Pure Trans-Resveratrol Nanoparticles Prepared by A Supercritical Antisolvent Process Using Alcohol and Dichloromethane Mixtures: Effect of Particle Size on Dissolution and Bioavailability in Rats.

The aim of this study was to prepare pure trans-resveratrol nanoparticles without additives (surfactants, polymers, and sugars) using a supercritical antisolvent (SAS) process with alcohol (methanol or ethanol) and dichloromethane mixtures. In addition, in order to investigate the effect of particle size on the dissolution and oral bioavailability of the trans-resveratrol, two microparticles with different sizes (1.94 μm and 18.75 μm) were prepared using two different milling processes, and compared to trans-resveratrol nanoparticles prepared by the SAS process. The solid-state properties of pure trans-resveratrol particles were characterized. By increasing the percentage of dichloromethane in the solvent mixtures, the mean particle size of trans-resveratrol was decreased, whereas its specific surface area was increased. The particle size could thus be controlled by solvent composition. Trans-resveratrol nanoparticle with a mean particle size of 0.17 μm was prepared by the SAS process using the ethanol/dichloromethane mixture at a ratio of 25/75 (w/w). The in vitro dissolution rate of trans-resveratrol in fasted state-simulated gastric fluid was significantly improved by the reduction of particle size, resulting in enhanced oral bioavailability in rats. The absolute bioavailability of trans-resveratrol nanoparticles was 25.2%. The maximum plasma concentration values were well correlated with the in vitro dissolution rate. These findings clearly indicate that the oral bioavailability of trans-resveratrol can be enhanced by preparing pure trans-resveratrol nanoparticles without additives (surfactants, polymers, and sugars) by the SAS process. These pure trans-resveratrol nanoparticles can be applied as an active ingredient for the development of health supplements, pharmaceutical products, and cosmetic products.


Introduction
Trans-resveratrol, a phytochemical with strong antioxidant property, is naturally found in grapes, berries, and peanuts. The benefits of trans-resveratrol have been reported in various diseases, including cancer, cardiovascular diseases, diabetes mellitus, neurological diseases, obesity, and other conditions associated with oxidative stress and inflammation [1][2][3][4][5]. Chemically, trans-resveratrol is converted into cis-resveratrol by light and has a very low aqueous solubility [6,7]. Further, its benefits are affected after its oral administration owing to its poor solubility and instability in physiological conditions as well as extensive metabolism [8,9]. To date, several formulation strategies such as amorphous solid dispersions, composite nanoparticles, emulsions, liposomes, polymeric micelles, self-emulsifying drug delivery systems, and solid lipid nanoparticles have been investigated to resolve this problem [10][11][12][13][14][15][16][17][18][19]. In addition, the enhanced oral bioavailability in humans has been reported with lipid formulation, micellar liquid formulation, and microparticles mixed with a liquid formulation [20][21][22]. Previously, the area under the plasma concentration versus time curve (AUC) and maximum plasma concentration (C max ) values obtained after oral administration at a dose of 40-mg trans-resveratrol were approximately 10-fold and 8.5-fold higher in the capsule containing solubilized resveratrol in polysorbate 20 and polyglyceryl-3-dioleate mixture than in powder capsule (original powder), respectively [20]. In a case study using capsules containing solubilized resveratrol in polysorbate 80, polysorbate 20, and medium chain triacylglycerol mixtures, the intake of a liquid micellar formulation containing 30 mg trans-resveratrol produced higher AUC (5.0-fold increase) and C max (10.6-fold increase) than the equivalent dose of the original powder [21]. In a human pharmacokinetic study using a suspension prepared by mixing resveratrol powder and the liquid formulation (docusate sodium solution and distilled water) at a daily dose of 5.0 g trans-resveratrol, the oral bioavailability of the micronized powder (particle size less than 5 µm) was significantly higher than that of the nonmicronized powder, with approximately 3.6-fold increases in the C max values at 14 days [22]. However, as a health supplement, a large amount of trans-resveratrol must be administered for long periods. Minimizing the use of additives such as surfactants (polysorbates and docusate sodium) can reduce its side effects.
Particle size reduction for poorly water-soluble phytochemicals is an important process for increasing the dissolution rate and, consequently, the oral bioavailability of compounds [23][24][25][26]. Generally, nanoparticles can be prepared via two different approaches, the 'top-down' and 'bottom-up' approaches. In the 'top-down' and 'bottom-up' methods, water containing surface stabilizers was used as a dispersion medium or an antisolvent. To transform the phytochemicals into the solid form, nanosuspensions are dried by spray-drying and fluidized bed granulation after the addition of dispersants (usually sucrose and mannitol) to reduce the aggregation of nanoparticles during solidification and enhance nanoparticle redispersion in water [27,28]. These methods have some disadvantages, such as the use of massive additives. Alternative technologies, such as the supercritical fluid technology, can produce solid nanoparticles in a one-step process without the use of additives and a subsequent water removal process [29][30][31]. Therefore, we attempted to prepare pure trans-resveratrol nanoparticles without the additives (surfactants, polymers, and sugars) by using the supercritical antisolvent (SAS) process and alcohol and dichloromethane mixtures. Herein, the solid-state properties of the pure trans-resveratrol nanoparticles were characterized. Further, particle size reduction of trans-resveratrol was performed using a Fitz mill and an air jet-mill, and a dissolution and pharmacokinetic study was conducted to investigate the effect of particle size on the dissolution and oral bioavailability of trans-resveratrol.

