TPGS Emulsified Zein Nanoparticles Enhanced Oral Bioavailability of

Shutt , D. A.; Cox , R. I. Steroid and phyto-oestrogen binding to sheep uterine receptors in vitro J. Endocrinol. 1972, 52 (2) 299– 310. [Crossref],...
0 downloads 0 Views 422KB Size
Article pubs.acs.org/molecularpharmaceutics

TPGS Emulsified Zein Nanoparticles Enhanced Oral Bioavailability of Daidzin: In Vitro Characteristics and In Vivo Performance Tao Zou and Liwei Gu* Food Science and Human Nutrition Department, Institute of Food and Agricultural Sciences, University of Florida, Gainesville, Florida 32611, United States S Supporting Information *

ABSTRACT: A novel drug delivery system, TPGS 1000 (TPGS) emulsified zein nanoparticles (TZN), were designed with an objective to improve the oral bioavailability of daidzin, an isoflavone glycoside with estrogenic activities. Zein nanoparticles (ZN) and TZN were fabricated using an antisolvent method. They were found to be spherical in shape with a mean size around 200 nm and a low polydispersity. Their zeta potentials were about +25 mV at pH 5.5 and −23 mV at pH 7.4. Adding TPGS as an emulsifier increased the encapsulation efficiency of daidzin in ZN from 53% to 63%. Daidzin loaded TZN had a slower daidzin release compared with daidzin loaded ZN in both simulated digestive fluids and a pH 7.4 buffer. Confocal laser scanning microscopy suggested that the cellular uptake of coumarin-6 labeled TZN in human intestinal epithelial Caco-2 cells were significantly higher than fluorescent ZN. Cellular uptake and transport studies revealed that daidzin in TZN were taken up more efficiently into Caco-2 cells and transported more quickly through Caco-2 monolayer than daidzin solution. A pharmacokinetic study demonstrated that the Cmax of daidzein in mice after oral administration of daidzin loaded TZN was 5.66 ± 0.16 μM, which was improved by 2.64-fold compared with that of daidzin solution (2.14 ± 0.04 μM). Moreover, the areas under the curve (AUC0−12 h) for daidzin loaded in TZN were enhanced by 2.4-fold compared with that of daidzin solution. These results suggested that TZN could be an effective strategy to improve the oral bioavailability of isoflavone glycosides like daidzin. KEYWORDS: TPGS, zein, isoflavone, daidzin, nanoparticles, phytoestrogen



INTRODUCTION Isoflavone 7-glucosides, genistin and daidzin, as well as their aglycones (genistein and daidzein) have been reported to have a variety of biological activities, including estrogenic,1 antioxidative,2,3 antiosteoporotic,4 and anticarcinogenic.5 Over 90% of isoflavones naturally exist as glycosides.6,7 Only a small portion exists as aglycones. Isoflavones have poor oral bioavailability due to two reasons. First, isoflavone glycosides or aglycones are absorbed via passive diffusion through intestinal epithelium because there are no active transporter for them.8 The diffusion rate and permeability of isoflavone glycosides on intestinal epithelial membrane are lower than aglycones.8 Compared with aglycones, isoflavone glycosides have a much lower adsorption rate9 or are not absorbed at all.10,11 Second, absorbed isoflavone aglycones are transformed into sulfates and glucuronides via phase II metabolism.10 These conjugates are subsequently pumped back into the intestinal lumen by the multidrug resistance protein MRP2 and P-glycoproteins (collectively called efflux pumps) in membranes of intestinal epithelium.11 The goal of this research is to enhance the absorption of isoflavone glycosides by overcoming the above obstacles using nanotechnology. First, we aim to improve the transport of isoflavone glycosides through intestinal epithelium by loading them into nanoparticles. Nanoparticles can improve the transport of © XXXX American Chemical Society

drugs with poor water solubility and/or low permeability to epithelial membrane by endocytosis.12 Zein is a plant protein and a food ingredient. Due to its inherent biodegradability and biocompatibility, zein nanoparticles were successfully applied as a carrier for controlled release of drugs or bioactive compounds (e.g., metformin, curcumin, cranberry procyanidins, etc.) in drugs and dietary supplements.13−15 Second, we will explore adding TPGS 1000 (TPGS) to inhibit the efflux of isoflavone conjugates. TPGS is a food additive approved by FDA and can be used as a reversing agent for P-glycoproteins (P-gp) mediated multidrug resistance.16 ATPase inhibition was found to be an inhibitory mechanism of TPGS 1000 on cellular P-gp efflux pumps.17 The bioavailability of paclitaxel was enhanced by 6.3-fold when it was administrated with TPGS compared with paclitaxel administrated alone.18 TPGS is more effective than verapamil, another known P-gp inhibitor.18,19 TPGS-PLGA (poly(lactic-co-glycolic acid)) nanoparticles have been explored for the controlled release of paclitaxel.20 Compared with PLGA, zein is safer, more biocompatible, and less expensive as a carrier. Received: February 15, 2013 Revised: March 22, 2013 Accepted: April 4, 2013

