Enhanced dissolution rate and oral bioavailability of ginkgo biloba extract by preparing nanoparticles via emulsion solvent evaporation combined with freeze drying (ESE-FR)

Abstract

Ginkgo biloba extract nanoparticles were successfully prepared by the method of emulsion solvent evaporation combined with freeze-drying (ESE-FR). By the single-factor method, the optimal parameters were determined: the concentration of GBE was 5 mg mL⁻¹, volume ratio of water to organic phase was 1 : 1, homogenate speed and time were 12 000 rpm and 3 min, respectively, and homogeneous pressure and times were 500 bar and 3, respectively. The mean particle size (MPS) of the GBE nano-emulsion prepared using the optimized conditions was 56.0 nm. SEM detection results indicated that the morphology of raw GBE was irregular. But the morphology of GBE nanoparticles was close to spherical and the particle size of GBE nanoparticles was much smaller than that of raw GBE. The particle size of the freeze-dried powder was 277.0 nm. The characteristics of prepared GBE nanoparticles were detected via XRD, DSC and TG, which showed that the crystal structure of GBE nanoparticles was the same as that of raw GBE. Through the test of solubility and dissolution in vitro, the solubility and dissolution of GBE nanoparticles in artificial gastric juice, artificial intestinal juice and deionized water were obviously improved compared with that of raw GBE. The results of bioavailability showed that the Area Under Plasma Concentration Curves (AUC) of flavonoids and terpene lactones in GBE nanoparticles were improved about 1.81 times and 1.75 times compared with that in raw GBE, respectively. In addition, the residual amounts of chloroform and ethanol were less than the ICH limit for class II.

---

Introduction

Ginkgo biloba has existed on the earth for more than 200 million years, and is well known as a “living fossil”.¹ In recent years, ginkgo biloba extracts (GBE) have been widely used in the world as a phytomedicine and a dietary supplement.² The standardized preparation of GBE is Egb761, which contains 24% ginkgo flavonoids, 6% terpene lactones, and no more than 5 parts per million ginkgolic acids, in which ginkgo flavonoids (quercetin, kaempferol, isorhamnetin) and terpene lactones (ginkgolide A, ginkgolide B, ginkgolide C, bilobalide) are the bioactive components in GBE.³ The chemical structures of these bioactive components are shown in Fig. 1. Increasing research shows that the bioactive components in GBE can improve blood circulation, prevent thrombosis, enhance the capillary walls of capillaries and protect nerve cells from harm of hypoxia.⁴ GBE is used to treat mental disorder, such as concentration difficulties and memory impairment.⁵ Moreover, GBE has significantly therapeutic effect on antioxidant,⁶ antiasthmatic,⁷ scavenge radicals,⁸ wound healing⁹ and neurodegenerative dementias associated with aging, Alzheimer's disease.¹⁰ At present, the market has a wide variety of pharmaceutical products of GBE, such as granules, tablets, capsules, pills, injection, etc.,¹¹ in those productions Kinnado is one of the best-selling GBE pharmaceutical products.¹²

Fig. 1 The chemical structures of quercetin, kaempferol, isorhamnetin, ginkgolide A, ginkgolide B, ginkgolide C and bilobalide.

At present, the pharmaceutical products of GBE on the market are mostly oral agent and injection. Comparing with the injection, oral agent has the advantages of high safety and convenience. However, the poor aqueous solubility and low dissolution rate of GBE limits the body's absorption of the active ingredients in GBE after oral administration.¹³ Recently, in order to improve the oral bioavailability, many researchers have investigated the methods of improving the aqueous solubility of the GBE. Wang and co-workers¹⁴ enhanced dissolution rate and oral bioavailability of GBE by preparing solid dispersion via hot-melt extrusion. Tang et al.^15,16 improved the vitro dissolution of GBE by rapid self-emulsification. Moreover, Chen and colleagues¹⁷ improved the oral bioavailability of GBE through preparing its phospholipids complexes and solid dispersions. Various approaches to overcome the poor aqueous solubility of drug candidates have been investigated in drug research and development, such as preparation of nanoparticles,¹⁸ crystal modification,¹⁹ cyclodextrin complexation²⁰ and pH modification.²¹ In recent years, preparation of nanoparticles has been used frequently in improving the water solubility and bioavailability of the drug.²² The preparation of nanoparticles mainly includes bottom-up and top-down technologies.²³ Emulsification is a kind of bottom-up technology, which is usually used to prepare nanoparticles. In the process of preparing nanoparticle by emulsification technology, the organic phase containing drugs and water phase is fully mixed into nano-emulsion. Nano-emulsion droplet sizes fall typically in the range of 20–200 nm and show narrow size distributions. Next step, the nano-emulsion is processed through high pressure homogenization and the organic solvent is removed by rotary evaporation. Finally, the remaining aqueous phase is removed by an appropriate method.²⁴ At present, there have been several kinds of nanoparticles such as genipin,²⁵ 10-hydroxycamptothecin,²⁶ poly(lactide-co-glycolide) (PLG)²⁷ and puerarin²⁸ were produced by this technique.

In this work, the preparation technology of nanoparticles of GBE was executed in experiment by the emulsion solvent evaporation and the freeze drying combination method (ESE-FR). The effect of various experimental parameters on particle size was investigated by single-factor method. These experimental parameters consisted of the concentration of GBE solution, volume ratio of water phase to oil phase, speed of homogenate and homogenate time. The optimal experimental parameters were selected. Then the nano-emulsion under the optimal experimental parameter was processed by high pressure homogenization. The influence of pressure and times of high pressure homogenization on the particle size were also studied, respectively. Lastly, the GBE nanoparticle was prepared by vacuum freeze-dry. The raw GBE and the optimized GBE nanoparticle were analyzed by SEM, FTIR, XRD, DSC, TG, HPLC, LC-MS/MS, dissolution test, residual solvent determination and bioavailability test.

Materials and methods

Materials

GBE was purchased from Xi'an Xiaocao Botanical Development Co. Ltd. (Xi'an, PR China), with 24% flavonoids and 6% terpene lactones, and no more than 5 parts per million ginkgolic acids. Hydroxypropyl-β-cyclodextrin (HP-β-CD) was obtained from Yuanye Bio Technology Co., Ltd. (Shanghai, China). Reference standards of quercetin (CAS: 117-39-5, QCT), kaempferol (CAS: 520-18-3, KMF), isorhamnetin (CAS: 480-19-3, ISR), bilobalide (CAS: 33570-04-6, BB), ginkgolide A (CAS: 15291-75-5, GA), ginkgolide B (CAS: 15291-77-7, GB), and ginkgolide C (CAS: 15291-76-6, GC) were all purchased from Shanghai Aladdin Biological Technology Co. Ltd. Ketoprofen (CAS: 22071-15-4) was purchased from National Institutes for Food and Drug Control (Beijing, China). Trichloromethane (purity >99.5%), ethanol (purity >99.5%) and the other reagents were all analytical grades.

