Analytical Investigation of Active Compound Contents of Panax ginseng C.A. Meyer and Ginkgo biloba L. Supplements Fortified with Apitherapy Products

Extracts of therapeutic plants of nature, such as Ginkgo biloba L. and Panax ginseng C.A. Meyer (P.ginseng) are highly requested. Recently admixtures of these extracts with apitherapy products are also available. In this study, as research materials, P.ginseng, and Ginkgo biloba L. supplements in the form of tablets, paste, and liquid extracts prepared as a mixture with bee products were compared in terms of their ginsenosides, phenolics, terpene lactones, and antioxidant capacities. Within this study, it was aimed to clarify the active ingredients P.ginseng extracts, Ginkgo biloba L. extracts, and their mixtures with bee products by developing advanced and novel analytical methods. Additionally, it was purposed to unveil the contribution of bee products to phytotherapy supplements of P.ginseng and Ginkgo biloba L. on the basis of secondary metabolites. Ultrasound-assisted hydroalcoholic extraction was applied at sample preparation and supernatants were analyzed at UPLC-MS/MS and UV-Visible spectrophotometer. The antioxidant activities were determined using the 2,2diphenyl-1-picrylhydrazyl (DPPH) method. As a result, total ginsenoside contents of the products per serving were ranged between 9.7 mg to 150.2 mg. Terpene lactones of the ginkgo products were ranged between 0.1 mg to 9.5 mg per serving. The antioxidant activities of all products were determined between 13% and 92%. Quercetin, rutin hydrate, CAPE, kaempferol, galangin, chrysin, gallic acid, pinocembrin, and isorhamnetin were found as the predominant phenolics. Due to their higher antioxidant capacities, superior phenolic concentrations, and diversities, it was enlightened that apitherapy products can augment the phytotherapy efficacies of P.ginseng and Ginkgo biloba L. supplements. ARTICLE HISTORY Received: Nov. 27, 2021 Revised: Apr. 21, 2021 Accepted: May 06, 2021


INTRODUCTION
With the development of modern medicines and revisions to the methods of treatment, phytotherapy which has been thrown into the background for a while has become popular again due to the difficulties in the treatment of chronic diseases and increased side effects from some modern therapies. Plant extracts of therapeutic nature, such as Panax ginseng C.A. Meyer and Ginkgo biloba L., dating back to ancient times, and mixtures containing these extracts are being activities are attributed to the synergistic activity of the flavonoids and the terpene trilactones (Smith & Luo, 2004;Stiker et al., 2000;van Beek, 2002). The flavonoids are composed of a large group of polyphenols and contain flavonol glycosides, flavones, flavan-3-ols, acylated flavonol glycosides and proanthocyanidins. Among these molecules, the flavonol glycosides are the most abundant group. Furthermore, numerous flavonol glycosides are sometimes found as their aglycones such as kaempferol, quercetin and isorhamnetin that are usually present in the leaves with small amounts (van Beek, 2002). The ginkgo terpene trilactones: bilobalide (sesquiterpene) and ginkgolides (diterpene); ginkgolide B, ginkgolide A, ginkgolide J and ginkgolide C are unique to Ginkgo biloba L. and they show broad spectrum of pharmacological activities (de Jager et al., 2006). Commercial Ginkgo biloba L. products are usually commercialized as mixtures, based on the content of terpene lactones and flavonol glycosides (van Beek & Montoro, 2009). A standardized leaf extract contains 22-27% of flavone glycosides and 5-7% of terpene lactones. Ginkgolide A, C, B with 2.8-3.4% and bilobalide with 2.6-3.2% along with less than 5 ppm ginkgolic acid are expected to be observed at standardized leaf extract (Kiefer, 2004). The most considered standardized leaf extract in studies is EGb761 (Mahadevan & Park, 2008). The standard dose of EGb761 is 120 mg (∼1.7 mg/kg) once or twice daily for clinical puposed usage; hence, a standard dose will contain ∼3-4 mg ginkgolides A, B, and C, 3-4 mg bilobalide, and 29 mg flavonoids (Nash & Shah, 2015). Quality control of Ginkgo biloba L. extracts and leaves can be routinely carried out by reversedphase HPLC (RP-HPLC) with evaporative light scattering detection (ELSD) (Dubber & Kanfer, 2006;Kaur et al., 2009), refractive index (RI) (Wang & Ju, 2000), diode array detector (DAD) (Tang et al., 2010) and mass spectrometry detection (Wang et al., 2017). As an orthogonal technique gas chromatography (GC) system equipped with a flame ionization detector (FID) or mass spectrometry (MS) detection (Czigle et al., 2019) is also used. At sample preparation steps; cold and hot extraction, pressurized solvent extraction and ultrasoundassisted extraction is typically applied to achieve the simultaneous analysis of terpene trilactones and flavonol aglycones (Kaur et al., 2009;Liu et al., 2015).
Apitherapy has been gaining great attention nowadays due to its superior health effects. Bee products like pollen, royal jelly, honey, and especially propolis are rich in flavonoids, phenolic acids, minerals, and diverse anti-inflammatory and immune booster substances. The antioxidant capacities, phenolic acids, and flavonoid contents of P. ginseng and Ginkgo biloba L. extracts mixed with bee products are expected to be increased significantly. This research aimed to determine and compare the bio-active phytochemical contents, antioxidant capacities, and phenolic profiles of P. ginseng and Ginkgo biloba L. extract products as well as their preparations in mixture with bee products applying developed methodologies. In addition to many studies that investigated the health benefits and pharmacological properties of bee products, our research addedly focused on the investigation of active substance diversities at commercially available supplements and tried to find answers to how apitherapy products may change the phenolic and antioxidant properties of P. ginseng and Ginkgo biloba L. extracts that have been used in traditional medicine for years. Moreover, a high-resolution phenolic profile determination method with a comprehensive compound library and using mass-based detection was targeted to be developed. Thanks to new methods in terms of ginsenoside, terpene lactone, and phenolic monitoring, we envisage that this research will contribute a lot to the literature for advanced phytotherapy research.