Solubility Study of Trans-Resveratrol in the Alcohol and Dichloromethane Mixtures
To understand the solubility behavior of trans-resveratrol in the solvents, the solubility of trans-resveratrol in alcohol (methanol or ethanol) and dichloromethane mixtures was determined by Antioxidants 2020, 9, 342 3 of 16 the shake-flask method through solid-liquid equilibrium [32][33][34]. Briefly, 15 g of the solvent mixture was prepared in an amber vial by mixing alcohol (methanol or ethanol) and dichloromethane at different weight ratios (alcohol:dichloromethane; 0:100, 25:75, 50:50, 75:25, 100:0 w/w). Excess trans-resveratrol (raw material) was mixed with the solvents in amber vials via vortex for 15 min and sonication in an ultrasonic bath (5800 model, Branson, USA) for 45 min. The amber vials were shaken at 60 rpm in a water-bath mechanical shaker (BS-21, Jeiotech Co., Ltd., Daejeon, Korea) at 25 • C for at least 24 h. For HPLC analysis, the sample was filtered using a 0.2 µm syringe filter to obtain a clear supernatant from a saturated solution. Thereafter, the filtrate was placed in a preweighed volumetric flask. The concentration of trans-resveratrol was determined using a Prominence HPLC-ultraviolet (HPLC-UV) system (Shimadzu, Tokyo, Japan) after dilution with methanol. Trans-resveratrol was separated by Gemini C18 column (150 × 4.6 mm, 5 µm, Phenomenex, Torrance, CA, USA). The mobile phase consisted of acetonitrile/water (40:60, v/v). The flow rate was 0.8 mL/min, column temperature was 30 • C, and the detection wavelength was 303 nm. All solubility measurements were repeated three times.

Preparation of Pure Trans-Resveratrol Nanoparticles Using the SAS Process
Pure trans-resveratrol nanoparticles were prepared by the SAS process, as previously reported [35][36][37]. To prepare the sample solution, raw trans-resveratrol (30 mg/g) was dissolved in alcohol (methanol or ethanol) and dichloromethane at different weight ratios (alcohol: dichloromethane; 25:75, 50:50, 75:25, 100:0 w/w) in an amber vial. Carbon dioxide liquefied with a cooler was heated to a desired point (40 • C) using a heat exchanger, and then injected into the particle formation vessel (1 L) through a two-way nozzle using a syringe pump (ISCO 260D dual-pump continuous flow systems, Teledyne Technologies Inc., Thousand Oaks, CA, USA). Once a pressure of 150 bar and temperature of 40 • C were achieved and equilibrium was reached, supercritical carbon dioxide was delivered at a rate of 40 g/min and the sample solution was simultaneously injected at a rate of 0.5 g/min using an HPLC pump (ReaXus LD Class, Teledyne Technologies Inc., Thousand Oaks, CA, USA). Upon injecting the sample solution, only carbon dioxide was further pumped at the same rate to remove the residual solvent that could dissolve the prepared particles. The SAS system was slowly depressurized to atmospheric pressure. The generated particles were collected on the walls and bottom of the particle formation vessel.

