Abstract
Ginkgo biloba extract (GBE) is an essential herbal supplement for diverse health perspectives. The present study aims to improve the oral bioavailability of GBE flavonoids. We formulated the GBE flavonoids-cyclodextrin (GBE-CD) complex by reacting GBE flavonoids with naringinase in the presence of γ-cyclodextrin (CD) using a patented method. The results showed an enhanced bioavailability of quercetin (QCT), kaempferol (KAE), and isorhamnetin (ISR) after β-glucuronidase catalysis of flavonoid glucuronide conjugates in plasma after GBE-CD (25 μmol/kg as QCT equivalence) administration, that was 5.4, 6.1, and 10.4-folds, respectively, compared to GBE administration (25 μmol/kg as QCT equivalence) in Sprague–Dawley rats. In summary, the GBE inclusion into the γ-CD cavity is found suitable to improve oral bioavailability and could be applicable to several health benefits and disease prevention.
Introduction
Ginkgo biloba extract (GBE) is one of the most commonly used herbal remedies and food supplements worldwide. 1 It has anti-neurological,2-7 anti-hepatic,8,9 anti-oxidant,10,11 and anti-cancer properties.12-14 The standardized extract EGb 761w (EGb) contains the following active constituents: ginkgo-flavonoids (> 24%), such as quercetin (QCT), isorhamnetin (ISR), and kaempferol (KAE), and terpene lactones (> 6%), such as ginkgolides and bilobalide.15,16
Intake of QCT, KAE, and ISR derivatives has various improvement effects against cerebrovascular, neurological, cardiovascular, oxidative, and hepatic disorders.17-23 However, the low bioavailability of GBE flavonoids (QCT, KAE, and ISR derivatives) limits these effects. Therefore, some groups have improved the bioavailability of flavonoids using liposome treatments, such as phospholipid treatment, solid dispersions, etc.24-26 Whereas, liposome phospholipid complexes have limitations such as instability and degradation due to hydrolysis and oxidation.27-29 Furthermore, solid dispersion has challenges related to physical and chemical stability, preparation method, and manufacturing processes. 30
Dos Santos Lima et al. 31 reported that an inclusion complex with cyclodextrin enhanced the bioavailability of some flavonoid compounds such as Taxifoline/γ-cyclodextrin (CD), Baicalin/β-CD, and Quercetin/Hydroxypropyl (HP)-β-CD, etc. Furthermore, Pandey 32 reported that CD-conjugated nanoparticles have higher drug loading capacity and stability than other carriers such as liposomes and microparticles.
GBE contains more than 30 flavonoids. Nevertheless, only glucose and rhamnose can be found as sugar molecules of mono-, di-, and tri-glycosides in the different binding patterns, and the main flavonol glycosides in GBE are derivatives of QCT, KAE, and ISR.2,33 Chen et al. 34 reported that the chromatographic fingerprint analysis of Ginkgo biloba products showed that most products, contained quercetin-rutinoside (RTN) as the largest peak in a total of 21 major peaks of flavonoids using high-performance liquid chromatography (HPLC). Moreover, Beck and Stengel 33 showed that Ginkgo biloba leaves contained flavonoid-rutinosides (FLRs) such as RTN, kaempferol-rutinoside (KAER), isorhamnetin-rutinoside (ISRR), and mearnsetin-/laricitrin-rutinoside using liquid chromatography with tandem mass spectrometry (LC-MS-MS), and reported the majority of flavonoids glycosides in GBE were FLRs.
The bioavailability of flavonoid-monglucosides (FMGs) such as isoquercitrin (IQC), naringenin-glucoside (NAGG), and hesperetin (HPT)-glucoside (HPTG) is several times higher than that of FLRs such as rutin (RTN), narirutin (NAR), and HSP in the human subjects.35-37 This is because lactase-phlorizin hydrolase in the intestine was shown to efficiently digest FMGs such as IQC, NAGG, and HPTG, to aglycones such as QCT, NAG, and HPT, and are easily absorbed through the small intestine. In contrast, FLRs are mainly absorbed in the large intestine after hydrolysis by colon microflora.38,39 This means that FMGs in GBE flavonoids are closely related to systemic absorption, that results in their enhanced physiological effects.
