal hydroxyl groups is essential in a pronounced antioxidant activity [67]. Flavonoids represent a very large class of plant secondary metabolites found in a wide range of fruits, vegetables, and herbs that can play a prominent role as natural antioxidants, due to the presence of hydroxyl groups and aromatic rings in their chemical structures [68]. In general, the presence of a pair of hydroxyls at ortho position to each other (cathechol configuration) on the B-ring is the most important active group, for the Separations 2022, 9,190 13 of 16 hydrogen atom transfer/single electron transfer to the hydroxyl/peroxyl radicals, thereby stabilizing them and giving rise to a relatively stable flavonoid radical [69]. Our results are therefore an indication that 7 and 2 have great capacities for hydrogen atom(s) transfer from B-hydroxyl groups to stabilize free radicals.
The highest antioxidant activity of compound 10 is in accordance with the theory that more hydroxyl functional groups on the flavan backbone structure lead to greater antioxidant activity, as described in previous studies [70,71]. This also confirms the significance of the 3 -and 4 -hydroxy groups (catechol) on the B-ring. Compound 8 on the other hand, having only one free OH at position 5, showed no significant activity in the FRAP antioxidant assay and relatively low activity in the TEAC antioxidant assay when compared to the other compounds. Compound 11 showed similar, relatively low antioxidant activity in the FRAP assay likely due to the methylated hydroxyl functional group at position 3. The significance of the free hydroxy on the C-3 is evident from the FRAP assay results, where compound 9, the only compound with a free 3-OH, yields significantly greater antioxidant activity when compared to the others (excluding 10), even though the positions C-6, C-7, and C-8 are all methoxylated. The total methanolic extract of H. pandurifolium, when compared to other methanolic extracts of Helichrysum, showed FRAP antioxidant activity of 43.7 ± 0.1 mg AAE/g, significantly lower than those of the various other Helichrysum species summarised by Albayrak, et al. [72].
Overall, the molecular docking gave insight into the observed results from the in vitro inhibitory alpha-glucosidase assay. The only major departure was observed for acarbose (the reference drug which is a known inhibitor of alpha-glucosidase). The molecular docking aligned with the reported use of acarbose as a potent inhibitor of alpha-glucosidase by the predicted high negative free binding energy (over −13 kcal/mol). The departure observed for in vitro inhibitory activity might be due to non-optimal assay conditions for acarbose (which is a maltotetrose analog, possessing a free end that may act as a reducing agent). An excessive amount of acarbose in the assay medium might react with 3, 5, dinitro salicylic acid (DNS) (the chromophore for this colorimetric assay) thereby causing a substantial~5-fold effect on acarbose IC 50 values when working just outside optimal concentration range [73].

Conclusions
The phytochemical analysis of the methanolic extracts of H. cymosum and H. pandurifolium yielded eleven known flavonoids. Their structure was elucidated by NMR spectroscopy and the spectroscopy data of the isolated compounds was compared to the literature data. The in vitro bio-evaluation of the extracts and their phytoconstituents revealed these plants to be a potential source of flavonoids with strong alpha-glucosidase inhibitory activities as well as significant antioxidant activity. These plants might become prominent natural therapeutic agents for the inhibition of alpha-glucosidase and oxidative stress, which play an important role in diabetic complications.
The present work is the first scientific report on alpha-glucosidase inhibitory activities and antioxidant capacities of H. cymosum constituents. This work is also the first scientific report on the phytochemical analysis of H. pandurifolium and the first inhibitory activity study of its total methanolic extract and its flavonoids on α-glucosidase and α-amylase.