A

dx.doi.org/10.1021/mp400086n | Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Molecular Pharmaceutics

Article

TPGS in the nanoparticles. The sediments were washed three times by water and lyophilized for 48 h to form dry particles and stored at −20 °C for further analyses. All samples were carefully protected from light throughout the experimental procedure. Particle Size and Zeta Potential Measurement. ZN and TZN were suspended in deionized (DI) water. The pH was adjusted to 5.5 or 7.4 using 0.01 mol/L HCl or NaOH solution, respectively. Particle size and zeta potential were measured using a Zetatrac (Microtrac Inc., Largo, FL). Each sample was analyzed in triplicate, and each replicate was measured three times to yield the average particle size and zeta potential. Scanning Electron Microscopy (SEM). The dry particle powders were used for morphology characterization using a field emission scanning electron microscope (model JSM633OF, JEOL Ltd., Tokyo, Japan). Particle Yield, Encapsulation Efficiency, and Content of TPGS in the Nanoparticles. The particle yield was calculated as (weight of dry particles)/(total weight of TPGS, zein, and daidzin or coumarin-6 used for particle preparation) × 100%. The content of TPGS in the nanoparticles was determined on an Agilent 1200 HPLC system using an Agilent C18 column (4.6 mm × 25 cm, 5 μm particle size). The binary mobile phase consisted of (A) acetonitrile and (B) isopropanol. The gradient reported by Kong et al. was used.21 The analysis was carried out at 35 °C with ultraviolet detection at 284 nm. A portion of 10 μL of each sample was injected into the HPLC system. The content of TPGS in the nanoparticles was calculated as (amount of TPGS in the nanoparticles)/(amount of dry nanoparticles) × 100%. The encapsulation efficiency of daidzin-loaded nanoparticles was determined on HPLC using an Agilent C18 column (4.6 mm × 25 cm, 5 μm particle size). The binary mobile phase consisted of (A) 0.2% (v/v) formic acid and (B) acetonitrile. The gradient reported by Klejdus et al. was used.22 The analysis was carried out at 40 °C with UV detection at 254 nm. A portiong of 10 μL of each sample was injected into the HPLC system. The encapsulation efficiency of coumarin-6 was determined by a Gemini XPS fluorescence microplate reader (Molecular Devices, California, λex 430 nm, λem 485 nm). The encapsulation efficiency was calculated as (amount of daidzin or coumarin-6 loaded in the nanoparticles)/(initial amount of daidzin or coumarin-6 used in the fabrication of the nanoparticles) × 100%. All measurements were done in triplicate. Release of Daidzin from the Nanoparticles in Simulated Gastric and Intestinal Fluids. The daidzin loaded ZN or TZN were incubated in 30 mL simulated gastric fluid (SGF) with 0.1% pepsin (w/v) (pH approximately 1.2) or simulated intestinal fluid (SIF) with 1.0% pancreatin (w/v) (pH 6.8) at 37.1 °C for 0.5 or 2 h, respectively. Then 1 mL of sample was centrifuged at 16 000 g for 5 min after incubation. The content of daidzin and daidzein in the supernatant was analyzed on HPLC, and the percentage of released daidzin in SGF or SIF was calculated. In Vitro Release at pH 7.4. In vitro release of the coumarin6 or daidzin loaded nanoparticles was carried out in a buffer (Hank’s balanced salt solution, HBSS, pH 7.4) at 37.1 °C. The 1 mg/mL nanoparticle suspension (5 mL) in a dialysis bag was incubated in 25 mL of HBSS solution in a water bath at 37.1 °C. A sample from the incubated solution was collected at designated time intervals, and an equal volume of fresh HBSS was replenished. The amount of coumarin-6 or daidzin in

In this study, TZN were fabricated to enhance the absorption of daidzin. The nanoparticles were then characterized by a particle size−zeta potential analyzer, high-performance liquid chromatography, and scanning electronic microscopy (SEM). The release of daidzin from the nanoparticles in simulated digestive fluids and physiological condition (pH 7.4 buffer) was evaluated. The cellular uptake of the fluorescent nanoparticles and the uptake and transport of the daidzin loaded nanoparticles were investigated on human intestinal epithelial Caco-2 cell monolayers. The pharmacokinetics of daidzin metabolites (daidzein and equol) was measured in mice after oral administration of daidzin loaded in TZN to assess the oral bioavailability and compare with that of daidzin solution.