However, the industrial production of FMGs from FLRs is difficult because (1) the acid hydrolysis of FLRs easily results in the over-hydrolysis of FMGs to produce flavonoid aglycones, (2) the hydrolysis of FLR(HSP) using rhamnosidase as naringinase limits the ability to produce FMG (HPTG) since FLR (HSP) is scarcely soluble, and (3) the production of FLRs (HSP, Naringin [NARN]) by naringinase produces aglycones (HPT, NAG).40,41
Recently, we showed the AUC0-24 values after the administration of FMGs-CD such as IQC-γ-CD- and HPTG-β-CD-inclusive complexes, industrially produced using FLRs, such as RTN or HSP with naringinase in the presence of CD following a patented method.42-44 The results successfully revealed bioavailability of namely 10 to 100 folds higher than that after the administration of QCT- or HSP-dextrin physical mixtures in healthy adult males.45,46 Studies also revealed the structure of inclusive complexes47,48 and their safeties.49,50 Therefore, we prepared GBE FMGs-CD complex (GBE-CD) by reacting GBE FLRs with naringinase in the presence of γ-CD using a patented method,42-44 to investigate the pharmacological activity of flavonoids. We compared the pharmacokinetic parameters of flavonoids (QCT, ISR, and KAE) after β-glucuronidase catalysis of QCT, ISR, and KAE glucuronide conjugate in the plasma using HPLC after administering GBE-CD and GBE (25 μmol/kg as QCT equivalence) in Sprague–Dawley (SD) rats.
Results and Discussion
Production of Ginkgo biloba Extract-γ-CD Complex and HPLC Analysis
The area rates (IQC from RTN, ISRG from ISRR, and KAEG from KAER, see HPLC analysis; area rate [%]) by treating flavonoids of GBE (26.1% flavones; China Material Company) with naringinase in the presence of α- or β-CD, or without CD under similar conditions 42-44 were 10% to 30%, while area rate was > 96% for γ-CD (data not shown). Therefore, GBE-CD was prepared by treating GBE with naringinase in the presence of γ-CD (Figure 1A).

(A) Formualtion of flavonoid-monoglucoside (quercetin-glucoside, isorhamnetin-glucoside, and kaempferol 3-glucoside)–γ-cyclodextrin complexes from Ginkgo biloba extract (GBE) using a patented method.45-47 (1) GBE (15%) and γ-CD (10%) were mixed in water. (2) Temperature and pH of the solution were adjusted to 73 °C and 3.8, using sodium hydroxide (NaOH) and sulfuric acid (H2SO4), respectively. (3) Naringinase (30 unit/g flavonoid) digestion was conducted for 12 h. (4) The solution containing the product was filtered and freeze-dried. Abbreviations: Glc, glucose; Rha, rhamnose. (B) Comparative HPLC analysis in (A) GBE
and (B) GBE-CD
a1 and b1 are peaks in GBE, and a2 and b2 are peaks in GBE-CD. λmax of a1 and a2 peaks: 255 and 352 nm, respectively; λmax of b1 and b2 peaks: 265 and 343 nm, respectively. Abbreviations: RTN, rutin (quercetin-rutinoside); IQC, isoquercitrin (quercetin-glucoside); KAER, kaempferol-rutinoside; KAEG, kaempferol-glucoside; ISRR, isorhamnetin-rutinoside; ISRG, isorhamnetin-glucoside.
A comparison of HPLC profiles between GBE, and GBE-CD (after12 h of enzymatic digestion), was shown in Figure 1B. The HPLC profile shows that GBE contains intense peaks (blue line) of FLRs such as RTN (retention time [RT]: 16.0 min), KAER (RT: 24.1 min), and ISRR (RT: 29.4 min), and small peaks of IQC, KAEG, and ISRG. Distinctly, the GBE-CD profile contains intense peaks (red line) to FMGs, such as IQC (RT: 18.0 min), kaempferol-glucoside (KAEG) (RT: 26.4 min), and isorhamnetin-glucoside (ISRG) (RT: 32.4 min), were shifted correlatively after12 h of the enzymatic reaction.