Particle Size Reduction of Trans-Resveratrol Using Fitz Mill or Air Jet-Mill
Two different milling processes using a Fitz mill and an air jet-mill were applied for particle size reduction of raw trans-resveratrol. To generate trans-resveratrol microparticles with the mean particle size of approximately 20 µm, raw trans-resveratrol with a mean particle size of 108.94 µm was milled in the Fitz mill (Fitz Mill L1A, Fitzpatrick Company, Westwood, MA, USA) under the following conditions: speed, 500-1000 rpm; feed rate, 0.5 g/min. To obtain trans-resveratrol microparticles with a mean particle size of approximately 2 µm, continuous grinding using an air jet-mill (McOne, Jetpharma, Balerna, Switzerland) was conducted under the following process conditions: injection pressure, 8 bar; grinding pressure, 8 bar; and feed rate, 0.5 g/min.

Analysis of Trans-Resveratrol Purity and Residual Solvent
The purity of trans-resveratrol after the SAS process and milling process was validated by HPLC analysis. The sample solution and standard solution were prepared by dissolving the processed trans-resveratrol and USP reference standard in methanol at 100 µg/mL concentration, respectively. A 10-µL volume of the solution was injected into a Prominence HPLC-ultraviolet (HPLC-UV) system (Shimadzu, Tokyo, Japan) as described above. Residual solvents of trans-resveratrol after the SAS process were measured by a gas chromatograph (GC) (Hewlett-Packard, 5890 SERIES II, Palo Alto, CA, USA) coupled with head space sampler (Hewlett-Packard, 7694 headspace sampler, Palo Alto, CA, USA) and flame ionization detector. Analysis was performed as previously described with slight modifications [38]. The fused-silica capillary column (Phenomenex, model ZB-624, 30 m length × 0.32 mm I.D. (internal diameter), Torrence, CA, USA) was selected for the simultaneous separation of ethanol, methanol, and dichloromethane in the sample solution. The following programmed GC conditions were employed: the initial temperature of 35 • C was held for 10 min after injection, then increased at a rate of 15 • C/min to 40 • C, where the temperature maintained for 10 min, then increased at a rate of 18 • C/min to 235 • C. After holding for 8 min at 235 • C, the temperature was returned to its initial value. The headspace conditions were: equilibration time, 60 min at 80 • C with gentle shaking; pressurization time, 0.5 min; and loop fill time, 0.1 min. The test sample solution was prepared by dissolving 100 mg of each sample dissolved in 10 mL of DMSO in 10 mL vials. Analyses were performed in triplicate for each sample. The morphology of the trans-resveratrol nanoparticles and microparticles was observed using a scanning electron microscope (SUPRA 25 or 40, Zeiss, Oberkochen, Germany). Electrically conductive samples were prepared by gold coating in vacuum for 1 min, before observation. Samples were visualized at an accelerating voltage of 5 kV.

Particle Size Measurements
For the nanoparticles, the z-average size and polydispersity index (PI) were measured using ELSZ-1000 (Otsuka Electronics, Tokyo, Japan) according to the dynamic light scattering (DLS) method. In addition, particle size and size distribution of trans-resveratrol nanoparticles and microparticles were determined using laser diffraction (LS 13 320, Coulter Beckman, Brea, CA, USA) according to the Mie scattering theory. Samples were sufficiently dispersed in water, and all particle size measurements were repeated three times.

Specific Surface Area Measurements
The specific surface area of trans-resveratrol nanoparticles and microparticles was determined using the Micromeritics TriStar II 3020 instrument (Micromeritics, Norcross, GA, USA) based on the Brunauer-Emmett-Teller method with the nitrogen adsorption analysis after degassing samples with helium (purity > 99.99%).

Differential Scanning Calorimetry (DSC) Analysis
DSC measurements were conducted with a Discovery DSC 25 instrument connected to an RSC90 cooling system (TA Instruments, Inc., New Castle, DE, USA). High purity indium and aluminum oxide sapphire were used to calibrate temperature and heat capacity. The trans-resveratrol samples ranging from 3 to 4 mg were accurately weighed and sealed in an aluminum pan. The same hermetic aluminum pan without sample was used as a reference. The measurements were carried out under nitrogen purge (300 mL/min) in the temperature range of 0 to 300 • C, with a heating rate of 10 • C/min. All measurements for each sample were performed in triplicate.