MATERIALS AND METHODS Materials. Daidzin (purity 98.9%) was purchased from Chengdu Biopurify Phytochemicals Ltd. (Chengdu, China). Daidzein (purity 99%) was purchased from Indofine Chemical Company (Hillsborough, NJ). (R,S)-Equol (purity 99%) was purchased from Alfa Aesar Chemical Company (Ward Hill, MA). Purified rodent diet free from flavonoids (D10012G) was purchased from Research Diets Inc. (New Brunswick, NJ). TPGS 1000 was kindly provided by ChayseChem Inc. (Yardley, PA). Sulfatase type H-1 from Helix pomatia was a product of Sigma-Aldrich (St. Louis, MO). Zein, coumarin-6, carboxymethyl cellulose sodium salt, simulated gastric fluid, simulated intestinal fluid, and other chemicals were purchased from Fisher Scientific (Pittsburgh, PA). Caco-2 cells were obtained from American Type Culture Collection (Manassas, VA). Animals. The in vivo experiment was carried out under the guideline approved by the Institutional Animal Care and Use Committee (IACUC) of University of Florida. Male CD-1 IGS mice (22−24 g) were purchased from Charles River Laboratories International, Inc. (Wilmington, MA). The mice were housed in the animal facility of Animal Care Service of University of Florida and acclimated for seven days using a purified diet free from flavonoids (D10012G) prior to the experiment. Preparation of ZN and TZN. ZN and TZN were prepared by a modified antisolvent method. Briefly, coumarin-6 or daidzin and zein were respectively dissolved in 2.5 mL of binary solvent mixture (ethanol−water at 80:20 v/v) to form a stock solution. Then 7.5 mL of aqueous solution with or without TPGS was added to the stock solution dropwise under stirring at 1000 rpm. The mixed solution was stirred for another 30 min. The different formulations that were used to fabricate nanoparticles are shown in Table 1. The nanoparticle Table 1. Different Formulations Used for Nanoparticle Preparationa samples

zein (mg/mL)

TPGS (mg/mL)

coumarin-6 (mg/mL)

daidzin (mg/mL)

A1 A2 B1 B2 C1 C2

1 1 1 1 1 1

0 0.75 0 0.75 0 0.75

0 0 0.0025 0.0025 0 0

0 0 0 0 0.1 0.1

a

Concentrations in mg/mL are the final content in the suspension.

suspensions were then centrifuged at 12 000 g for 5 min. The supernatant was removed for calculations of the encapsulation efficiency of coumarin-6 and daidzin, as well as the content of B

dx.doi.org/10.1021/mp400086n | Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Molecular Pharmaceutics

Article

the Caco-2 monolayers was measured above 700 Ω × cm2 using an epithelial volt-ohm meter (Millicell ERS-2, Millipore Corp, Billerica, MA), cellular uptake and transport was carried out in triplicates as previously described.8 Briefly, the inserts were washed with HBSS and equilibrated for 30 min in the incubator. Then 10 mmol/L of daidzin in dimethyl sulfoxide was diluted with HBSS, and the final concentration of daidzin was adjusted to 10 μmol/L. Samples of 1.5 mL of daidzin and daidzin loaded TZN at the same daidzin concentration of 10 μmol/L in HBSS were added on the apical side of Caco-2 monolayers, and 2.6 mL of HBSS was added to the basolateral chamber. After incubation for 0.5, 1.0, and 2.0 h at 37 °C, the TEER of the monolayer was measured. Subsequently, the apical and the basolateral solutions were collected. Cellular extracts were prepared by incubating the whole inserts with methanol for 30 min. The apical and basolateral solutions and the cellular extracts were each divided into two aliquots and evaporated using a SpeedVac concentrator (ISS110, Thermo Scientific). One aliquot of apical solution, basolateral solution or cellular extracts was mixed with 20 units sulfatase type H-1 in 100 mmol/L acetate buffer (0.1 mL, pH 5.0) and incubated at 37 °C for 45 min. Subsequently, the same volume of methanol was added to the mixture and centrifuged at 12 000 × g for 10 min. The resultant supernatant solutions were used as sulfatase-treated samples. The second aliquot was dissolved and used as untreated samples. Since sulfatase type H-1 possesses sulfatase and glucuronidase activities, the amounts of the metabolites (including glucuronides and sulfates) were calculated by subtracting the amounts of daidzein and daidzin obtained in untreated samples from that of total daidzein in sulfatase-treated samples. Other metabolized forms, such as methylated forms, were not identified in this study. The samples were analyzed with HPLC using a published method with minor modifications.22 An Agilent Zorbax SB C18 column (4.6 mm × 250 mm) was used. The binary mobile phase consisted of 0.2% (v/v) formic acid: water (A) and acetonitrile (B). A linear gradient profile from 12 up to 22% B (v/v) from 0 to 20 min, up to 50% B to 25 min, up to 55% to 30 min, and down to 12% B at 35 min was used for separation. The injection volume was 100 μL with a flow rate of 1 mL/min. The temperature of the column was set at 40 °C. Isoflavones were detected at 254 nm. Daidzin and daidzein were used as an external standard. The apparent permeability coefficient (Papp) of daidzin was calculated using the following equation:

HBSS was analyzed on a fluorescence microplate reader or HPLC. Cell Culture. Caco-2 cells of passages between 35 and 40 were used. Caco-2 cells were cultured in Dulbecco’s modified Eagle’s media containing 20% fetal bovine serum (FBS), 1% penicillin−streptomycin, and 1% nonessential amino acids. Cells were cultured at 37.1 °C in a humidified atmosphere containing 5% CO2, and the medium was replenished every other day. Confocal Laser Scanning Microscopy. Caco-2 cells were seeded at a level of 1 × 105 cells/cm2 in a four-well covered glass chamber (Lab-Tek, Nalge Nunc, IL, USA). Cells were cultured to about 70% confluence. On the day of the experiment, the growth medium was replaced by HBSS (Hank’s balanced salt solution, HBSS, pH 7.4). After equilibration with HBSS at 4 or 37.1 °C for 30 min, the buffer was replaced with coumarin-6 loaded ZN or TZN suspension (200 μg/mL in HBSS), and then the cells were further incubated for 2 h. At the end of experiment, the cells were washed 3 times with fresh prewarmed HBSS to remove excess nanoparticles. Cells were then fixed with 200 μL of 70% ethanol for 20 min. The cells were further washed with phosphate-buffered saline (PBS, pH 7.4) three times, and the nuclei were counterstained with propidium iodide (PI) (300 μL 500 nM) for 5 min. The fixed cell was finally washed with PBS for three times and examined using a confocal laser scanning microscopy (Zeiss Pascal LSM 5 confocal laser scanning microscopy). Uptake of Fluorescent Marker Loaded Nanoparticles. Caco-2 cells were seeded in 96-well plates (Costar, IL, USA) and incubated until they formed a confluent monolayer. Upon reaching confluence, the culture medium was replaced by HBSS and preincubated at 4 or 37.1 °C for 30 min. After equilibration, cellular uptake of nanoparticles was initiated by exchanging the transport medium with 100 μL of coumarin-6 loaded ZN or TZN suspension at 200 μg/mL in HBSS for different incubation times of 0.5, 1, 2, or 4 h. The cellular uptake analysis was also carried out at different nanoparticle concentrations of 100, 200, 400, and 800 μg/mL at 2 h. For each sample, a total of 16 wells in two columns were used. The first column was kept intact after incubation as a positive control, which represented the fluorescence intensity associated with total amount of the nanoparticles. The second column was used as a sample for investigation. After the incubation and removal of the nanoparticle suspension, cells in the sample columns were washed three times with 0.1 mL PBS (4 °C) to remove excess particles. Cell membrane was disrupted with 50 μL of 0.5% Triton X-100 in 0.2 mol/L NaOH solution to expose the internalized nanoparticles for the quantitative measurement. Cell-associated nanoparticles were quantified by analyzing the cell lysate on a fluorescence microplate reader (λex 430 nm, λem 485 nm). The fluorescence intensity of the wells with the cells alone represented the background intensity and was used as a negative control. The efficiency of cellular uptake of the nanoparticles was then expressed as (I: fluorescence intensity):