Moreover, we observed the characteristic peaks (a1, a2, b1, and b2) of flavonoids with two λmax peaks: a1 (RT: 11.58 min) in GBE, a2 (RT: 13.8 min) (λmax: 255 and 352 nm) in GBE-CD, b1 (RT: 15.15 min) in GBE and b2 (RT: 18.8 min) (λmax: 265 and 343 nm) in GEE-CD using photo-diode array detection in HPLC.51-53 It was postulated that the unknown FLR in GBE changed to FMG because of the two same λmax peaks and the shifted RT. This further confirmed the shifted peak from FLR to FMG in the prepared GBE-CD inclusion complex.
Pharmacokinetic Study in Rats
The mean plasma flavonoid concentration-time profiles of KER, KAE, ISR, and TAM after β-glucuronidase digestion of KER, KAE, ISR, and tamarixetin (TAM) glucuronide conjugate in plasma at 15, 30 min, 1, 2, 3, 6, 9, 12, and 24 h after the administration of GBE-CD and GBE are analyzed.
The bioavailability of QCT, KAE, and ISR, respectively, in the plasma after GBE or GBE-CD administration in SD rats, is shown in Table 1(a) to (d) and Figure 2(A) to (D). The AUC0-24 of the flavonoids QCT, KAE, and ISR after GBE-CD administration was 5.4, 6.1, and 10.4 times higher, respectively than after GBE administration in rats. Further, TAM was detected in the plasma after the GBE-CD administration, but not after the GBE administration.

Pharmacokinetic profiles of (A) quercetin, (B) kaempferol, (C) isorhamnetin, (D) tamarixetin, and (E) total four flavonoids in plasma after the administration of Ginkgo biloba extract–γ-cyclodextrin complex (GBE-CD) or Ginkgo biloba extract (GBE) in Sprague–Dawley (SD) rats. Values are mean ± standard error of mean (SEM, n = 6). ***P < 0.001, **P < 0.01, *P < 0.05, GBE
and GBE-CD
.
Pharmacokinetic parameters as the area under the concentration–time curve from 0 to 24 h (AUC0-24 μM.h), maximum concentration (Cmax), and time to reach the peak concentration (Tmax) of quercetin, kaempferol, isorhamnetin, tamarixetin, and four flavonoids after the administration of Ginkgo biloba extract-γ-cyclodextrin complex (GBE-CD) and Ginkgo biloba extract (GBE) to Sprague–Dawley (SD) rats.
Four flavonoids [mixture of (a) to (d)]; significance: P < 0.05; Student t-test.
Zheng et al. 24 reported that the AUC0-∞ of the flavonoids QCT, KAE, and ISR after HCL hydrolysis of flavonoids metabolites in plasma after administration of a formulation of GBE pro-liposomes containing egg yolk phosphatidylcholine and sodium deoxycholate (NaDC) were 2.45, 2.11, and 2.64, respectively, compared to those after GBE administration in rats. Additionally, Chen et al. 25 demonstrated that the bioavailability of QCT, KAE, and ISR after HCL hydrolysis of flavonoids metabolites in plasma after the GBE-phospholipid (GBP) or GBE-solid dispersion (GBS) complexes administration increased significantly by 2.42, 1.96, 2.36, or 2.08, 1.43, and 1.5- folds, respectively, compared to those after GBE administration in rats. These results show the bioavailability of QCT, KAE, and ISR in the plasma after GBE-CD is notably 2 to 7 folds higher than those of previous reports. 24 25
Moreover, in Figure 2(E) (total four flavonoids ), the plasma profile of total flavonoid metabolites showed two intense peaks at ∼ 30 min and 3 h after GBE-CD administration, which decreased steadily for 24 h thereafter, whereas a small peak from 30 min to 12 h, decreasing gradually for 24 h was identified after GBE administration. Table 1(e) shows the AUC0-24, Cmax, and time to reach the peak concentration (Tmax) of the total four flavonoids after GBE-CD complex administration (equivalent to QCT), which were 7.4 times greater, 6 times higher, and markedly 3 h faster than those of GBE, respectively.
Therefore, it is anticiapted that flavonoids after GBE-CD administration could show physiological effects at targeted sites more effectively than those after GBE, GBP, and GBS administration.
Conclusions
In summary, the present study indicates that the oral administration of the GBD-CD inclusion complex leads to 5 to 10- folds higher levels of KER, KAE, and ISR (Table 1) after β-glucuronidase digestion in plasma compared to after GBE intake in rats. The formulation of treating FLRs using CD and naringinase would have great potential to improve the oral absorption of herbal remedies and food supplements with poor bioavailability, including FLRs.
