Metabolites of Geum aleppicum and Sibbaldianthe bifurca: Diversity and α-Glucosidase Inhibitory Potential

α-Glucosidase inhibitors are essential in the treatment of diabetes mellitus. Plant-derived drugs are promising sources of new compounds with glucosidase-inhibiting ability. The Geum aleppicum Jacq. and Sibbaldianthe bifurca (L.) Kurtto & T.Erikss. herbs are used in many traditional medical systems to treat diabetes. In this study, metabolites of the G. aleppicum and S. bifurca herbs in active growth, flowering, and fruiting stages were investigated using high-performance liquid chromatography with photodiode array and electrospray ionization triple quadrupole mass spectrometric detection (HPLC-PDA-ESI-tQ-MS/MS). In total, 29 compounds in G. aleppicum and 41 components in S. bifurca were identified including carbohydrates, organic acids, benzoic and ellagic acid derivatives, ellagitannins, flavonoids, and triterpenoids. Gemin A, miquelianin, niga-ichigoside F1, and 3,4-dihydroxybenzoic acid 4-O-glucoside were the dominant compounds in the G. aleppicum herb, while guaiaverin, miquelianin, tellimagrandin II2, casuarictin, and glucose were prevailing compounds in the S. bifurca herb. On the basis of HPLC activity-based profiling of the G. aleppicum herb extract, the most pronounced inhibition of α-glucosidase was observed for gemin A and quercetin-3-O-glucuronide. The latter compound and quercetin-3-O-arabinoside demonstrated maximal inhibition of α-glucosidase in the S. bifurca herb extract. The obtained results confirm the prospects of using these plant compounds as possible sources of hypoglycemic nutraceuticals.


Introduction
Glycemic control is an essential therapy for patients with diabetes mellitus. Monitoring of postprandial hyperglycemia by inhibiting carbohydrate hydrolases (such as α-glucosidase) can decrease the risk of complications such as cardiovascular disease, neuropathy, nephropathy, and angiopathy [1,2]. Inhibition of α-glucosidase can slow down the digestion of complex carbohydrates and, thus, reduce the release of glucose into the blood [3]. The clinically used α-glucosidase inhibitors (acarbose, voglibose, and miglitol) have common side effects, such as diarrhea and flatulence, with corresponding liver dysfunction and abdominal pain [4,5]. Thus, the search for new possible α-glucosidase inhibitors with few side effects is an important goal.
Plant-derived drugs contain natural compounds of various structures and are promising sources of α-glucosidase inhibitors [6,7]. Previously, we screened the most common tea species of the Rosaceae family growing in Siberia. High inhibitory activity of α-glucosidase (IC 50 < 50 µg/mL) was a selection criterion and was used to identify promising plant species. Herb extracts of Geum aleppicum Jacq. and Sibbaldianthe bifurca (L.) Kurtto & T.Erikss. were the most active inhibitors of α-glucosidase according to their results [8]. G. aleppicum (Colurieae tribe) and S. bifurca (Potentilleae tribe) are closely related and belong to the Rosoideae subfamily [9].  (8 July) and fruiting phases (12 September). The species were authenticated by Prof. Tamara A. Aseeva (IGEB SB RAS, Ulan-Ude, Russia). Experimental samples of herb were collected in the morning (between 9 and 11 h). The herb samples were sealed in plastic bags and placed in a cooler with ice for transport to the laboratory. The collected herb samples were dried in a ventilated hood at a temperature of 24 • C to a moisture content of 7-9%. The herb samples were stored at 4 • C before analysis in a Plant Repository of the Institute of General and Experimental Biology. To obtain Metabolites 2023, 13, 689 3 of 16 herb samples with different growth periods, herbs from each collection date were pooled. After combining the herb samples from each collection date, three total samples of each growth period (active growth, May; flowering, July; fruiting, September) were obtained for both plants. No. GAL/Pro0522/12 (total sample, May), GAL/Pro0722/10 (total sample, July), GAL/Pro0922/12 (total sample, September) were the numbers of voucher specimens of G. aleppicum herb in the Plant Repository. No. SBI/Ros0522/10 (total sample, May), SBI/Ros0722/12 (total sample, July), SBI/Ros0922/12 (total sample, September) were the numbers of voucher specimens of S. bifurca herb in the Plant Repository. The samples were ground before analysis in an A11 basic analytical mill (IKA ® -WerkeGmbh & Co. KG, Staufen, Germany). Hereinafter, herb samples were sieved up to an average particle diameter of 0.5 mm using sieving machine ERL-M1 (Zernotekhnika, Moscow, Russia).