Sample Collection
Products containing P. ginseng extract (n=12) and samples containing Ginkgo biloba L. extract (n=9) were purchased from the pharmacies and herbalists. Two of the samples that contain Ginkgo biloba L. extracts and seven of the samples containing P. ginseng extract were in form of mixtures with alternative bee products. Information of the samples regarding their serving sizes and their forms are given in Table 1.

Terpene Lactone Analysis
100 mg of Ginkgo biloba L. product was weighed in a 50 ml falcon tube and 20 ml of 90% MeOH solution was added. The sample was extracted using an ultrasonic bath (Sonorex ® RK 154 BH, Bandelin Electronic ® GmbH & Co.KG, Berlin, Germany) at 50°C for 15 min. It was then centrifugated at 3000 rpm for 10 min. (Hettich ® , Rotina 35R Type, Tuttlingen, Germany) and the supernatant was transferred to a new falcon tube. This extraction step was repeated twice more. Transferred supernatants were combined (~50 ml) vortexed-mixed and completed to a volume of 50 ml with deionized water. Prepared samples were transferred to glass vial by diluting 1:10 (w/v) with deionized water. Calibration standards at 0.5, 1.0, 2.0, and 4.0 μg/ml were also prepared by diluting the standard mixture solution of terpene lactones using deionized water. All standards and samples were kept at 4°C until analysis. The chromatographic analysis was performed on Waters ® ACQUITY UPLC (Waters ® , Milford, MA, USA) system equipped with a binary solvent delivery system and autosampler. The chromatographic separation was achieved on Waters ® ACQUITY UPLC HSS (High Strength Silica) T3 column (2.1 mm x 75 mm, 1.8 µm). The mobile phase consists of solution A (water containing 0.005% acetic acid) and solution B (methanol containing 0.005% acetic acid). The gradient elution was optimized as follows: 10% B (0-0.5 min.), 10-70% B (0.5-7 min.), 70-10% B (7-7.01 min.), 10% B (7.01-8 min.). The flow rate was 0.35 ml/min. The column and autosampler temperatures were maintained at 40°C and 4°C, respectively. The injection volume was 10 µL. The detection was done via Waters ® Xevo TQ tandem quadrupole mass spectrometer (Micromass ® MS Technologies, Manchester, UK) with electrospray ionization (ESI) interface in multiple reaction monitoring (MRM) and positive ionization mode. The needle capillary voltage was set at 3 kV. The flow rate and temperature of desolvation gas were 850 L/h and 450 °C, respectively. The flow rate of cone gas was 50 L/h and source temperature operated at 150°C. These conditions were optimized by analyzing the reference compounds and comparing the signal-to-noise ratio (S/N) for each analyte. All ESI and mass spectrometer conditions were optimized individually for each target compound and listed in Table 2. Dwell times were automatically set by the software. Data acquisitions and quantifications were performed using Waters ® Mass-Lynx software with the Target Lynx program. 2.4. Ginsenoside Analysis 500 mg of P. ginseng product was weighed and 15 ml of 90% MeOH was added. Ultrasoundassisted extraction was performed at 50°C for 15 min. The sample was centrifugated at 3000 rpm for 10 min. After the transfer of supernatant to a new falcon tube, the extraction process was repeated two more times and all supernatants were combined (~45 ml) after each extraction step. Extracts were vortexed-mixed and completed to volume of 50 ml. Final prepared samples were transferred to glass vial by diluting 1:10 (w/v) with ultra-pure water. Calibration standards at 0.5, 1.0, 2.0, and 4.0 μg/ml concentrations were prepared by appropriate dilution of the standard mix. solutions of ginsenosides. Waters ® ACQUITY UPLC system equipped with a binary