However, inclusive elucidation proofs using spectroscopic techniques, various safety measures, and pharmacokinetic analysis for absorption, metabolism, and excretion are needed to further explore the potential of GBE flavonoids.. Also, additional animal tests should be conducted before performing human clinical studiesto determine the GBE flavonoid's effectiveness in health benefits and disease prevention.
Materials and Methods
Materials
γ-Cyclodextrin (purity > 98%) was obtained from CycloChem Co., Ltd, and quercetin (97.3%), kaempferol (99.7%), and isorhamnetin (98.6%) were purchased from FUJIFILM Wako Pure Chemical Corporation. Naringinase was obtained from Amano Enzyme Inc. Other flavonoids used in HPLC analysis were of very high purity, and were obtained from FUJIFILM Wako Pure Chemical Corporation.
Production of Ginkgo biloba Extract-γ-CD Complex and HPLC Analysis
GBE-CD was prepared by treating GBE (26.1% flavones; China Material Company) in the presence of γ-CD, with naringinase, using a patented method (Figure 1).42-44
The production of FMGs (IQC, ISRG, and KAEG) from RTN, ISRR, and KAER correlated with the maximum peak area at 12 h after the enzyme reaction (Figure 2).
The FMG area rate (%) greater than 96% was determined using HPLC with the following equation:
The enzyme reaction was analyzed using the Nexera XR HPLC system (Shimadzu Co., Ltd) with the following specifications: column, UG-120 (5 μm, 150 mm × 4.6 mm ID; Shiseido); eluent, 15% (v/v) acetonitrile/0.1% aqueous phosphoric acid solution; column temperature, 70 °C, detection, absorbance at 351 nm; and flow rate, 0.3 mL/min. The presence of respective flavonoids was confirmed using HPLC analysis with RTN (RT: 16.0 min), IQC (RT: 18.0 min), KAER (RT: 24.1 min), KAEG (RT: 26.4 min), ISRR (RT: 29.4 min), and ISRG (RT: 32.4 min) (Figure 2).
Analysis of Flavone Contents Using HPLC
We analyzed the flavone content in GBE and GBE-CD using HPLC according to a previously described method of sample preparation and acid hydrolysis. 54
The HPLC analysis for the prepared samples was performed using the Nexera XR HPLC system (Shimadzu Co., Ltd) with the following specifications: column, UG-120 (5 μm, 150 mm × 4.6 mm ID; Shiseido); column temperature, 60 °C; mobile phase, 0.1% phosphoric acid (solvent A), and acetonitrile (solvent B; B/A 30/70 for 12 min); detection, absorbance at 360 nm; flow rate, 1.5 mL/min. The presence of respective flavonoids was confirmed using HPLC analysis with QCT (RT: 2.8 min), KAE (RT: 4.3���min), and ISR (RT: 4.9 min).
GBE: flavones 26.1%a (QER: 4.45%, KAE: 4%, ISR: 1.9%), GBE-CD: flavones 14.6%c (QER: 2.5%, KAE: 2.24%, ISR: 1.0%), GBE ([enzyme-treated]: 57%b), γ-CD: 39%, and other: 4% saccharides, etc. The flavonoid content (%) was calculated using the following formula55,56:
Pharmacokinetic Study in Rats
Male 4 to 5 week-old SD (Crl: CD (SD)) rats were obtained from Atsugi Breeding Center (Charles River Laboratories). The rats were placed in a temperature-controlled room under the following conditions: temperature, 20 °C to 26 °C; relative humidity, 30% to 70%; air ventilation rate, 10 to 15 times/h; and 12 h light/12 h dark cycles. Moreover, the rats were raised in breeding cages (44 cm width × 27.5 cm length × 18 cm height; Hanyu Seimitsu Co., Ltd; two rats/cage) with bedding. The experimental rats were allowed free access to solid CR-LPF feed (Oriental Yeast Co., Ltd) and tap water during the feeding period.