Papp = (dQ /dt ) × (1/AC0)

where dQ/dt is the permeability rate (μmol/s), C0 is the initial concentration in the donor chamber (μmol/mL), and A is the surface area of filter (cm2), which was 4.67 cm2 in this study. In Vivo Pharmacokinetics of Daidzin Loaded TZN in Mice. Mice (n = 52) were fasted overnight (8 pm to 8 a.m.) with free access to water before pharmacokinetic studies. The daidzin solution (daidzin dispersed in 0.5% carboxymethyl cellulose sodium salt solution) and daidzin loaded TZN (5 mg/mL) were administrated to mice by oral gavage (50 mg daidzin/kg), respectively. Mice had free access to food and water after dosing. At different time intervals (0, 0.5, 1, 2, 4, 8, and 12 h), mice were anesthetized using isoflurane inhalation. Blood samples were drawn by cardiac puncture and collected into heparinized tubes at each time point. Blood was centrifuged immediately at 3000 rpm for 10 min at 4 °C to separate plasma. Plasma was stored at −80 °C before analysis.

uptake efficiency% = (Isample − Inegative)/(Ipositive − Inegative) × 100%

Cellular Uptake and Transport of Daidzin Loaded Nanoparticles by Caco-2 Cells. Caco-2 cells were seeded and cultured in 6-well transwell plates (Corning Inc. Corning, NY). When the transepithelial electrical resistance (TEER) of C

dx.doi.org/10.1021/mp400086n | Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Molecular Pharmaceutics

Article

Table 2. Particle Size, Zeta Potential, Particle Yield, and Encapsulation Efficiency of Nanoparticles Prepared Using Different Formulation and Suspended in DI Water at pH 5.5a samples A1 A2 B1 B2 C1 C2 a

particle size (nm) 159.7 182.4 170.7 191.3 185.1 210.6

± ± ± ± ± ±

8.9c 14.3b 15.3bc 6.5ab 5.1b 19.2a

polydispersity index 0.207 0.106 0.266 0.154 0.216 0.141

± ± ± ± ± ±

zeta potential (mV)

encapsulation efficiency (%)

± ± ± ± ± ±

NA NA 34.4 ± 3.42c 35.87 ± 2.05c 52.67 ± 3.51b 63.33 ± 5.03a

0.040ab 0.043b 0.124a 0.095ab 0.033ab 0.067ab

27.69 25.03 26.76 24.08 26.48 22.16

1.03a 0.81bc 0.83a 0.82c 0.82ab 1.10d

particle yield (%) 87.7 72.0 85.1 67.7 76.1 62.3

± ± ± ± ± ±

2.5a 3.7bc 3.5a 4.0cd 7.1b 3.8d

Data are mean ± standard deviation for triplicate tests. Means within a column followed by the different letter are significantly different at p ≤ 0.05.

A sample of 200 μL of plasma in ammonium acetate buffer (1 M, 1.8 mL, pH 6.0) was mixed with 100 U of a sulfatase type H-1 solution in ammonium acetate buffer (1 M, 1 mL, pH 6.0) and incubated at 37 °C for three hours. The sample after enzymatic hydrolysis was acidified with 100 μL of glacial acetic acid and was defatted with 5 mL of hexane. The hexane fraction was discarded by pipetting. Residual hexane on the top of the aqueous phase was dried using Speedvac for 15 min. Ethyl acetate (5 mL) was added to extract daidzein and equol. This was repeated one more time, and two ethyl acetate extracts were combined and dried using Speedvac. The dried extracts were reconstituted in 0.2 mL of 50% methanol/water prior to HPLC-MS analysis. The samples were analyzed with HPLC-MS using a published method with minor modifications.22 An Agilent Zorbax SB-C18 column (4.6 mm × 250 mm) was used. The binary mobile phase consisted of 0.2% (v/v) formic acid: water (A) and acetonitrile (B). A linear gradient profile from 12 up to 22% B (v/v) from 0 to 20 min, up to 50% B to 25 min, up to 55% to 30 min, and down to 12% B at 35 min was used for separation. The injection volume was 100 μL with a flow rate of 1 mL/min. The temperature of the column was set at 40 °C. Electrospray ionization in negative mode was performed using nebulizer 50 psi, drying gas 10 L/min, drying temperature 350 °C, and capillary 4000 V. The SIM (Selected Ion Monitoring) mode was used to monitor the ions (M − H)− of m/z 241 (equol) and 253 (daidzein). Daidzein and equol were used as external standards. Quantification was conducted using QuantAnalysis (Version 2.0, Bruker Daltonics Inc., Billerica, MA). Statistical Analyses. Data were expressed using mean ± standard deviation. Mean values between two groups were compared using t tests. One-way ANOVA with Tukey HSD tests were done using JMP software (Version 8.0, SAS Institute Inc., Cary, NC). A difference with p ≤ 0.05 was considered significant.