Preparation of the GBE nanoparticles

The GBE nanoparticles were prepared by the method of ESE-FR. The preparation process of the GBE nanoparticles is shown in Fig. 2.

Fig. 2 Diagram of experimental process to prepare the GBE nanoparticles.

Preparation of the GBE nano-emulsion

In this experiment, the mixed solution of chloroform and ethanol (the volume ratio was 7 : 3) was chosen as the oil phase and the deionized water was selected as water phase. Firstly, the GBE was dissolved into the oil phase to form a solution with certain concentration. Secondly, the GBE solution was added into the deionized water slowly under the high-speed stirring state via homogenate. After a certain period of high speed homogenate, primary GBE nano-emulsion was formed. The GBE nano-emulsion was obtained after high-pressure homogenization by using Nano homogenize machine (AH100D, ATS Engineering Inc., Shanghai, China), which was equipped with emulsion valve. Firstly the second-stage valve was regulate the pressure to the total pressure of 10% and then first-stage valve was increased the pressure to the total pressure.

Single-factor method was used to study the optimal operating parameters affecting particle size of GBE nano-emulsion during the preparation process. The parameters include two parts: (a) in the preparation of the primary GBE nano-emulsion process, there are four parameters: the concentration of GBE solution, volume ration of water phase to oil phase, speed of homogenate and time of homogenate, can affect the particle size of primary GBE nano-emulsion; (b) during the homogeneous process, the homogeneous pressure and times have an effect on the size of GBE nano-emulsion. The concentration of GBE solution were 1–5 mg mL⁻¹, the volume ration of water phase to oil phase were 1 : 1–9 : 1, the speed of homogenate were 8000–16 000 rpm, the homogenate time were 1–9 min, the homogeneous pressure were 100–900 bar, the homogeneous times were 3–11, respectively. Example of implementation methods was as follows: when concentration of GBE solution was studied, the other factors were set under the following conditions: the volume ration of water phase to oil phase, speed of homogenate, homogenate time, homogeneous pressure and homogeneous times were set at 5 : 1, 12 000 rpm, 5 min, 500 bar and 7, respectively. The same method was used to accomplish the investigation of other factors. Finally, the optimal condition was selected according to the minimum particle size of each factor. The GBE nano-emulsion was prepared under the optimum condition.

Preparation of GBE nano-suspension

The oil phase in nano-emulsion was removed by a rotary evaporator R201BL (SENCO, Shanghai, China) under the condition of temperature of 40 °C and vacuum pressure of −0.06 Pa. During the process, an equal volume of deionized water with GBE nano-emulsion was added into the GBE nano-emulsion. The GBE nano-suspension was obtained after removing the oil phase.

Lyophilization

After the GBE nano-suspension was obtained, the lyoprotectant was added into GBE nano-suspension and mixed fully. In this experiment, according to pre-experimental screening result and previous report,²⁹ the HP-β-CD was selected as the lyoprotectant, and the mass ration of the HP-β-CD and GBE in nano-suspension was 3 : 1. The GBE suspension was frozen at −40 °C to become a solid state. Finally, the GBE nanoparticles were obtained after freeze-drying at −50 °C, −10 Pa for 72 h.

Detection of main components

Detection of flavonoids. An aliquot 35 mg sample was mixed with 25 mL methanol–25% hydrochloric acid solution (4 : 1), hydrolyzed for 30 minutes, cooled to room temperature rapidly, then transferred to a 50 mL volumetric flask and added methanol to scale line. All samples were filtered with 0.22 μm membrane before injected into an HPLC system (Waters Company, USA). The chromatographic column used Diamonsil C₁₈ (200 mm × 4.6 mm, 5 μm), with methanol–0.4% phosphoric acid solution (52 : 48) as mobile phase, detection wavelength: 360 nm, the flow rate: 1.0 mL min⁻¹, the injection volume: 10 μL. The column temperature: indoor temperature.³⁰ The total flavonoids (TFs) were calculated according to eqn (1):³¹

Total flavonoids (TFs) = (QCT + KMF + ISR) × 2.51 (1)

Detection of the content of terpenoid lactones. 150 mg sample was dissolved in 10 mL water which mixed with 50 μL 2% hydrochloric acid solution and warmed in 50 °C water bath. Then 2 mL ethyl acetate was added into the suspension of GBE and shaking fully, and the ethyl acetate layer was separated by centrifugation. This operation was repeated four times. The terpenoid lactones were measured by LC-MS/MS (Applied Biosystems, CA), equipping with electrospray mass spectrometry detector API3000 LC-MS/MS for quantitative analysis of terpenoid lactones. Chromatographic conditions: using Agilent Eclipse XDB-C₁₈ chromatographic column (150 mm × 4.6 mm, 5 microns Agilent, USA), methanol : water = 7 : 3 (containing 0.2% ammonium acetate) as mobile phase, the flow rate been setting as 1.0 mL min⁻¹ and injected sample volume been 5 μL. For operation in MS/MS mode, negative ion scanning; under the condition of the thermal stability of ion source in front of the sample for 30 minutes; multiple reaction monitoring ion to choose: ginkgo lactone A was m/z 407.5 to 351.3; ginkgo lactone B was m/z 423.4 to 367.1; ginkgo lactone C was m/z 439.5 to 383.3; bilobalide was m/z 325.4 to 163.1. The other parameters were as following: nebulizing gas, curtain gas and collision gas were 12, 10 and 6 L min⁻¹, respectively, dwell time was 1.5 s, ion spray voltage was 5000 V, ion source temperature was 350 °C, declustering potentials (DP) were −42 V for ginkgolide A, −25 V for ginkgolide B, −28 V for ginkgolide C, −20 V for bilobalide, focusing potential (FP) and entrance potential (EP) were 400 and 10 V, respectively; collision energies (CE) were −25 V for ginkgolide A, −26 V for ginkgolide B, −24 V for ginkgolide C, −28 V for bilobalide, respectively; collision cell exit potentials (CXP) were −6 V for ginkgolide A, −7 V for ginkgolide B, −5 V for ginkgolide C, −5 V for bilobalide, respectively.³² The total terpene lactones (TTLs) were calculated as the sum of BB, GA, GB and GC (eqn (2)).

Total terpene lactones (TTLs) = BB + GA + GB + GC (2)

Characterization of GBE nanoparticle

Particle size. Mean particle size (MPS) of GBE nano-emulsion and GBE freeze-dried powder was detected by laser light scattering using a Zeta PALS Analyzer (Brookhaven Instruments, Holtsville, NY, USA). The measurement was performed with angle detection at 90° in 4 mm diameter cells with field strength of 10 V cm⁻¹. The GBE freeze-dried powder was dispersed in pure water for detection. Every measurement was repeated three times.