solvent delivery system and autosampler was used to perform the separation. A chromatographic resolution was crucial and this was achieved using a Waters ® Acquity BEH (Bridge Ethylene Hybrid) C18 column (2.1 mm x 50 mm, 1.7 µm). The liquid chromatography phases consisted of water as mobile phase A and acetonitrile (ACN) as mobile phase B without any modifiers. The gradient elution was optimized as follows: 15% B (0-1 min.), 15-35% B (1-9 min.), 35-75% B (9-12 min.), 75-15% B (12-12.01 min.), 15% B (12.01-14 min.). The flow rate was 0.4 ml/min. The column oven set to 40°C and autosampler temperature was maintained at 4°C. The injection volume was 10 µl. The high accuracy mass spectrometric data were recorded on Waters ® Xevo TQ tandem quadrupole mass spectrometer with ESI source in MRM mode. The optimized MS parameters were as follows: the capillary voltage: 3 kV, flow rate, and temperature of desolvation gas: 800 L/h and 450 °C respectively, flow rate of cone gas: 50 L/h, and source temperature: 150°C. Ionization states and MS parameters were optimized individually for each target compound and they were listed in Table 3. Dwell times were automatically set by the Mass-Lynx software. Data acquisition was controlled by the Target Lynx software from Waters ® . . It was centrifugated at 3000 rpm for 5 min. and the supernatant was separated. Extraction was repeated and the total volume of supernatants was completed to 100 ml with EtOH. Samples were diluted to appropriate volume depending on the phenolic concentration of the sample to be able to quantify between the calibration range with deionized water and filtered to glass vials using a 0.45 µm PVDF filter (Interlab ® , Istanbul, Turkey) prior to chromatographic analysis. Calibration standards at 0.05, 0.1, 0.25, 0.5, 1.0 and 2.0 μg/ml were prepared from stock standard mixture by applying serial dilutions. The chromatographic system was Waters ® ACQUITY UPLC, which consisted of a binary solvent delivery system and autosampler. The separation was performed on Waters ® CORTECS T3 column (1,6 µm 2,1 x 150 mm) using a gradient elution of (A) water containing 0. ). The column and autosampler tray temperatures were 30°C and 10°C, respectively. The effluent from the LC outlet was directed into the ionization electrospray source of Waters ® Xevo TQ tandem quadrupole mass spectrometer after 1.0 min. delay using the embedded valve on the device. The ion source and desolvation temperature were held at 150 °C and 450 °C, respectively. The flow rate of desolvation and cone gas were optimized at 850 L/h and 50 L/h respectively. The needle capillary voltage was determined at 2 kV for ideal responses. MRM mode was employed for analysis. Peak areas for all components were automatically integrated using Mass-Lynx software with the Target Lynx program (Waters ® ). All ESI and MS parameters were optimized individually for each target compound and listed in Table 4.

Free Radical Scavenging Activity (Antioxidant Capacity Assay)
The percentage of antioxidant activity (AA%) of each sample was assessed by DPPH free radical scavenging assay. Hydroalcoholic extracts prepared for phenolic profiling analysis were used as representative samples and reacted with the reagent of DPPH radical. For this, 2.3 ml of sample extract was mixed with 0.3 ml of DPPH solution (0.1 mmol/L) and it was incubated in the dark for 15 min. The changes in color (from deep violet to light yellow) were observed and the absorbance of sample solutions was read at 517 nm against ethanol/DPPH solution as a reagent blank using a UV-VIS spectrophotometer (Thermo Fisher Scientific ® Inc., Electron Evolution 300 UV-VIS Spectrophotometer, Waltham, MA, USA).