Powder X-Ray Diffraction (PXRD) Analysis
PXRD analysis was carried out using an X-ray diffractometer (XPert 3, Panalytical, Almelo, Netherlands), with Cu-Kα radiation at a voltage of 40 kV and a current of 40 mA. The samples were scanned from 5 • to 50 • (2θ), with a scanning speed of 3 • /min and a step size of 0.01 • .

Dissolution Study
Dissolution under the physiological conditions of the gastrointestinal tract is the first process in the oral absorption of trans-resveratrol. To investigate the effect of particle size on dissolution and to derive the relationship between in vitro dissolution rate and in vivo pharmacokinetic parameters, we conducted dissolution testing of trans-resveratrol nanoparticles and microparticles in fasted state-simulated gastric fluid (FaSSGF) and fasted state-simulated intestinal fluid (FaSSIF) as biorelevant dissolution media. First, the solubility of trans-resveratrol (raw material) in FaSSGF and FaSSIF was determined at 37 • C as described in the solubility measurement method above. FaSSGF and FaSSIF were prepared with sodium chloride, hydrochloric acid, and Biorelevant powder (Biorelevant.com Ltd., London, UK) and sodium hydroxide, sodium phosphate monobasic, sodium chloride, and Biorelevant powder, respectively, according to the method retrieved from Biorelevant.com. The dissolution test was carried out in accordance with the USP paddle method using a Varian VK 7010 dissolution system (Vankel Industries, Edison, NJ, USA) at 50 rpm and 37 ± 0.5 • C. To neglect the effect of wettability, trans-resveratrol nanoparticles and microparticles were dispersed in water at 50 mg/mL prior to the dissolution study. A sample of 50 mg trans-resveratrol was placed in 900 mL of dissolution media (FaSSGF). Dissolved trans-resveratrol was quantified using Pion Rainbow spectrometers with 2-mm path length UV fiber optic probes (Pion Inc., Billerica, MA, USA) at a wavelength of 306 nm. Data were collected every 1 min for 60 min, and all dissolution measurements were repeated six times.

Pharmacokinetic Study in Rats
To investigate the effect of particle size on bioavailability, the pharmacokinetic parameters of trans-resveratrol nanoparticles and microparticles were obtained following their oral administration to male Sprague-Dawley (SD) rats. The animal study protocol complied with the institutional guidelines for the care and use of laboratory animals and was approved by the ethics committee of Kyungsung University (No. 19-008A). Twenty-four male SD rats (200-230 g; HanaBiotech, Suwon, Korea) were divided into four treatment groups (n = 6 per group). Three groups received either trans-resveratrol nanoparticles (0.17 µm) prepared by the SAS process, trans-resveratrol microparticles (1.94 µm) milled by the air jet-mill, and trans-resveratrol microparticles (18.25 µm) milled by the Fitz mill at a dose of 50 mg/kg using an oral dosing apparatus. Each particle was dispersed in 1 mL of water immediately before dosing [39]. To estimate the bioavailability (F%) of trans-resveratrol nanoparticles and microparticles, one group was intravenously administered via the tail vein a dose of 15 mg/2.5 mL/kg (solution prepared by dissolving trans-resveratrol in DMSO and PEG 300 mixtures, 15:85 v/v) [40]. Serial blood samples (approximately 300 µL each) were collected from the jugular vein at different time intervals. The blood samples were centrifuged (12,000 rpm, 5 min), and plasma was collected and stored at −70 • C until HPLC analysis. The concentration of trans-resveratrol in plasma was determined using a previously reported method and under HPLC analysis conditions [16,41]. The AUC 0→8h was calculated using the linear trapezoidal method. C max and the time required to reach C max (T max ) were directly obtained from plasma data.

Data Analysis
Data are expressed as mean ± standard deviation (n = 3, 4, or 6). One-way analysis of variance followed by the least significant difference test and the Student-Newman-Keuls test was performed using SPSS (version 25.0; IBM SPSS Statistics, IBM Corporation, Armonk, NY, USA).