Morphology analysis. SEM (Quanta 200, FEI) was employed to analyze the particle shape of raw GBE, GBE nanoparticles and GBE nanoparticles without HP-β-CD. The raw GBE and GBE nanoparticles were observed in dry samples. The GBE nanoparticles without HP-β-CD were observed in liquid samples. GBE nanoparticles were dispersed in deionized water, and the GBE nanoparticles without HP-β-CD was obtained by centrifugal and the GBE nanoparticles without HP-β-CD need to be washed with deionized water to remove the HP-β-CD. Each dried sample was onto the surface of an aluminum alloy sample table with conductive adhesive directly. The liquid samples were dripped into the foil which was stuck to the aluminum alloy station by using double-sided adhesive. Before the analysis, the samples were sputter-coated with gold under the argon atmosphere.

Crystal structure analysis. The crystal structures of raw GBE, HP-β-CD, GBE nanoparticles and physical mixture of raw GBE with HP-β-CD were detected via XRD, DSC and TG. X-ray diffraction analysis was used to detect the crystal form of the raw GBE, HP-β-CD, GBE nanoparticles and physical mixture of raw GBE with HP-β-CD by using an X-ray diffractometer with a rotating anode (Philips, Xpert-Pro, Netherlands). The scanning range was from 5° to 60° with a step size of 0.02°, and the scanning rate was 5° min⁻¹. The samples were irradiated using a Cu target tube at 30 mA and 40 kV. Thermal stability analysis was employed to researched the heating stability of the raw GBE, HP-β-CD, GBE nanoparticles, physical mixture of raw GBE with HP-β-CD by DSC (TA instruments, model, DSC 204), and TG (TGA, Diamond TG/DTA PerkinElmer, USA). The test temperature range of DSC was from 30 °C to 300 °C with a heating rate of 10 °C min⁻¹, and that of TG was from 30 °C to 600 °C with a heating rate of 10 °C min⁻¹.

Solubility analysis. The solubility of raw GBE, GBE nanoparticles and physical mixture of raw GBE with HP-β-CD in artificial gastric juice (PH 1.2), artificial intestinal juice (PH 6.8) and deionized water (pH 7.2) was detected by using the HPLC method. According to the preparation methods in Pharmacopoeia of the People's Republic of China,³³ the artificial gastric juice and artificial intestinal juice were prepared. At the beginning of the experiment, excess of raw GBE, GBE nanoparticles and physical mixture of raw GBE with HP-β-CD were added into three vials, in each vial been equipped with 3 mL artificial gastric juice, artificial intestinal fluid or deionized water. Then the vials were put in the water bath of 37 °C, 100 rpm for 48 h. After this, taking 1 mL suspension, centrifuging under 10 000 rpm for 10 min, then taking 100 μL supernatant and diluting with methanol to 1 mL. The sample solution was detected by using HPLC after centrifuging under 10 000 rpm for 10 min.

Dissolution analysis. The dissolution analysis of raw GBE, GBE nanoparticles, physical mixture of raw GBE with HP-β-CD in the artificial gastric juice (pH 1.2), artificial intestinal juice (pH 6.8) and deionized water (pH 7.2) were performed by dialysis method. Raw GBE suspension (GBE concentration of 3.2 mg mL⁻¹, 7 mL), GBE nanoparticles (GBE concentration of 3.2 mg mL⁻¹, 7 mL) and physical mixture of GBE with HP-β-CD (GBE concentration of 3.2 mg mL⁻¹, 7 mL) were separately placed in treated dialysis bags (MWCO3500; Sigma, St. Louis, USA) and then the dialysis bags were immersed in three identical beakers containing 200 mL artificial gastric juice, artificial intestinal juice or deionized water, respectively. Then the beakers were placed in a water bath at 37 °C, and the stirring rate was set at 100 rpm. At the time point of 2, 5, 10, 15, 25, 35, 45, 60, 75, 90, 120 min, the solution with 2 mL was taken out and filtered using 0.22 μm cellulose nitrate membrane. The removed solution was replaced with 2 mL fresh medium.

The dissolution rate was determined by plotting the cumulative amount of solute dissolved against time. Dissolution is calculated according to the following equation.¹⁴

D = dissolution, C_n = drug concentration at each time interval, V_n = sampling volume at each time interval, M = total amount of active ingredients.

Bioavailability analysis. Twelve male Sprague–Dawley rats, weighing 200–250 g, were divided randomly into two equal groups: reference group and experiment group. The animal use and care protocol was reviewed and approved by the ethics committee of the Harbin Medical University. All experiments were performed in compliance with relevant laws and followed institutional guidelines; animals were housed under standard conditions of temperature, humidity and light with food and water provided freely and allowed to acclimatize in the laboratory for at least 1 week prior to the experiment. Before the experiment, all the rats were fasted for 12 h and water was given ad libitum. At the beginning of the experiment, the rats in reference group were given the raw GBE at a dose of 40 mg kg⁻¹ according to the method of oral administration, and the rats in experiment group were given the GBE nanoparticles with the same dose and method to the reference group. To determine the drug concentration and calculate the pharmacokinetic parameters, a series of blood samples were collected at the following times: 0.083, 0.167, 0.333, 0.5, 1.0, 2.0, 4.0, 6.0, 8.0, 10.0, 12.0, 24.0 h after dosing. The samples were immediately centrifuged at 3000 rpm for 10 min and plasma was separated from sediment cells with pipet. Samples were stored at −20 °C until analysis.

Plasma sample preparation. The plasma samples were extracted by using liquid–liquid extraction technique. 400 μL of plasma was added into an equal volume of 25% hydrochloric acid and put the plasma under vortex movement for 3 min. Then the plasma was hydrolyzed in a water bath at 80 °C for 60 min. After this, the hydrolyzed sample was added 2 mL ethyl acetate and under vortex movement for 3 min. The mixed solution was centrifuged at 3000 rpm for 10 min to separate the ethyl acetate layer. Then 2 mL ethyl acetate was added into the remaining water layer and mixed by vortex for 3 min. The ethyl acetate layer was separated from the mixed solution again by centrifuged at 3000 rpm for 10 min, and was combined with the previous ethyl acetate layer. The obtained ethyl acetate was evaporated to dryness under a gentle stream of nitrogen. The residue was reconstituted with 200 μL methanol, and then the solution was centrifuged at 15 000 rpm for 10 min before been injected into HPLC and LC-MS/MS system.