RESULTS and DISCUSSION
3.1. Ginsenosides 12 commercial samples were gathered and analyzed using the developed Ultra Performance Liquid Chromatography Multiple Reaction Monitoring Mass Spectrometry (UPLC/MRM-MS) method for ginsenoside contents. Thanks to our high-resolution chromatographic separation method, as shown in Figure 1, ginsenoside molecules were resolved and analyzed in high specificity and in short analysis time. Only the ginsenoside Rb2 and ginsenoside Rb3 components were co-eluted and their concentrations are reported as total amount.
The majority of these samples are in mixture with different bee products; A1: pollen and honey, A2: pollen, honey, propolis, and royal jelly, A3: royal jelly and propolis, A4: pollen, royal jelly, and propolis, A5: royal jelly A8: propolis and honey A12: propolis, royal jelly, and honey. All quantified results were converted to their corresponding percentage for each ginsenoside component in the P. ginseng sample. The analysis results are given in Table 5.  Total ginsenoside percentage values in the products were ranged between 0.14% and 11.80%. Label fraud has been detected at A2 coded sample, since the ginsenoside content could not be quantified as a result of the analysis. Approximate percentages of total ginsenoside contents were similar at investigated samples except for A6 and A12 coded samples which have considerably higher quantities for daily intake. In addition to the total ginsenoside content, the compound diversity of the product and amount of each separate ginsenoside molecule is also a very important point of evaluation regarding the raw material quality. Considering that the therapeutic effects of each ginsenoside compound may vary, the diversity of ginsenoside types in P. ginseng extracts is a matter to be considered. In the samples coded A9 and A10, the percentage distribution of the compounds can be seen as homogeneous, while Ginsenoside Rd was detected as the predominant substance over the other components in 6 of the analyzed samples (A1, A3, A4, A7, A8, A10). Ginsenoside Re, Ginsenoside Rb2, and Rb3 were other high abundance ginsenoside species at A1, A3 and A4 coded samples alongside a high level of Ginsenoside Rd. While the highest content of ginsenoside in A5 coded sample was determined as Rg2, ginsenoside Ra1 with Rg2 in sample coded A6 stands out as the most dominant species. In the sample coded A8, in addition to the ginsenoside Rd, Rb2, Rb3 components, unlike other samples, high levels of ginsenoside Rb1 and ginsenoside Rc were detected. In the A11 coded sample, ginsenoside Rg1 and ginsenoside Re components were in high amount, while in the A12 coded sample, ginsenoside Rc and ginsenoside Rb1 molecules were detected at high levels. As can be seen from the results, the ginsenoside types in the samples differ greatly due to factors such as extraction technique and ginseng types and/or origins. Results regarding the ginsenoside compound diversity are given in Table 6. Whether the products include apitherapy additives or not, they were observed that P. ginseng contents were appropriate for recommended daily consumption, and except for one sample, all products mostly meet the values stated on the labels. Daily intake amounts of the products were ranged between 13.3 mg/day and 150.2 mg/day. The recommended daily intake amounts were at the highest levels in two samples containing P. ginseng which are mixed with bee products (A3: 87.47 mg and A12: 150.20 mg).

Terpene Lactones
Ginkgo biloba L. contains mainly two active components; the terpene lactones and the ginkgo flavon glycosides, which together have been proven to be responsible for the polyvalent activities of Ginkgo biloba L. containing preparations. The nine commercial samples were analyzed by the developed UPLC/MS-MS method to quantify the levels of terpene lactones in these samples. Figure 2 shows typical chromatograms for the analysis of 5 of terpene lactone standards using the UPLC/MS-MS method under optimized instrument conditions. According to the analysis results, the B4 coded sample provided the highest free form terpene lactone content (16.3 mg) for daily intake. The samples with the codes of B6 and B7 provided a higher amount of terpene lactone than other products and followed the B4 coded sample in terms of the total quantity. Only two of the analyzed products (B2, B6) were bee product mixtures. Among these products, the B6 coded sample gave the second-highest daily intake value among the analyzed products in terms of terpene lactone values. The results in Table 7 are shown that daily intake amounts of terpene lactones (mg) were ranged between 16.30 mg and 0.10 mg. While the ginkgolide B and C are found in all samples, the ginkgolide A was not quantified at only B2 coded sample. Ginkgolide J and the bilobalide have not been observed in B1, B2 and B3 coded samples. Ginkgolide J was also not quantified in the sample of B6. Analysis results of the investigation for the diversity of terpene lactones in the products are given in Table 8 The homogeneous distribution could only be detected in only a few samples, similar to the results acquired at P. ginseng analyzes. Samples coded B4, B5, B7, B8, B9 became prominent compared to other samples due to the reason that they contain all analyzed terpene lactone parameters and contain four terpene lactone species except ginkgolide J at high and similar percentages.