The zeta potential of all nanoparticles at pH 5.5 was positive and above +20 mV, whereas the zeta potential of all nanoparticles at pH 7.4 was negative and below −20 mV. This is because zein has an isoelectric pH of about 6.8.23 After emulsification using TPGS, the zeta potential was found to be slightly decreased compared with that of ZN. The encapsulation efficiency of the daidzin loaded TZN was about 63%, which was higher than that of the daidzin loaded ZN 53%. It was speculated that the leakage of daidzin was prohibited by using TPGS during particle fabrication. This was in agreement with an earlier study using chitosan coating.24 The scanning electron microscopic images of the nanoparticles revealed their spherical shapes with smooth surface morphology (Figure 1).



Figure 1. Scanning electron microscope images of ZN (A1), TZN (A2), coumarin-6 loaded ZN (B1), coumarin-6 loaded TZN (B2), daidzin loaded ZN (C1), and daidzin loaded TZN (C2).

RESULTS Characterization of Daidzin or Coumarin-6 Loaded ZN or TZN. The particle size, polydispersity index (PDI), and zeta potential of the nanoparticles in pH 5.5 water were summarized in Table 2. The particle size and zeta-potential in pH 7.4 water were listed in Supplemental Table S1. The particle size of all nanoparticles prepared in this study ranged from 160 to 230 nm. The PDI of all nanoparticles is less than 0.3. A low PDI suggested that the nanoparticles were nearly monodispersed in nature. The particle sizes of blank, coumarin-6, or daidzin loaded TZN were larger than that of corresponding ZN, suggesting the coating and embedding of TPGS in ZN. The content of TPGS in TZN was determined to be 29.1 ± 3.1% using HPLC.

Release of Daidzin from the Nanoparticles in Simulated Digestive Fluids. Release of daidzin from nanoparticles in SGF after 0.5 h and SIF after 2 h at 37 °C is shown in Figure S1. The release of daidzin from ZN was 62.0% in SGF and 50.9% in SIF, respectively. In contrast, the release of daidzin from TZN was 44.9% in SGF and 36.2% in SIF. The release of daidzin from TZN was significantly lower than that of ZN in both SGF and SIF, suggesting TPGS retarded the release of daidzin in simulated digestive fluids. In Vitro Release in pH 7.4 Buffer. The in vitro release of coumarin-6 from TZN and ZN was 3.42% and 5.27%, D

dx.doi.org/10.1021/mp400086n | Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Molecular Pharmaceutics

Article

20.3%, and 16.4% at 37.1 °C using nanoparticle concentrations of 100, 200, 400, and 800 μg/mL, respectively. The uptake efficiency of ZN was 10.3%, 12.9%, 9.9%, and 7.5% at 4 °C using nanoparticle concentrations of 100, 200, 400, and 800 μg/mL, respectively. In addition to nanoparticle concentration, the incubation time is also an important factor to determine the cellular uptake efficiency. Figure 3B shows the cellular uptake of the coumarin-6 loaded ZN or TZN after 0.5, 1, 2, and 4 h incubation at 200 μg/mL nanoparticle concentrations at 37.1 or 4 °C. The uptake efficiency of TZN was 20.4%, 26.2%, 32.8%, and 38.9% after 0.5, 1, 2, and 4 h incubation at 37.1 °C, respectively. The uptake efficiency of TZN was 8.5%, 11.1%, 15.6%, and 18.1% after 0.5, 1, 2, and 4 h incubation at 4 °C, respectively. The uptake efficiency of ZN was 13.4%, 18.3%, 24.8%, and 28.4% after 0.5, 1, 2, and 4 h incubation at 37.1 °C, respectively. The uptake efficiency of ZN was 5.6%, 7.5%, 12.9%, and 14.2% after 0.5, 1, 2, and 4 h incubation at 4 °C, respectively. It can be seen that the uptake of nanoparticles by the Caco-2 cells increased with the incubation time over the incubation period. Figure 4 shows confocal microscopic images of Caco-2 cells after 2 h incubation with coumarin-6 loaded ZN or TZN at a nanoparticle concentration of 200 μg/mL at 37.1 or 4 °C. The nucleus stained by propidium iodide was surrounded by green fluorescence of coumarin-6, suggesting that the nanoparticles were internalized in the cytoplasm. The cells incubated with TZN at 37.1 or 4 °C exhibited a thicker layer and stronger fluorescence than those incubated with the ZN under the same conditions. Furthermore, the cells incubated with nanoparticles at 37.1 °C also exhibited a thicker layer or stronger fluorescence than those incubated at 4 °C under the same conditions. Both supported the previous quantitative measurements of the cellular uptake of nanoparticles in Figure 3. Cellular Uptake and Transport of Daidzin-Loaded Nanoparticles by Caco-2 Monolayers. The TEER of Caco2 monolayer treated by daidzin solution did not significantly decrease during the incubation period (Figure 5). The TEER percentage of the initial value was 94.7% at 30 min and remained to be 87.5% after 120 min. No significant differences on TEER percentage were observed between daidzin solution and daidzin loaded TZN at 30, 60, or 120 min.