The general conditions of the rats (extracorporeal appearance, nutritional situation, bearing, movement, and feces) were observed once daily. After a quarantine-acclimation period of at least 1 week, the rats were ingested with GBE-CD or GBE. The dosing formulations of GBE-CD and GBE were provided as QCT equivalence (2.5 mmol/L) in water for oral administration (Japan Pharmacopeia, Otsuka Pharmaceutical Factory, Inc.), and six rats each were administered the test substances individually via oral gavage (25 μmol as QCT/kg body weight (BW), i.e., 10 mL/kg BW).
Approximately 0.3 mL of blood was collected from the jugular vein of the rat with a 1 mL heparinized syringe equipped with a 23-gauge needle before and 15, 30 min, 1, 2, 3, 6, 9, 12, and 24 h after the administration of the GBE-CD or GBE (n = 6), without anesthesia. The collected blood samples were transferred to a polypropylene tube, cooled on ice, and centrifuged at 6000 × g for 2 min at 4 °C to obtain the plasma (∼120 µL), which was frozen at ≤ −80 °C in a freezer until analysis.
Pharmacokinetic Analysis of Metabolites of Quercetin, Kaempferol, Isorhamnetin, and Tamarixetin in Rat Plasma
The plasma was obtained from the blood samples in the presence of heparin, and analyzed to detect QCT, KAE, ISR, and TAM metabolites using a previously described method, 40 with the following minor modifications. 70 μL of plasma sample and 20 μL of β-glucuronidase (EC No. 232-606-8) from Helix Pomatia (#G0751, 34110 units/mL β-glucuronidase and > 5 × 102 units/mL of sulfatase; Sigma-Aldrich) were mixed in 0.58 M acetate buffer at pH 4.9, and incubated for 60 min at 37 °C. Subsequently, 25 μL of 10 mM sodium metaphosphate solution containing 1 M L-ascorbic acid and 9 μL of myricetin (internal standard, 20 μg/mL) were mixed with the reaction mixture, which was then freeze-dried.
The freeze-dried mixture was dissolved in 1 mL of MeOH and centrifuged (6000 × g for 10 min), and then the supernatant (800 μL) was transferred to a new tube and dried under nitrogen flow. The residue was dissolved in 150 μL of MeOH/200 mM hydrochloric acid (HCl, 1:1) and subsequently centrifuged (6000 × g for 7 min). Forty (40) microliters of the supernatant were applied to HPLC analysis using the Nexera XR system with the following methods; column, YMC carotenoid (5 μm, 4.6 mm × 250 mm; YMC Co., Ltd); mobile phase, 0.1% phosphoric acid (solvent A), and acetonitrile (solvent B; B–A (from 25:75 to 35:65) for 5 min, 35:65 from 5 to 18 min, and 59.5:40.5 from 18 to 30 min, linear gradient); flow rate, 1.0 mL/min; column temperature, 35 °C; detection, absorbance at 370 nm; and RT for myricetin, QCT, KAE, ISR, and tamarixetin (TAM), were 8.5, 12.8, 20.8, 24.1, and 25.3 min, respectively. Linear regression of the concentration range of 0.1 to 300 μM for QCT, KAE, ISR, and TAM by the peak area ratio of these compounds to myricetin was calibrated using the least-squares method (r2 = 0.99).
Statistical Analysis
The pharmacokinetic profiles of plasma QCT, KAE, ISR, and TAM metabolites in SD rats were statistically analyzed. In the baseline comparison of the results of the rat experiments, a two-tailed t-test for two independent samples was performed. Results with ***P < 0.001, **P < 0.01, and *P < 0.05 were considered statistically significant. The results are shown as mean ± standard error of mean (SEM) for normal distribution variables.
Footnotes
Abbreviations
Acknowledgements
The authors would like to thank the colleagues at Taiyo Kagaku Co. Ltd., Japan, and BoZo Research Center Inc., Japan for thier kind support during this study.
Declaration of Conflicting Interests
The authors declare no known potential conflict of interest with respect to the research, authorship, and/or publication of this article.
Ethical Approval
The pharmacokinetic study was approved by the Institutional Animal Care and Use Committee (Number G200226), and all proceedings were performed according to the Act on Welfare and Management of Animals (October 1, 1973; Law No. 105).
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Statement of Human and Animal Rights
The experimental rats were treated following the Standards for Breeding and Storage of Laboratory Animals and Reduction of Pain (April 28, 2006; Notification No. 88; Ministry of the Environment).
Informed Consent
This study did not involve human participants.