Pure Trans-Resveratrol Nanoparticles Prepared by the SAS Process
To prepare the trans-resveratrol nanoparticles using the SAS process, the solubility of trans-resveratrol was determined in alcohol (methanol or ethanol) and dichloromethane mixtures at different weight ratios (alcohol: dichloromethane; 0:100, 25:75, 50:50, 75:25, 100:0 w/w) and the measured data were graphically plotted ( Figure 1). The solubility of trans-resveratrol in dichloromethane was 5.38 µg/mg, indicating very low solubility. However, the solubility of trans-resveratrol was markedly increased by the addition of a small amount of alcohol, and reached 38.26 mg/g in the ethanol and dichloromethane mixture at 25:75 (w/w). As shown in Figure 1, solubility was proportionally increased with an increase in the alcohol ratio with mixtures of methanol and ethanol. Based on solubility, dichloromethane and alcohol (methanol and ethanol) were recognized as "poor solvent" and "good solvent," respectively, with respect to trans-resveratrol.
Based on the solubility data, we prepared trans-resveratrol nanoparticles using the SAS process and 30 mg/g of trans-resveratrol to investigate the effect of solvent composition (alcohol: dichloromethane; 25:75, 50:50, 75:25, 100:0 w/w) on particle size and morphology. As shown in Figure 2 and Table 1, the trans-resveratrol precipitated from the methanol-dichloromethane mixtures had smaller nanoparticles than those generated from methanol. By increasing the percentage of dichloromethane in the solvents, the mean particle size decreased and the specific surface area increased. The PI values for trans-resveratrol precipitated from the methanol/dichloromethane mixtures at ratios 25:75 and 50:50 (w/w) were below 0.2, indicating narrow particle size distribution [42,43]. Cohesion of the particles was observed, which may be due to the relatively high surface energy of fine nanoparticles. From the results obtained, nanoparticles were easily dispersed in water, and the particle sizes measured by DLS method were in good agreement with the SEM images, and it was confirmed that the particles were not aggregated via irreversible bonding. In most particle formation studies using the SAS process, a single solvent is employed to dissolve compounds. By performing some case studies using solvent mixtures, De Marco et al. [44,45] reported that the particle size and morphology of compounds were controlled by the solvent mixtures with a "poor solvent" and "good solvent" through the enhanced mixing effect of the "poor solvent." When ethanol and the ethanol-dichloromethane mixtures were used, a similar effect to that of dichloromethane in the methanol-dichloromethane mixtures was observed as the particle size of trans-resveratrol was reduced. In particular, the trans-resveratrol nanoparticles prepared at a ratio of 25:75 (ethanol:dichloromethane) had a mean particle size of 151.2 nm, with 60.23 m 2 /g. Additionally, it had the smallest particle size among the prepared samples.
As shown in Figure 3 and Table S1 (Supplementary Material), the melting temperature, fusion enthalpy, and X-ray characteristic peaks of trans-resveratrol in all processed samples were the same as those of the unprocessed raw trans-resveratrol (p > 0.05), which means that there was no difference in crystal polymorphism and crystallinity between nanoparticle samples prepared by the SAS process. The process yield was above 85% for all experiments. All trans-resveratrol nanoparticles prepared using the SAS process exhibited low residual solvents; methanol < 20 ppm, ethanol < 20 ppm, and dichloromethane < 10 ppm according to the residual solvent analysis. The purity (%) of all