Phenolic profiles and antioxidant capacities
Previous analysis results have been shown that the forms of P. ginseng and Ginkgo biloba L. extracts mixed with apitherapy products have equivalent concentrations of the active ingredient (ginsenosides and terpene lactones) compared to non-mixed products on the market. The concentration of total and individual phenolic contents is given in Table 9. It can be observed that the content of the 43 components varies greatly among the different samples. Quercetin, rutin hydrate, CAPE, kaempferol, galangin, chrysin, gallic acid, pinocembrin, and isorhamnetin were monitored as predominant phenolic substances as a result of the phenolic profiling analysis. As shown in Figure 3, all these 43 components are analyzed in about 35 min. using novel UPLC/MS-MS method. Figure 4 shows the comparison of the phenolic contents of the bee product supplemented P. ginseng samples (A1, A2, A3, A4, A5, A8, A11, and A12) and pure P. ginseng extracts (A6, A7, A9, and A10).  The same effect can also be seen in Figure 5 representing the Ginkgo biloba L. products containing bee products and comparison with those of not containing.  Table 9a and Table 9b also show the antioxidant activity results of the samples which were between 12% and 23%. According to these results, Ginkgo biloba L. and P. ginseng extracts supported by bee products such as pollen, propolis, and honey also provide high antioxidant capacity besides their phenolic diversities. Getting systematic knowledge about phenolic compounds and mostly flavonoids in the natural products are highly important for the phytotherapy product development strategies but also for the assessment of the therapeutic effects. Therefore, we thought, performing phenolic profiling investigations instead of total phenolic content analysis is much amenable for the samples. It was found that 4hydroxybenzoic acid, transcinnamic acid, protocatechuic acid, p-Coumaric acid, homogentisic acid, gallic acid, caffeic acid, ferulic acid, 3,4-dimethoxy cinnamic acid, chrysin, pinocembrin, apigenin, galangin, naringenin, pinobanksin, CAPE, catechin, quercetin, isorhamnetin, chlorogenic acid, rosmarinic acid, and methylsyringate were the abundant phenolic derivates, which especially arises with the addition of bee products. Table 9a. Total antioxidant activities and the phenolic profiles of the P. ginseng products.

Analyte
Sample Results (mg/kg)     According to results, propolis and pollen were mainly contributed to the phenolic profiles more than the royal jelly and honey supports.
Previous research based on the identification of the secondary metabolites of P. ginseng and Ginkgo biloba L. supplements includes tedious sample preparation steps, insensitive detection methods, and identifications based on passive detector acquisitions such as refractive index detection, UV detection, and so forth (Wang & Ju, 2000;Li et al., 2005;Tang et al., 2010). UPLC-ESI-MS/MS and developed methods enabled to identify and quantify the compounds of interest more accurately, specifically owing to their unique three-dimensional data (retention time, abundance, and mass to charge ratio). In addition to this, comprehensive phenolic profiling analysis of these herbal extracts was not achieved until this novel method was implemented. In this study, ultrasound-assisted solid-liquid extracts of the samples were analyzed directly in the UPLC-ESI-MS/MS system for three different phytochemical classes (terpene lactones, ginsenosides, and phenolics) without any additional purification, concentration, or derivatization. By using a single extract obtained from these samples, antioxidant and active ingredient measurements could be made simultaneously. In this way, the preliminary process has been simplified.

CONCLUSION
The active ingredient contents of P. ginseng and Ginkgo biloba L. supplements available in the market and the forms that supplemented with bee products were enlightened. Contents of the supplied samples were evaluated in terms of potential phytotherapy efficacy by means of research carried out using novel developed analytical methods. Within the scope of the study, phenolic substances analysis method including 43 compounds was developed using Ultra-Performance Liquid Chromatography, electrospray ionization (ESI) Tandem Mass Spectrometry (UPLC-ESI-MS/MS) system. Thus, sensitive and high-resolution profiling study was achieved. In addition, analytical methods were developed based on MS detection for the monitoring of ginkgolide, bilobalide, and ginsenoside active compounds. At our investigation, it has been determined that P. ginseng and Ginkgo biloba L. supplements available in the market predominantly contain sufficient amounts of active ingredients, as well as the analyzed amounts and compositions, were variable. The novel and practical analytical methods proved that the health benefits of the aforementioned products will increase owing to the fact that an increase in the phytochemical molecule diversity and antioxidant capacities if the content of the raw extract is supported with alternative apitherapy products.