Detection of residual solvents. The residual chloroform in the GBE nanoparticle was analyzed by using an Agilent 7890A gas chromatograph (GC, Agilent Technologies, Palo Alto, CA, USA) with HP-5 5% phenyl methyl siloxane capillary column (30.0 m × 250 mm × 0.25 mm, nominal) equipping with an G1540N-210 FID detector. Peaks areas were used for obtaining quantitative data. The analysis conditions of GC for chloroform were as following: oven temperature was maintained at 40 °C for 5 min initially, and then raised at the rate of 20 °C min⁻¹ to 200 °C maintaining for 2 min at last. Nitrogen was used as carrier gas at a flow rate of 30 mL min⁻¹. Then 10 mg of GBE nanoparticles were dissolved by 1 mL methanol and centrifuged at 8000 rpm for 5 min, and 1 μL samples were manually injected in split mode with a split ratio 25 : 1. Hydrogen gas and air flow rate were 30 mL min⁻¹ and 400 mL min⁻¹, respectively. The conditions of GC analysis of ethanol were as follows: the oven temperature was initially maintained at 40 °C for 8 min and then ramped at a rate of 40 °C min⁻¹ to 240 °C and held for 10 min. The injector temperature was set at 200 °C and the detector was 280 °C, respectively. Nitrogen was used as carrier gas at a flow rate of 25 mL min⁻¹. Then 10 mg of GBE nanoparticles were dissolved by 1 mL deionized water and centrifuged at 8000 rpm for 5 min. 1 μL samples were injected manually in split mode, with a split ratio 2 : 1. The hydrogen gas and air flow rates were 30 and 400 mL min⁻¹, respectively.

Results and discussion

Optimization study

The optimal condition is chosen by using single-factor method and the optimization of experimental parameters was listed in Table 1 and 2. When determining one of the factors, the other factors were selected as fixed value. During the process of homogenate, the following factors affecting the particle size of primary GBE nano-emulsion were determined: the concentration of GBE solution, volume ration of water phase to oil phase, speed of homogenate and homogenate time. During the process of homogeneous, the optimal value of the homogeneous pressure and the homogeneous times were determined.

Table 1 Experimental parameters of homogenate process

No.	Concentration of GBE solution (mg mL^−1)	Volume ratio (v/v)	Speed of homogenate (rpm)	Homogenate time (min)	MPS of primary nano-emulsion (nm)
1	1	5	12 000	5	96.6
2	2	5	12 000	5	74.7
3	3	5	12 000	5	62.2
4	4	5	12 000	5	63.3
5	5	5	12 000	5	61.5
6	5	1	12 000	5	23.1
7	5	3	12 000	5	43.2
8	5	5	12 000	5	66.8
9	5	7	12 000	5	85.4
10	5	9	12 000	5	102.1
11	5	1	8000	5	114.9
12	5	1	10 000	5	83.8
13	5	1	12 000	5	52.2
14	5	1	14 000	5	56.2
15	5	1	16 000	5	57.6
16	5	1	12 000	1	64.2
17	5	1	12 000	3	48.4
18	5	1	12 000	5	58.0
19	5	1	12 000	7	68.1
20	5	1	12 000	9	85.4

---

Table 2 Experimental parameters of homogenization process

No.	Homogeneous pressure (bar)	Homogeneous times	MPS of nano-emulsion (nm)
1	100	7	123.9
2	300	7	69.2
3	500	7	57.5
4	700	7	74.0
5	900	7	83.8
6	500	3	94.3
7	500	5	67.0
8	500	7	62.4
9	500	9	60.3
10	500	11	56.0

---

Effect of the concentration of GBE solution

The concentration of GBE was set within the range from 1 mg mL⁻¹ to 5 mg mL⁻¹. The MPS of primary GBE nano-emulsion with the increasing concentration of GBE was presented in Fig. 3a. As shown in Fig. 3a, the MPS of primary GBE nano-emulsion decreased from 96.6 nm to 62.2 nm with the concentration of GBE changing from 1 mg mL⁻¹ to 3 mg mL⁻¹, and when the concentration of GBE solution was greater than 3 mg mL⁻¹, the MPS of primary GBE nano-emulsion almost unchanged. The reason for this phenomenon is that flavonoids can act as stabilizers of oil-in-water (O/W) emulsions through pickering stabilization. Due to the flavonoids existing as insoluble particles in the aqueous phase and tending to absorb on the oil-water interface, the flavonoids could provide an efficient steric barrier against the coalescence of emulsion oil globules.³⁴ However, when the concentration of GBE solution was higher than 5 mg mL⁻¹, the GBE nanoparticles in suspension were easy to agglomerate and precipitate in the process of removing the organic phase. In order to improve the yield and ensure the quality of products, the GBE solution with 5 mg mL⁻¹ was selected as the optimal concentration.

Fig. 3 Influence of four parameters on the mean particle size of GBE nano-emulsion during the process of homogenate. (a) Concentration; (b) volume ratio of water phase to organic phase; (c) speed of homogenate; (d) homogenate time.

Effect of the volume ratio of water phase to oil phase

The MPS of primary GBE nano-emulsion with the increasing volume ratio of water phase to oil phase was shown in Fig. 3b. From Fig. 3b it can be seen that the MPS of primary GBE nano-emulsion increased gradually when the volume ratio of water phase to oil phase increased from 1 : 1 to 9 : 1.

The reason for this phenomenon is that the increase of volume ratio of water phase to oil phase results in the decrease of concentration of flavonoids in the emulsion, and weakens the stability of the nano-emulsion. When the volume ratio was 1 : 1, the MPS was 23.1 nm, which was the smallest MPS. However, when the volume ratio of water phase to oil phase was lower than 1 : 1, the primary GBE nano-emulsion was heterogeneous and layered easily. So the optimal volume ratio of water phase to oil phase was 1 : 1.

Effect of homogenate speed and time

The MPS of primary GBE nano-emulsion with the increased homogenate speed were presented in Fig. 3c, in which the homogenate speed was in the range of 8000–16 000 rpm. It can be seen from Fig. 3c that the MPS of primary GBE nano-emulsion changed from 114.9 nm to 52.2 nm with the homogenate speed increasing from 8000 rpm to 12 000 rpm. When the homogenate speed was greater than 12 000 rpm, the MPS of primary GBE nano-emulsion increased slightly. Therefore, the optimal homogenate speed was fixed to 12 000 rpm.

The effect of the homogenate time on the MPS of primary GBE nano-emulsion was displayed in Fig. 3d. As shown in Fig. 3d, the MPS of primary GBE nano-emulsion decreased from 64.2 nm to 48.4 nm when the homogenate time changed from 1 min to 3 min. While, when the homogenate time was greater than 3 min, the MPS of primary GBE nano-emulsion increased obviously. So the optimal homogenate time was fixed to 3 min.

The above results show that the longer the homogenate time, the faster the homogenization speed, the smaller the size of the nano-emulsion. But after the homogenate time and the homogenization speed reaching a certain value, the MPS of primary GBE nano-emulsion was increased. The reason for the effect of homogenate speed and time may be high speed transfer to the emulsion system of higher energy, the system tends to be unstable state.

From the above results, one can get the optimal experimental parameters in the process of homogenate, which were as following: the concentration of GBE solution was 5 mg mL⁻¹, volume ratio of water phase to oil phase was 1 : 1, homogenate speed was 12 000 rpm and homogenate time was 3 min. Under the optimal conditions, the particle size of the primary GBE nano-emulsion was 48.4 nm.