Plant Extracts Preparation
To prepare plant extracts, 10 g of dry and grounded herb of G. aleppicum and S. bifurca were extracted twice via stirring in a glass flask (200 mL) with 70% methanol (100 mL) with sonication at 40 • C using a Sapphire 2.8 bath (Sapphire Ltd., Moscow, Russia) for 30 min, ultrasound power 100 W, and frequency 35 kHz. The obtained methanolic extracts were combined, filtered through a cellulose filter and concentrated under reduced pressure until dryness. Obtained extracts were stored at 4 • C before use for HPLC analysis and α-glucosidase inhibiting activity study. The yields of total extracts of G. aleppicum were 3.3 g (May sample), 3.5 g (July sample), 3.2 g (September sample). The yields of total extracts of S. bifurca were 2.8 g (May sample), 3.2 g (July sample), 3.0 g (September sample). Before analysis, dry extract (100 mg) was dissolved in 10 mL 70% methanol using measuring flask (10 mL) and filtered through 0.22 µm syringe filters. To analyze the chemical profile of G. aleppicum and S. bifurca herb extracts, the previously described method used high-performance liquid chromatography with photodiode array detection and electrospray ionization triple quadrupole mass spectrometric detection (HPLC-PDA-ESI-tQ-MS/MS) was applied [8]. Chromatographic separation of compounds  was realized with liquid chromatograph LC-20 Prominence coupled with a photodiode  array detector, SPD-M30A (wavelength range of 200-600 nm), and a triple-quadrupole mass spectrometer, LCMS 8050 (all Shimadzu, Columbia, MD, USA). Column GLC Mastro C18 (2.1 × 150 mm, 3 µm) was used. Column temperature was 30 • C. The following eluents were used: A (0.4% formic acid in water) and B (0.4% formic acid in acetonitrile). The injection volume was 1 µL, and elution flow rate was 80 µL/min. The negative electrospray ionization was applied for mass spectrometric detection (-3 kV source voltage, range of m/z 100-1900, collision energy 5-40 eV). There were following temperature levels of ESI interface (300 • C), desolvation line (250 • C), and heat block (400 • C). There were following flow rates of nebulizing gas (N 2 , 3 L/min), heating gas (air, 10 L/min), collision-induced dissociation gas (Ar, 0.3 mL/min). The data were processed with LabSolution's workstation software (Shimadzu) equipped with the inner LC-MS library. The identification of metabolites was realized via the analysis of their retention time, ultraviolet, and mass-spectrometric data comparing the same criteria with the reference standards and literature data.

HPLC Activity-Based Profiling
To perform HPLC activity-based profiling, aliquots (100 µL) of G. aleppicum herb extract solution (10 mg/mL) and S. bifurca herb extract solution (10 mg/mL) were separated under analytical HPLC-PDA-ESI-tQ-MS/MS conditions as described in Section 2.4. The collection of eluates (40 µL) was performed every 30 s in 96-well plates. Then, the eluates were dried and redissolved in 10 µL of phosphate-buffered saline (PBS) followed by analysis as described previously [20]. α-Glucosidase from Saccharomyces cerevisiae was dissolved in PBS (pH 6.8), which contained bovine serum albumin (0.2%) up to 0.5 U/mL concentration, then 125 µL of PBS and 60 µL p-nitrophenyl-α-D-glucopyranoside (5 mM) were added. The incubation of the samples was realized at 37 • C for 5 min. Thereafter, 60 µL of α-glucosidase (0.4 U/mL) was added. Then, the samples were incubated at 37 • C for 15 min and 50 µL of sodium carbonate (200 mM) was added. Absorbance was determined at 400 nm. Epicatechin gallate was the reference compound. The activity of the microfractions as a percentage from the activity of the reference compound was displayed on the chromatogram as bars.