respectively. TZN had a slower coumarin-6 release compared with the ZN due to the retardation of TPGS on the release of coumarin-6. The in vitro daidzin release profiles of the daidzin loaded ZN or TZN at pH 7.4 are also shown in Figure 2. After 24 h, about

Figure 2. Release profiles of coumarin-6 or daidzin loaded ZN or TZN in transport buffer (Hank’s balanced salt solution, HBSS, pH 7.4) at 37.1 °C.

15% and 11% of daidzin was released from ZN and TZN, respectively. The ZN showed faster daidzin release compared with the TZN. Cellular Uptake of Coumarin-6 Loaded ZN or TZN. Figure 3 shows the cellular uptake of the coumarin-6 loaded ZN or TZN. Uptake efficiency was measured after Caco-2 cells were cultured for 2 h at 37.1 or 4 °C using 100, 200, 400, and 800 μg/mL nanoparticle concentrations, respectively (Figure 3A). The uptake efficiency of TZN was 28.2%, 34.8%, 30.3%, and 24.1% at 37.1 °C using nanoparticle concentration of 100, 200, 400, and 800 μg/mL, respectively. The uptake efficiency of TZN was 13.4%, 15.6%, 12.4%, and 9.7% at 4 °C using nanoparticle concentrations of 100, 200, 400, and 800 μg/mL, respectively. The uptake efficiency of ZN was 21.3%, 24.8%,

Figure 3. Concentration- and time-dependent change in uptake efficiency of the coumarin-6-loaded TZN and ZN at 37 and 4 °C on Caco-2 cells. An incubation time of 2 h was applied when studying concentration dependence. A concentration of 200 μg/mL was applied when studying time dependence. Data points within a same category marked with different letters differ significantly. Data points within a same concentration or time marked with different numbers differ significantly. E

dx.doi.org/10.1021/mp400086n | Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Molecular Pharmaceutics

Article

(Figure 6). The Papp of daidzin from the daidzin solution in this study were comparable with a previous study.25

Figure 4. Confocal microscopic images of Caco-2 cells after 2 h incubation at 37.1 °C with coumarin-6 loaded TZN (A) or ZN (B); after 2 h incubation at 4 °C with coumarin-6 loaded TZN (C) or ZN (D). The cells were stained by propidium iodide (red) and uptake of green fluorescent coumarin-6 loaded nanoparticles in Caco-2 cells was visualized by overlaying images obtained by fluorescein isothiocyanate filter and rhodamine B−isothiocyanate filter.

Figure 6. Apparent permeability coefficient (Papp) of daidzin after incubation daidzin solution or daidzin loaded TZN with Caco-2 cells for 0.5−2.0 h. Data points at a same time point marked with * differ significantly.

Pharmacokinetic Studies. Sulfatase from Helix pomatia contains both sulfatase and glucuronidase, and it was used to hydrolyze isoflavone conjugates in plasma samples. The concentration of daidzein and equol in mouse plasma samples was determined with a HPLC-MS method after enzymatic hydrolysis. The concentration−time curves were shown in Figure 7. At each time point after dosing, the plasma concentrations of daidzein in mice administrated with daidzin loaded in TZN were significantly higher than those treated with daidzin solution (Figure 7A). The Cmax,1h of daidzein after oral administration of daidzin loaded in TZN was 5.66 ± 0.16 μM, which was over 2-fold higher than that of daidzin solution (2.14 ± 0.04 μM). In particular, the areas under curve (AUC0−12h) of daidzin loaded TZN were enhanced by 2.4-fold compared with that of daidzin solution. Similarly, the plasma concentrations of equol in mice administrated with daidzin loaded TZN were significantly greater than those treated with daidzin solution except time point 1 and 2 h (Figure 7B).