Effect of homogeneous pressure and times

During the homogeneous process, the effect of the homogeneous pressure on the MPS of GBE nano-emulsion was investigated in this experiment. The MPS of GBE nano-emulsion with the increasing homogeneous pressure was shown in Fig. 4a. It can be seen from Fig. 4a that the MPS of GBE nano-emulsion decreased from 123.9 nm to 57.5 nm with the homogeneous pressure rising from 300 bar to 500 bar. However, when the homogeneous pressure was higher than 500 bar, the MPS of GBE nano-emulsion began to increase. So the optimal value of the homogeneous was 500 bar. That maybe because when the homogeneous pressure is higher than 500 bar, the energy delivering to the system is too large, resulting in instability of the system, the particles of the nano-emulsion occur agglomeration.

Fig. 4 Influence of two parameters on the mean particle size of GBE nano-emulsion during the process of homogenization. (a) Homogenization pressure; (b) homogenization times, and (c) normal distribution map of particle size of primary GBE nano-emulsion and GBE nano-emulsion.

During the homogeneous process, the effect of the homogeneous times on MPS of GBE nano-emulsion was taken into account, and the results were shown in Fig. 4b. It can be seen from Fig. 4b that the MPS of GBE nano-emulsion reduced from 94.3 nm to 67.0 nm with the homogeneous times increasing from 3 times to 5 times. When the homogeneous times was over 3 times, the influence of the homogeneous times on the MPS of GBE nano-emulsion was unobvious. For improving the productivity, the homogeneous times was chosen 3. The MPS of GBE nano-emulsion prepared under the optimal conditions was 67.0 nm. The GBE freeze-drying powder was obtained after the nano-emulsion was processed through rotary evaporation and freeze-drying. GBE freeze-dried powder with was dispersed in the deionized water, and the MPS was detected by dynamic laser light scattering. The MPS of GBE freeze-drying powder without lyoprotectant was more than 1 μm, this is due to the agglomeration of GBE nanoparticles during the process of freeze-drying. After testing and screening, the optimal MPS of GBE freeze-drying powder was 277.0 nm when the HP-β-CD was selected as the lyoprotactant and the mass ration of the HP-β-CD and GBE was 3 : 1. In addition to freeze-drying, atomization/freeze-drying is a commonly used method for the preparation of nanoparticles, which may be worth pursuing in the future study.³⁵

The particle size distribution of primary GBE nano-emulsion and GBE nano-emulsion was shown in Fig. 4c. From Fig. 4c it can be known that the particle size of primary GBE nano-emulsion was nonuniform, resulting in the system been unstable and the oil phase and water phase been easy to be stratified in a short time. The particle size distribution of the GBE nano-emulsion after homogenization was more uniform comparing with that of primary GBE nano-emulsion, and the system was more stable. Through above results can be seen, the MPS of GBE nano-emulsion was larger slightly than that of primary GBE nano-emulsion after homogenization, but the distribution of particle size of GBE nano-emulsion was concentrated, indicating that homogenization was an indispensable step in the prepared process of GBE nanoparticles.

Physicochemical characterization

Morphology analysis. The morphology of raw GBE, GBE nanoparticles and GBE nanoparticle without HP-β-CD was tested by SEM and the results were presented in Fig. 5a–c. As shown in Fig. 5a, the morphology of raw GBE exhibited various block shape, and the size of raw GBE was unequal. From Fig. 5b it can be seen that the GBE nanoparticles with HP-β-CD presented a structure like a net. The HP-β-CD with net structure had a good protective effect on GBE nanoparticles, which preventing the aggregation of nanoparticles. As shown in Fig. 5c, the morphology of GBE nanoparticles without HP-β-CD presented a fluffy and sparse state and nearly spherical, which had a smaller particle size than that of raw GBE.

Fig. 5 Pictures of scanning electron microscopy. (a) Raw GBE; (b) GBE nanoparticles; (c) GBE nanoparticle without HP-β-CD.

Crystal structure. The crystal structure of raw GBE, HP-β-CD, GBE nanoparticles and physical mixture of raw GBE with HP-β-CD was detected via XRD, DSC and TG. The result of XRD was shown in Fig. 6. From Fig. 6a it can be seen that there was no distinct peaks in the diffractogram of the raw GBE, indicating that raw GBE exists as amorphous state. This may be due to the complex composition of the raw GBE. As shown in Fig. 6b–d, there were similar peaks at 2θ = 19.82° for GBE nanoparticles (Fig. 6b) and physical mixture of GBE with HP-β-CD (Fig. 6c) with that of HP-β-CD (Fig. 6d). In the diffractograms of GBE nanoparticles and the physical mixture of GBE with HP-β-CD, there were no other new crystals forming except the HP-β-CD. This result shows that the nanoparticles exist as the amorphous state.

Fig. 6 XRD results of each sample. (a) Raw GBE; (b) GBE nanoparticles; (c) physical mixture of GBE with HP-β-CD; (d) HP-β-CD.

The DSC of raw GBE, HP-β-CD, GBE nanoparticles and the physical mixture of raw GBE with HP-β-CD were presented in Fig. 7. As shown in Fig. 7a, for raw GBE, there was a small absorption peak at 257.02 °C, indicating that there were a few technical components of crystal morphology in raw GBE and the overall state of raw GBE was close to the amorphous state. It can be seen from Fig. 7b–d that the absorption peaks of GBE nanoparticles, the mixture of raw GBE with HP-β-CD, and HP-β-CD were unobvious, and the absorption peaks of GBE nanoparticles and the physical mixture of raw GBE with HP-β-CD were similar to that of HP-β-CD, implying that the GBE in GBE nanoparticles was same as that in physical mixture of raw GBE with HP-β-CD, closed to the amorphous state. This result is consistent with the test results of XRD.

Fig. 7 DSC thermograms of each sample. (a) Raw GBE; (b) physical mixture of GBE with HP-β-CD; (c) GBE nanoparticles; (d) HP-β-CD.

The tested TG curves of raw GBE, HP-β-CD, GBE nanoparticles and the physical mixture of raw GBE with HP-β-CD were presented in Fig. 8, and it can be seen from Fig. 8 that there did not appear suddenly weightless point in the TG of raw GBE. This result implied the raw GBE existed as amorphous state, which was consistent with the result of DSC. The masses of raw GBE, GBE nanoparticles, the mixture of raw GBE with HP-β-CD decreased suddenly at the range of 300 °C to 380 °C, indicating that the existence of HP-β-CD resulted in the sudden decrease of the mass of GBE nanoparticles and the mixture of raw GBE with HP-β-CD. Combined with the results from XRD and DSC, it implied that the raw GBE and GBE nanoparticles existed as the amorphous state.

Fig. 8 TG curves of each sample. (a) Raw GBE; (b) GBE nanoparticles; (c) physical mixture of GBE with HP-β-CD; (d) HP-β-CD.