Statistical Analysis
Statistical analyses were performed using one-way analysis of variance, and the significance of the mean difference was determined using Duncan's multiple range test. Differences at p < 0.05 were considered statistically significant. The results are presented as the mean ± S.D. The linear regression analysis and generation of calibration graphs were conducted using Advanced Grapher 2.2 (Alentum Software, Inc., Ramat-Gan, Israel).

Metabolites of Geum aleppicum Herb: HPLC-PDA-ESI-tQ-MS/MS Profile
The Geum genus is characterized by the presence of numerous chemical classes of compounds with definite chromatographic behavior [29]. High-performance liquid chromatography with photodiode array and electrospray ionization triple quadrupole mass spectrometric detection (HPLC-PDA-ESI-tQ-MS/MS) was applied to separate compounds from the G. aleppicum herb extract. Analysis of chromatographic mobility, UV parameters, and mass spectral data and subsequent comparison of the obtained results with reference standards and/or literature information led to the identification of 29 compounds of various chemical classes (Figure 1a,b; Table 1).
were added. The incubation of the samples was realized at 37 °C for 5 min. Thereafter, 60 µL of α-glucosidase (0.4 U/mL) was added. Then, the samples were incubated at 37 °C for 15 min and 50 µL of sodium carbonate (200 mM) was added. Absorbance was determined at 400 nm. Epicatechin gallate was the reference compound. The activity of the microfractions as a percentage from the activity of the reference compound was displayed on the chromatogram as bars.

Statistical Analysis
Statistical analyses were performed using one-way analysis of variance, and the significance of the mean difference was determined using Duncan's multiple range test. Differences at p < 0.05 were considered statistically significant. The results are presented as the mean ± S.D. The linear regression analysis and generation of calibration graphs were conducted using Advanced Grapher 2.2 (Alentum Software, Inc., Ramat-Gan, Israel).

Metabolites of Geum aleppicum Herb: HPLC-PDA-ESI-tQ-MS/MS Profile
The Geum genus is characterized by the presence of numerous chemical classes of compounds with definite chromatographic behavior [29]. High-performance liquid chromatography with photodiode array and electrospray ionization triple quadrupole mass spectrometric detection (HPLC-PDA-ESI-tQ-MS/MS) was applied to separate compounds from the G. aleppicum herb extract. Analysis of chromatographic mobility, UV parameters, and mass spectral data and subsequent comparison of the obtained results with reference standards and/or literature information led to the identification of 29 compounds of various chemical classes (Figure 1a,b; Table 1).   Table 1. Two carbohydrates were discovered in G. aleppicum herb extract including saccharose (1) and glucose (2). Earlier, glucose was revealed in the herb and roots of G. urbanum and leaves of G. montanum [30]. Additionally, saccharose was detected in the herbs of G. montanum [30] and G. rivale [31] and roots of G. iranicum [32].

Organic Acids
The presence of malic (3) and citric (4) acids was noted for the G. aleppicum herb. Previously, malic acid was found in the aerial parts of G. reptans, G. montanum, G. bulgaricum, and G. hybrid [33]. There are no data on the detection of citric acid in other species of the genus Geum.

Benzoic Acid Derivatives
Three benzoic acid derivatives were determined in the G. aleppicum herb. 3,4-Dihydroxybenzoic acid 4-O-glucoside (6) and 3,4,5-trihydroxybenzaldehyde (7) were identified by comparing these with reference standards. The mass spectrometric analysis of compound 9 demonstrated the loss of a hexose fragment (162 Da) and the remaining fragment with m/z 121 corresponding to a benzoic acid moiety. The assumed structure of compound 9 was found to be a benzoic acid, O-hexoside. 3,4,5-Trihydroxybenzaldehyde was revealed earlier in G. japonicum [34], while 3,4-dihydroxybenzoic acid 4-O-glucoside was found in Geum for the first time.

Carbohydrates and Organic Acids
The carbohydrates [saccharose (1) and glucose (2)] and two organic acids [malic (3) and citric (4)] were discovered in the S. bifurca herb upon comparison of t R , UV, and mass spectra data with reference standards. Previously, these compounds were not detected in the Sibbaldianthe genus.