Figure 5. Transepithelial electric resistance (TEER) percentage of initial value after incubating daidzin solution or daidzin loaded TZN with Caco-2 cell monolayers for 0.5−2.0 h.



Daidzin solution or daidzin in TZN was applied to the apical side of Caco-2 monolayers using the same daidzin concentration of 10 μmol/L. The daidzin solution did not change the daidzin concentration in the apical solution of the monolayer. In contrast, TZN significantly decreased the concentration of daidzin in the apical solution of the monolayer throughout the incubation period (Figure S2A). Daidzein (aglycone of daidzin), daidzein glucuronides, and sulfates appeared at the apical side of monolayers. Their contents after nanoparticle treatment were significantly higher than those treated with the daidzin solution (Figures S2B and S2C). The amounts of daidzin, daidzein, and daidzein glucuronides/sulfates in the cellular extract and the basolateral side of the monolayers treated with the nanoparticles were much higher than those treated with the daidzin solution (Figure S3 and S4). This confirmed that daidzin loaded TZN were taken up more efficiently into Caco-2 cells and transported more quickly through Caco-2 monolayers than the daidzin solution. The apparent permeability coefficient (Papp) of daidzin from apical side to basolateral side was calculated after incubating daidzin solution or daidzin TZN with Caco-2 cells for 0.5− 2.0 h. The Papp of daidzin from daidzin loaded TZN was about 1.9−2.9 × 10−6 cm/s for up to 2 h, which were significantly higher than those of daidzin solution at about 0.3−0.6 × 10−6 cm/s

DISCUSSION Daidzin was loaded on the nanoparticles using polyphenol− protein affinity. Hydroxyl groups of daidzin bind with zein via hydrogen binds and hydrophobic interactions.13,26 Particle size is a key factor affecting cellular internalization. Particles with a size below 200 nm diffuse through the mucus and avoid elimination by mucilliary clearance.27 The average size of TZN is 210.6 and 221.3 nm at pH 5.5 and 7.4, respectively. Therefore, most of TZN have potential to avoid elimination by mucilliary clearance. In general, the ability of nanoparticles to escape the endolysosomes depends on the surface charge. Nanoparticles with surface charge changing from anionic at pH 7.4 to cationic in the acidic endosomal pH (4−5) were able to escape the endosomal compartment.28 Therefore, nanoparticles in the present study may have the potential to escape the endosomes after internalization by cells. TPGS consists of a lipophilic α-tocopherol and a hydrophilic polyethylene glycol chain. It can be speculated that αtocopherol of the TPGS was partly inserted in the ZN and the polyethylene glycol chains protrude to cover the surface of the ZN. This may reduce the zeta potential of ZN and make the F

dx.doi.org/10.1021/mp400086n | Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Molecular Pharmaceutics

Article

uptake of the coumarin-6 molecules encapsulated in the ZN or TZN, instead of from release of coumarin-6 from nanoparticles into the medium. Results indicated that less than 6% of the coumarin-6 was released from the nanoparticles after 24 h of incubation (Figure 2). This suggested that most of the coumarin-6 remained in the nanoparticles and that the fluorescence signal detected in the cells was attributed to the coumarin-6 encapsulated within the nanoparticles. Such observations were confirmed by quantitative measurements in Figure 3. Cellular uptake efficiency of nanoparticles by Caco-2 cells reached a peak at a particle concentration of 200 μg/mL. This result suggested a saturated and limited capability of cells to take up nanoparticles. Similar observations were also reported by Dong and Feng.30 Cellular uptake efficiency of TZN was significantly higher than that of ZN at all tested particle concentrations at 37.1 °C. This supports the assumption that TPGS can be used as a reversing agent for P-glycoprotein mediated multidrug resistance. It was also seen that there was a significant reduction in nanoparticle uptake by the Caco-2 cells at 4 °C, compared with the uptake at 37.1 °C at equivalent particle concentration and incubation time. This suggested that the nanoparticle uptake by the Caco-2 cells were an energydependent endocytic process.31 TZN had better stability and cellular uptake efficiency than ZN based on the previous experiments. Therefore they were selected for additional cellular uptake and transport experiments as well as pharmacokinetic experiments. The influence of daidzin loaded TZN on the TEER value of Caco-2 cells was compared with daidzin solution (Figure 5). The TEER decrease of Caco-2 cells incubated with the nanoparticles was