Solubility study. Fig. 9a showed the results of solubility of GBE in raw GBE, GBE nanoparticles and the physical mixture of raw GBE with HP-β-CD in artificial gastric juice, artificial intestinal juice and deionized water. From the testing results by HPLC can be known, the saturated solubility of GBE in raw GBE, GBE nanoparticles and the mixture of raw GBE with HP-β-CD were 9.78 mg mL⁻¹, 17.33 mg mL⁻¹ and 10.21 mg mL⁻¹ in artificial gastric juice, respectively, whereas that were 7.18 mg mL⁻¹, 17.25 mg mL⁻¹, 9.66 mg mL⁻¹ in artificial intestinal juice and 7.15 mg mL⁻¹, 17.64 mg mL⁻¹, 9.93 mg mL⁻¹ in deionized water, respectively. From the above results can be drawn that the solubility of GBE in GBE nanoparticles was much higher than that of in raw GBE, but that of in the mixture of raw GBE with HP-β-CD had no significant improvement compared with that of in raw GBE, indicating that the increase in solubility of GBE in GBE nanoparticles was not a result of the assistant dissolve effect of HP-β-CD. In previous researches,^36–40 the reasons resulting in the higher solubility of nano drug than the raw mainly have two types: (a) in the process of prepared nanoparticles, the crystal type of drugs changes from crystalline state into amorphous state; (b) the particle size of nanoparticles is much smaller than that of raw, decreasing particle size leads to the increase in specific surface area. According to the Noyes Whitney equation,⁴¹ the solubility of drug will be significantly increased with the increase of the specific surface area. In this study, according to the results of section “Crystal structure”, one can know that before and after the preparation, the drug crystal was close to the amorphous form, so the improvement of solubility was mainly caused by the decreased particle size. In addition, the solubility of GBE in raw GBE, GBE nanoparticles and the mixture of raw GBE with HP-β-CD in artificial gastric juice was higher than that in artificial intestinal juice and deionized water. This is due to artificial gastric juice containing a small amount of Tween, which has the assistant dissolve effect. In addition, the solution state of raw GBE and GBE nanoparticles with equivalent mass in 5 mL deionized water were shown in Fig. 9b, and from Fig. 9b can intuitively see that the dissolved status of GBE nanoparticles was significantly better than that of raw GBE.

Fig. 9 (a) Solubility of GBE in raw GBE, physical mixture of raw GBE with HP-β-CD and GBE nanoparticles: (1) artificial gastric juice; (2) artificial intestinal juice; (3) deionized water. (b) The dissolution state of (1) raw GBE and (2) GBE nanoparticles in deionized water with concentration of GBE was 5 mg mL⁻¹.

Dissolution results. The results of the dissolution for raw GBE and GBE nanoparticles in artificial gastric juice were shown in Fig. 10. From Fig. 10a can be seen that in artificial gastric juice, the release of amount of flavonoids in raw GBE was close to the equilibrium at 12 h and reached to 20.57% at 24 h. While the release of flavonoids in GBE nanoparticles was close to the equilibrium at 12 h and reached to 50.41%, improving 1.45 times comparing with that in raw GBE. As shown in Fig. 10b, it can be found that the release amount of terpene lactones in raw GBE closed to the equilibrium state at 10 h, and the final release was 26.68%. The release amount of terpene lactones in GBE nanoparticle closed to the equilibrium state at 10 h and the final release amount was 33.32%, which was 0.25 times more than that in raw GBE.

Fig. 10 Dissolution profiles of raw GBE, GBE nanoparticles in artificial gastric juice. (a) The accumulative release of total flavonoids; (b) the accumulative release of total terpene lactones.

The dissolution results of raw GBE and GBE nanoparticles in artificial intestinal juice were presented in Fig. 11. It can be seen from Fig. 11a that up to 12 h, the release amount of flavonoids in raw GBE reached to the equilibrium state with the final release of 23.78%. After 12 h, the release amount of flavonoids in GBE nanoparticles reached to a balance level (59.64%), which was increased by 1.51 times comparing with that in raw GBE. From Fig. 11b can be seen that up to 12 h, the release of terpene lactones in raw GBE reached to the equilibrium state with the final release amount of 20.66%. For the terpene lactones in GBE nanoparticles, after 12 h, the release amount reached to the equilibrium state with the final release of 44.11%, which was increased by 1.14 times comparing with that in raw GBE.

Fig. 11 Dissolution profiles of raw GBE, GBE nanoparticles in artificial intestinal juice. (a) The accumulative release of total flavonoids; (b) the accumulative release of total terpene lactones.

In addition, the dissolution results of raw GBE and GBE nanoparticles in deionized water were presented in Fig. 12. From Fig. 12a can be seen that in deionized water, after 4 h, the release amount of flavonoids closed to the equilibrium state (20.61%). Meanwhile, the release amount of flavonoids in GBE nanoparticles in deionized water reached to the maximum value of 78.53% at 6 h. It was noting that the release amount of ginkgo flavones in GBE nanoparticles decreased comparing with that in 6 h, and tended to 71.46%, which was 2.47 times higher than that in raw GBE. As shown in Fig. 12b, the dissolution rate of terpene lactones in raw GBE, after 8 h, the release tended to the equilibrium state with the final release of 29.56%. For the terpene lactones in GBE nanoparticles, the release almost reached to equilibrium state with release of 50.62% at 12 h, which was 0.71 times higher than that in raw GBE.

Fig. 12 Dissolution profiles of raw GBE, GBE nanoparticles in deionized water. (a) The accumulative release of total flavonoids; (b) the accumulative release of total terpene lactones.

In the three systems, the saturated solubility and dissolution of ginkgo flavones and terpene lactones in GBE nanoparticles are higher than that in raw GBE. This because the particle size of GBE nanoparticles is much smaller than that of raw GBE, and the decrease of particle size leads to an increase in the surface area, which increases the solubility and dissolution.

Results from the results above, it can be known, in the three systems, due to the poor solubility of raw GBE, the release amount of ginkgo flavones and terpene lactones in vitro dissolution shows unobvious. But for GBE nanoparticles with outstanding water solubility, the release amount of ginkgo flavones and terpene lactones in vitro dissolution realizes a significant difference, in which the release amount of ginkgo flavones in artificial gastric juice shows the minimum and the maximum in deionized water, owing to the different pH of three systems. The dissolution rate of ginkgo flavones and terpene lactones could be inhibited under the acidic condition. Therefore, the release amount of ginkgo flavones and terpene lactones increases with the elevated pH in these three systems.

Bioavailability analysis. The plasma concentration of ginkgo flavones in plasma for reference group and experiment group were presented in Fig. 13a. As shown in Fig. 13a, the charts appeared two peaks for the two groups. For the reference group (raw GBE), the first peak appeared at 0.5 h with plasma concentration of 3.80 μg mL⁻¹ and the second peak at 4 h with plasma concentration of 2.70 μg mL⁻¹. For the experiment group (GBE nanoparticles), the first peak was located at 0.083 h with plasma concentration of 8.34 μg mL⁻¹ and the second peak at 0.5 h with plasma concentration of 6.79 μg mL⁻¹.