Quantitative Content and Seasonal Variation of Profile of Geum aleppicum and Sibbaldianthe bifurca Herb
To identify possible patterns in the chemical profile of the G. aleppicum herb and S. bifurca herb, these species were collected and investigated at different growth phases: active growth (May), flowering (July), and fruiting (September). The maximum content of the majority of compounds in both species was observed during the flowering period. In particular, gemin A, miquelianin (quercetin-3-O-glucuronide), niga-ichigoside F1, 3,4dihydroxybenzoic acid 4-O-glucoside, and glucose were the dominant compounds of the G. aleppicum herb. The contents of different ellagitannins increased towards the flowering phase and then decreased upon fruiting. Thus, the content of the dominant ellagitannin, gemin A, in the active growth phase (10.18 mg/g) increased more than five times by the flowering phase (53.26 mg/g), and then, it gradually decreased in the fruiting stage (42.11 mg/g). A similar trend was observed for both casuariin (1.26 mg/g → 2.57 mg/g → 2.03 mg/g) and pedunculagin (trace → 0.26 mg/g → trace). In contrast, the content of ellagic acid was the maximum in the fruiting phase (5.63 mg/g), which occurred possibly because ellagic acid was released during the hydrolysis of ellagitannins [54]. Flavonoids, both derivatives of quercetin and kaempferol, accumulated the most in the flowering phase of the G. aleppicum herb. The content of the prevalent quercetin derivative, miquelianin, in the active growth phase increased from 5.20 mg/g to 26.83 mg/g in the flowering period. One possible reason for the maximum accumulation of flavonoids in the G. aleppicum herb in the flowering phase may be the high UV radiation and air temperature. Previously, similar accumulations of flavonols at high growth temperatures were observed in other representatives of the Rosaceae family [21,55].
Guaiaverin (quercetin-3-O-arabinoside), miquelianin, tellimagrandin II 2 , casuarictin, and glucose were the dominant compounds in the S. bifurca herb. The accumulation of the dominant flavonoids, guaiaverin and miquelianin, was also observed during the growth phase (21.59 and 19.62 mg/g, respectively). The concentrations of dominant ellagitannins, tellimagrandin II 2 and casuarictin, increased until the flowering stage and then consistently decreased until the fruiting period (3.62 mg/g → 7.83 mg/g → 5.16 mg/g and 1.60 mg/g → 5.28 mg/g → 4.16 mg/g, respectively). The maximum content of gallotannins was observed in samples during the growth phase, followed by a decrease in the flowering and fruiting phase samples. This can likely be explained by the fact that galloyl hexoses are precursors of complex hydrolysable tannins, the biosynthesis of which is carried out via oxidative binding of galloyl groups [56,57]. Thus, the maximum accumulation of dominant ellagitannins and flavonoids in the G. aleppicum and S. bifurca herbs under Siberian conditions was observed during the flowering phase in July.

Chemotaxonomic Significance of G. aleppicum and S. bifurca Metabolites
As a result of the chromatographic investigation of the G. allepicum and S. bifurca herbs, 70 metabolites of different chemical classes were identified. To select compounds of chemotaxonomic significance for these species, particular attention should be paid to 2-pyrone-4,6-dicarboxylic acid, hydrolysable ellagitannins, and flavonols.
Currently, S. bifurca belongs to the Potentilleae tribe [58], and G. aleppicum belongs to the Colluria tribe, although earlier, experts attributed it to the tribe Dryadeae [9]. Both species are closely related and belong to the Rosoideae subfamily [9]. 2-Pyrone-4,6-dicarboxylic acid is a breakdown product of phenolic compounds and is a taxonomic marker of the Rosoideae subfamily [59]. Both the G. aleppicum and S. bifurca herbs contained 2-pyrone-4,6-dicarboxylic acid, which confirmed the results reported by Wilkes et al. on the presence of this compound in representatives of the Rosoideae subfamily [59].
Ellagitannins have a wide distribution in the Rosaceae family [60], while oligomeric hydrolysable tannins are limited to the Rosoideae subfamily [16]. According to this theory, most species of the Rosoideae subfamily contain one or two oligomers that are used as chemotaxonomic markers. In the studied plant objects, dimer gemin A for the G. aleppicum herb and dimers rugosin E and agrimoniin for the S. bifurca herb may have chemotaxonomic significance.
Flavonoids have also been proposed as a chemotaxonomic marker of the Rosaceae family [61]. Derivatives of kaempferol and quercetin were revealed in both the G. aleppicum and S. bifurca herbs. However, these species did not contain any specific flavonoids that would allow us to discuss chemosystematic markers. Thus, the exact chemosystematic significance of flavonols in the Rosoideae subfamily is not definitively due to their wide presence in Rosaceae in general.