trans-resveratrol nanoparticles was above 99.3%, indicating no degradation and conversion to cis-resveratrol during the SAS process. Therefore, pure trans-resveratrol nanoparticles were successfully prepared using the SAS process with alcohol and dichloromethane and their particle size could be controlled by solvent composition. Based on the above results, trans-resveratrol nanoparticles prepared at a ratio of 25:75 (ethanol:dichloromethane, w/w) were selected as the best sample for further investigation. Based on the solubility data, we prepared trans-resveratrol nanoparticles using the SAS process and 30 mg/g of trans-resveratrol to investigate the effect of solvent composition (alcohol: dichloromethane; 25:75, 50:50, 75:25, 100:0 w/w) on particle size and morphology. As shown in Figure  2 and Table 1, the trans-resveratrol precipitated from the methanol-dichloromethane mixtures had smaller nanoparticles than those generated from methanol. By increasing the percentage of dichloromethane in the solvents, the mean particle size decreased and the specific surface area increased. The PI values for trans-resveratrol precipitated from the methanol/dichloromethane mixtures at ratios 25:75 and 50:50 (w/w) were below 0.2, indicating narrow particle size distribution [42,43]. Cohesion of the particles was observed, which may be due to the relatively high surface energy of fine nanoparticles. From the results obtained, nanoparticles were easily dispersed in water, and the particle sizes measured by DLS method were in good agreement with the SEM images, and it was confirmed that the particles were not aggregated via irreversible bonding. In most particle formation studies using the SAS process, a single solvent is employed to dissolve compounds. By performing some case studies using solvent mixtures, De Marco et al. [44,45] reported that the particle size and morphology of compounds were controlled by the solvent mixtures with a "poor solvent" and "good solvent" through the enhanced mixing effect of the "poor solvent." When ethanol and the ethanol-dichloromethane mixtures were used, a similar effect to that of dichloromethane in the methanol-dichloromethane mixtures was observed as the particle size of trans-resveratrol was reduced. In particular, the trans-resveratrol nanoparticles prepared at a ratio of 25:75 (ethanol:dichloromethane) had a mean particle size of 151.2 nm, with 60.23 m 2 /g. Additionally, it had the smallest particle size among the prepared samples.
As shown in Figure 3 and Table S1 (Supplementary material), the melting temperature, fusion enthalpy, and X-ray characteristic peaks of trans-resveratrol in all processed samples were the same as those of the unprocessed raw trans-resveratrol (p > 0.05), which means that there was no difference in crystal polymorphism and crystallinity between nanoparticle samples prepared by the SAS process. The process yield was above 85% for all experiments. All trans-resveratrol nanoparticles prepared using the SAS process exhibited low residual solvents; methanol < 20 ppm, ethanol < 20 ppm, and dichloromethane < 10 ppm according to the residual solvent analysis. The purity (%) of all trans-resveratrol nanoparticles was above 99.3%, indicating no degradation and conversion to cisresveratrol during the SAS process. Therefore, pure trans-resveratrol nanoparticles were successfully prepared using the SAS process with alcohol and dichloromethane and their particle size could be controlled by solvent composition. Based on the above results, trans-resveratrol nanoparticles prepared at a ratio of 25:75 (ethanol:dichloromethane, w/w) were selected as the best sample for further investigation.