Fig. 13 Plasma concentration of raw GBE, GBE nanoparticles. (a) The plasma concentration of total flavonoids; (b) the plasma concentration of total terpene lactones.

Moreover, the drug concentration of terpene lactones in plasma for reference group and experiment group were presented in Fig. 13b. It can be seen from Fig. 13b that the charts appeared three peaks for the two groups. For the reference group, the first peak appeared at 1 h with plasma concentration of 0.11 μg mL⁻¹, and the second peak was located at 4 h with plasma concentration of 0.09 μg mL⁻¹. In addition, the third peak lied at 8 h with plasma concentration of 0.05 μg mL⁻¹. For the experiment group, the first peak appeared at 0.167 h with plasma concentration of 0.18 μg mL⁻¹ and the second peak was located at 1 h with plasma concentration of 0.13 μg mL⁻¹. Meanwhile, the third peak lied at 4 h with plasma concentration of 0.12 μg mL⁻¹.

From the above results, one can find that the detected plasma concentration of ginkgo flavones and terpene lactones in the rats of experiment group were higher than that in the rats of reference group, and the absorption rate to ginkgo flavones and terpene lactones of the rats in the experiment group was faster than that in the reference group. This is because the particle size of GBE nanoparticles is much smaller than that of raw GBE, which enhances the water solubility of GBE nanoparticles, thereby improving the absorption effect and rate in the rats. Meanwhile, the plasma concentration curves of ginkgo flavones and terpene lactones in raw GBE and GBE nanoparticles appearing two peaks owes to the ginkgo flavones are mainly composed of quercetin, kaempferol and isorhamnetin, and the absorption rate to these three components is different in vivo. When a component has been absorbed to reach the maximum value, the other ingredients may just began to absorb, when the absorption of first component declines, that of the other components reaches maximum, leading to the two peaks in the plasma concentration curves. For the same reason, terpene lactones contain the components of ginkgolide A, ginkgolide B, ginkgolide C and bilobalide, the absorption rate of these substances is different in vivo, leading to the three peaks occur in the plasma concentration of the rats. The oral relative bioavailability of GBE nanoparticles was calculated by the way of comparing the corresponding AUC values of the two groups. The result demonstrated that the AUC values of the flavonoids and terpene lactones in GBE nanoparticles was about 1.81 and 1.75 times compared with raw GBE, respectively. Therefore, the oral bioavailability of the GBE nanoparticles was improved significantly compared with the raw GBE.

From the above results, the method of ESE-FR achieved good results in improving the oral bioavailability of GBE. In addition to GBE, this method also significantly improved the oral bioavailability of genipin,²⁵ 10-hydroxycamptothecin,²⁶ poly(lactide-co-glycolide) (PLG)²⁷ and puerarin.²⁸ In recent years, preparation of GBE solid dispersion via hot-melt extrusion¹⁴ and phospholipid complexes¹⁷ had improved its oral bioavailability obviously. However, comparing with the method of ESE-FR, the operation of preparing GBE phospholipid complexes and solid dispersion are very complex and not conducive to the expansion of production scale. These two methods cost much higher than the method of ESE-FR. In addition, the preservation of GBE phospholipid complexes is difficult. In the process of preparing GBE solid dispersion by hot melt extrusion, due to the difference between the melting point of the drug and the auxiliary materials, the concentration distribution of solid dispersion is difficult to ensure uniformity.

The detection of residual solvents. In the actual production process, the residual solvent is also considered in drug production. In this work, the GBE nanoparticles were prepared successfully by using emulsion solvent evaporation method, which used the International Conference on Harmonization (ICH) class II solvent ethanol and chloroform with low toxicity. Fig. 14a and b showed the chromatograms results of 10 mg mL⁻¹ ethanol solution of the GBE nanoparticles and 0.05 mg mL⁻¹ chloroform of ethanol solution, respectively. From Fig. 14a and b it can be seen that, there were two solvent peaks, the retention time of chloroform was 3.130 min after confirmation. Meanwhile, from Fig. 14b, the peak of chloroform was significantly smaller than that of chloroform in Fig. 14a, the percentages of residual chloroform was about 0.00217% in GBE nanoparticles. The residual chloroform content in GBE nanoparticles was less than the ICH limit for chloroform in class II solvents of 0.006% for solvents.

Fig. 14 Gas chromatograms of (a) chloroform standard solution; (b) GBE nanoparticles; (c) ethanol standard solution; (d) GBE nanoparticles.

Fig. 14c and d presented the chromatograms results of 0.05 mg mL⁻¹ ethanol of deionized water solution and 10 mg mL⁻¹ deionized water solution of the GBE nanoparticles, respectively. In Fig. 14c, there was a peak at retention time of 2.192 min, which represented ethanol after confirmation. From Fig. 14d, the peak of ethanol was significantly smaller than that of ethanol in Fig. 14c, the percentages of residual ethanol was about 0.07% in GBE nanoparticles. The residual ethanol content in GBE nanoparticles was less than the ICH limit for ethanol in class II solvents of 0.5% for solvents. GBE nanoparticles fully meet the requirements of pharmaceutical use.

Conclusions

By using the method of ESE-FR, the GBE nanoparticles were prepared successfully. By means of a single factor method, the optimal parameters were as follows: the GBE concentration was 5 mg mL⁻¹, volume ratio of water to organic phase was 1, homogenate speed was 12 000 rpm, homogenate time was 3 min, homogeneous pressure and times were 500 bar and 3 times, respectively. According to pre-experimental screening result, the HP-β-CD was selected as the lyoprotectant, and the mass ration of the HP-β-CD and GBE in nano-suspension was 3 : 1. The particle size of prepared GBE nanoparticles was 56 nm under the optimal parameters. One has come to the conclusion by means of the following test: (a) the detection of XRD, DSC and TG shows that the GBE nanoparticles have the same crystal type with raw GBE. (b) By detecting the solubility, dissolution in vitro and bioavailability, it can be known that the saturated solubility of GBE nanoparticles is 1.77 times, 2.47 times and 2.40 times of raw GBE in artificial gastric juice, artificial intestinal juice and deionized water, respectively. And the release percentage of GBE nanoparticles in vitro has been greatly improved compared with that of raw GBE. By detecting the bioavailability, the plasma concentration of GBE nanoparticles in the rats is much higher than that of raw GBE, and the absorption rate of GBE nanoparticles in the rats is significantly accelerated comparing with that of raw GBE. (c) The results of solvent residue show that the residual amounts of chloroform and ethanol are 0.00217% and 0.07%, respectively, completely consistent with the requirements of ICH. In a word, the method of ESE-FR has great potential value to the preparation of oral GBE drugs.

Acknowledgements

The authors are grateful for the precious comments and careful corrections made by anonymous reviewers. The authors would also like to acknowledge the financial support from the National Key Technology R&D Program (2012BAD21B0403) and National Natural Science Foundation of China (no. 21473023).