α-Glucosidase Inhibiting Activity of Geum aleppicum and Sibbaldianthe bifurca Herb Extract: HPLC Activity-Based Profiling
To reveal the components of G. aleppicum and S. bifurca herb extracts with α-glucosidaseinhibiting properties, the HPLC activity-based profiling method was used. This is a highly multipurpose strategy to miniaturize and accelerate identification of active substances in analyzed extracts by analytical HPLC [62,63]. HPLC activity-based profiling is performed via post-column collection of microfractions in a plate after a certain period of time, their subsequent drying, and the addition of reagents for biological evaluation [64,65]. The activity of the microfractions as a percentage of the activity of the reference compound is displayed on the chromatogram as bars (Figure 1c). Epicatechin gallate was chosen as the reference compound due to its high ability to inhibit α-glucosidase with IC 50 value of 4.03 ± 0.01 µg/mL [66]. Epicatechin gallate isolated from Rhodiola crenulata roots inhibited α-glucosidase with IC 50 0.71 ± 0.01 [67].
As a result of HPLC activity-based profiling of the G. aleppicum herb extract, inhibition of α-glucosidase by 3,4-dihydroxybenzoic acid 4-O-glucoside, gemin A, quercetin-3-O-glucuronide, niga-ichigoside F1, and rosamultin was found. The most pronounced inhibition of α-glucosidase was observed for ellagitannin gemin A and flavonol quercetin-3-O-glucuronide. The same procedure for the S. bifurca herb extract revealed inhibition of α-glucosidase by tetragalloyl hexose, quercetin 3-O-glucuronide, quercetin-7-O-glucoside, and quercetin-3-O-arabinoside (Figure 2c). Maximal inhibition of α-glucosidase was found for the flavonols quercetin-3-O-glucuronide and quercetin-3-O-arabinoside. Previously, it was suggested that the mechanism of the inhibitory activity of ellagitannins against αglucosidase involves their binding to proteins, followed by changes in the conformation of the enzyme and a decrease in its activity [68,69]. In turn, flavonols bind to glucosidase with high affinity through hydrogen bonds and van der Waals forces, and then, these complexes lead to conformational changes in α-glucosidase [70,71]. Thus, microfractionation of the G. aleppicum and S. bifurca herb extracts allowed direct evaluation of the α-glucosidase inhibitory activity of the components in their biological matrices. The obtained results can serve as a basis for using these plant compounds as possible sources for hypoglycemic nutraceutical production.

Conclusions
The Geum aleppicum and Sibbaldianthe bifurca herbs are used in traditional medicine as antidiabetic remedies. In an attempt to identify compounds with antidiabetic potential, these closely related plant species were first characterized via HPLC-PDA-ESI-tQ-MS/MS and, as a result, data on 70 compounds were obtained. Carbohydrates, organic acids, derivatives of benzoic and ellagic acids, ellagitannins, flavonoids and triterpenoids were identified in both plant species. Then, HPLC activity-based profiling was applied, which allowed one to miniaturize and accelerate the identification of active substances in the analyzed extracts. Gemin A, quercetin-3-O-glucuronide and quercetin-3-O-arabinoside showed the most pronounced results in terms of α-glucosidase inhibition. In this study, the G. aleppicum and S. bifurca herbs were shown to be natural sources of metabolites with α-glucosidase-inhibiting properties. Further in vivo investigations of these plant extracts are necessary for the wide introduction of new biologically active agents into therapeutic practice for the treatment of diabetes mellitus.