References

1. R. T. Major, Science, 1967, 157, 1270–1273.
2. K. Stromgaard and K. Nakanishi, Angew. Chem., Int. Ed., 2004, 43, 1640–1658.
3. J. Kleijnen and P. Knipschild, Lancet, 1992, 340, 1136–1139.
4. P. Ren, X. Zhao, J. Zhang, R. Shi, Z. Yuan and C. Wang, React. Funct. Polym., 2008, 68, 899–909.
5. B. Singh, P. Kaur, G. R. D. Singh and P. S. Ahuja, Fitoterapia, 2008, 79, 401–418.
6. K. Nakanishi, Bioorg. Med. Chem., 2005, 13, 4987–5000.
7. F. Mahmoud, H. Abul, B. Onadeko, M. Khadadah, D. Haines and G. Morgan, Jpn. J. Pharmacol., 2000, 83, 241–245.
8. A. Louajri, S. Harraga, V. Godot, G. Toubin, J. P. Kantelip and P. Magnin, Biol. Pharm. Bull., 2001, 24, 710–712.
9. C. Zhang, Y. Zhu, J. Wan, H. Xu, H. Shi and X. Lu, Liver Int., 2006, 26, 1283–1290.
10. C. Shi, J. Liu, F. Wu and D. T. Yew, Int. J. Mol. Sci., 2010, 11, 107–123.
11. Y. Jin, J. Y. Wen, S. Garg, D. Liu, Y. L. Zhou, L. R. Teng and W. Y. Zhang, Int. J. Nanomed., 2013, 8, 421–430.
12. Y.-k. Deng, F. Wei and B.-q. An, Zhongguo Zhongxiyi Jiehe Zazhi, 2006, 26, 316–318.
13. F. Chen, L. Li, F. Xu, Y. Sun, F. Du, X. Ma, C. Zhong, X. Li, F. Wang, N. Zhang and C. Li, Br. J. Pharmacol., 2013, 170, 440–457.
14. W. Wang, Q. Kang, N. Liu, Q. Zhang, Y. Zhang, H. Li, B. Zhao, Y. Chen, Y. Lan, Q. Ma and Q. Wu, Fitoterapia, 2015, 102, 189–197.
15. J. Tang, J. Sun, F. Cui and Z. He, Asian J. Tradit. Med., 2006, 138–141.
16. J. L. Tang, J. Sun, F. D. Cui, T. H. Zhang, X. H. Liu and Z. G. He, Drug Delivery, 2008, 15, 477–484.
17. Z.-p. Chen, J. Sun, H.-x. Chen, Y.-y. Xiao, D. Liu, J. Chen, H. Cai and B.-c. Cai, Fitoterapia, 2010, 81, 1045–1052.
18. E. M. Merisko-Liversidge and G. G. Liversidge, Toxicol. Pathol., 2008, 36, 43–48.
19. A. Nokhodchi, N. Bolourtchian and R. Dinarvand, Int. J. Pharm., 2003, 250, 85–97.
20. Y. Zu, W. Wu, X. Zhao, Y. Li, C. Zhong and Y. Zhang, Int. J. Pharm., 2014, 477, 148–158.
21. C. Taniguchi, Y. Kawabata, K. Wada, S. Yamada and S. Onoue, Expert Opin. Drug Delivery, 2014, 11, 505–516.
22. Y. K. Choi, B. K. Poudel, N. Marasini, K. Y. Yang, J. W. Kim, J. O. Kim, H.-G. Choi and C. S. Yong, Int. J. Pharm., 2012, 434, 264–271.
23. H.-K. Chan and P. C. L. Kwok, Adv. Drug Delivery Rev., 2011, 63, 406–416.
24. D.-J. Jang, S. T. Kim, E. Oh and K. Lee, Bio-Med. Mater. Eng., 2014, 24, 917–930.
25. Y. Zu, X. Yu, X. Zhao, W. Wang and K. Wang, J. Nanomater., 2014 DOI:10.1155/2014/240950.
26. L. M. Yuangang Zu, X. Zhao, Y. Ge, X. Yu, Y. Zhang and Y. Deng, Int. J. Nanomed., 2013, 8, 1207–1222.
27. Y. Chernyak, F. Henon, R. B. Harris, R. D. Gould, R. K. Franklin, J. R. Edwards, J. M. DeSimone and R. G. Carbonell, Ind. Eng. Chem. Res., 2001, 40, 6118–6126.
28. Y. Zhang, Y. Li, X. Zhao, Y. Zu, W. Wang, W. Wu, C. Zhong, M. Wu and Z. Li, RSC Adv., 2016, 6, 69889–69901.
29. S. F. Chow, K. Y. Wan, K. K. Cheng, K. W. Wong, C. C. Sun, L. Baum and A. H. L. Chow, Eur. J. Pharm. Biopharm., 2015, 94, 436–449.
30. D. E. Gray, D. Messer, A. Porter, B. Hefner, D. Logan, R. K. Harris, A. P. Clark, J. A. Algaier, J. D. Overstreet and C. S. Smith, J. AOAC Int., 2007, 90, 1203–1209.
31. C. J. Zhao, L. Wang, Y. G. Zu, C. Y. Li, S. H. Liu, L. Yang, X. H. Zhao and B. S. Zu, Powder Technol., 2011, 209, 73–80.
32. J. Xie, C. Ding, Q. Ge, Z. Zhou and X. Zhi, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2008, 864, 87–94.
33. Chinese Pharmacopoeia Commission, Chinese Pharmacopoeia, Beijing, China Medical Science Press, 2010.
34. D. Yang, X.-Y. Wang, L.-J. Gan, H. Zhang, J.-A. Shin, K.-T. Lee and S.-T. Hong, Food Chem., 2015, 174, 124–131.
35. M. Z. Yates, W. Yin, X. Yang and M. Nair, Process for preparing polymer particles containing metallic flakes, US Pat., US20110262654, 2011.
36. Y. Gao and K. W. Olsen, Mol. Pharmaceutics, 2013, 10, 905–917.
37. F. Lai, C. Sinico, G. Ennas, F. Marongiu, G. Marongiu and A. M. Fadda, Int. J. Pharm., 2009, 373, 124–132.
38. T. Zhang, Y. Yang, H. Wang, F. Sun, X. Zhao, J. Jia, J. Liu, W. Guo, X. Cui, J. Gu and G. Zhu, Cryst. Growth Des., 2013, 13, 5261–5266.
39. E. Myllykyla, T. Lavonen, M. Stennett, C. Corkhill, K. Ollila and N. Hyatt, Radiochim. Acta, 2015, 103, 565–576.
40. T. Thao Truong-Dinh, T. Kiet Anh and T. Phuong Ha-Lien, Ultrason. Sonochem., 2015, 24, 256–263.
41. D. Aristides, P. Vasiliki and M. Panos, Pharm. Res., 2006, 23, 256–261.

---