Effect of Particle Size on Dissolution and Oral Bioavailability
To investigate the effect of particle size on dissolution and oral bioavailability, two different sizes of trans-resveratrol microparticles were prepared using a Fitz mill and an air jet-mill in contrast to the nanoparticles prepared by the SAS process. As a result, the particle size of raw trans-resveratrol with a mean particle size of 108.94 µm was decreased to 18.25 µm by the milling process using the Fitz mill and 1.94 µm by the air jet-mill. In contrast, the trans-resveratrol nanoparticles with mean particle sizes of 0.17 µm and 60.23 m 2 /g were prepared by the SAS process using the ethanol: dichloromethane mixture at a ratio of 25:75. The physicochemical properties of the trans-resveratrol nanoparticles and microparticles are summarized in Table 2. The melting temperature and fusion enthalpy of trans-resveratrol nanoparticles and microparticles were not significantly different (p > 0.05) to those of the unprocessed raw trans-resveratrol (Table S1, Supplementary material). The results of DSC and PXRD confirmed that there was no difference in crystal polymorphism and crystallinity between nanoparticles and microparticles.
To investigate the effect of particle size on dissolution and derive the relationship between in vitro dissolution rate and in vivo pharmacokinetic parameters, we carried out a dissolution test with the trans-resveratrol nanoparticles and microparticles in FaSSGF and FaSSIF as biorelevant dissolution media. To date, the solubility of trans-resveratrol in these dissolution media has not been reported in the literature. Based on this study, the solubility values of trans-resveratrol in FaSSGF and FaSSIF at 37 • C were 61.59 ± 1.01 µg/mL and 87.15 ± 1.31 µg/mL, respectively. The solubility in FaSSIF (pH 6.5) was superior to that in FaSSGF (pH 1.5) and could be attributed to micellar solubilization instead of pH as a result of the higher concentration of sodium taurocholate and lecithin; this is because trans-resveratrol exhibited a similar solubility in the pH range of 1.2-6.8 [10,46]. To predict the in vivo dissolution and investigate the effect of particle size, 50 mg of trans-resveratrol was applied for solubilization in 900 mL of FaSSGF or FaSSIF. The dissolution profiles of the trans-resveratrol nanoparticles and microparticles in FaSSGF and FaSSIF are shown in Figure 4. Evidently, the dissolution rate of trans-resveratrol was affected by its particle size in both FaSSGF and FaSSIF. As a nanoparticle (0.17 µm), trans-resveratrol was approximately 90% dissolved within 3 min; however, as a microparticle (18.25 µm), only~50% dissolved at 60 min in both FaSSGF and FaSSIF. This enhancement in the dissolution rate is a consequence of the increase in surface area and decrease in diffusion layer thickness by particle size reduction. For further quantitative analysis, the Hixson-Crowell cube root equation was applied to estimate the resveratrol release rate and simulated 50% dissolution time (t 50% ) [47]. As shown in Table 3, the resveratrol release rate and simulated 50% dissolution time (t 50% ) for the trans-resveratrol nanoparticles and microparticles were increased and decreased, respectively, as the particle size was reduced.         For comparison, the particle size and size distribution of the trans-resveratrol nanoparticles and microparticles were determined using the laser diffraction method. Span = (d90−d10)/d50, where d10, d50, and d90 are the diameters and the given percentage value is the percentage of particles smaller than that size. DSC: differential scanning calorimetry, PXRD: powder Xray diffractometry. For comparison, the particle size and size distribution of the trans-resveratrol nanoparticles and microparticles were determined using the laser diffraction method. Span = (d90−d10)/d50, where d10, d50, and d90 are the diameters and the given percentage value is the percentage of particles smaller than that size. DSC: differential scanning calorimetry, PXRD: powder Xray diffractometry. For comparison, the particle size and size distribution of the trans-resveratrol nanoparticles and microparticles were determined using the laser diffraction method. Span = (d90−d10)/d50, where d10, d50, and d90 are the diameters and the given percentage value is the percentage of particles smaller than that size. DSC: differential scanning calorimetry, PXRD: powder Xray diffractometry. For comparison, the particle size and size distribution of the trans-resveratrol nanoparticles and microparticles were determined using the laser diffraction method. Span = (d 90 −d 10 )/d 50 , where d 10 , d 50 , and d 90 are the diameters and the given percentage value is the percentage of particles smaller than that size. DSC: differential scanning calorimetry, PXRD: powder X-ray diffractometry.  To investigate the effect of particle size on the bioavailability, the pharmacokinetic parameters of trans-resveratrol nanoparticles and microparticles were obtained following their oral administration to male SD rats (Table 4). In addition, the absolute oral bioavailability (F%) of the transresveratrol nanoparticles and microparticles were estimated using the mean value of the intravenously administered AUC0 → 8 h. As shown in Figure 5, the nanoparticles achieved Cmax within 0.5 h after oral administration and had a higher concentration than the microparticles. For the pure transresveratrol particles lacking solubilizing additives, the rapid dissolution generated by the 0.17 μm nanoparticles led to rapid oral absorption, resulting in enhanced Cmax. However, the degree of enhancement via particle size reduction was relatively small for the AUC values, which may be due to the rapid metabolism of resveratrol [48]. Nonetheless, the absolute bioavailability of pure transresveratrol nanoparticles prepared using the SAS process was 25.2%. Previously, the absolute bioavailability of trans-resveratrol was reported to exist between 2.6% and 46.4%, thereby varying according to the sample preparation (the type and amount of suspending agent and surfactant) employed for oral administration in rats [38,39,[49][50][51].   To investigate the effect of particle size on the bioavailability, the pharmacokinetic parameters of trans-resveratrol nanoparticles and microparticles were obtained following their oral administration to male SD rats (Table 4). In addition, the absolute oral bioavailability (F%) of the trans-resveratrol nanoparticles and microparticles were estimated using the mean value of the intravenously administered AUC 0→8 h . As shown in Figure 5, the nanoparticles achieved C max within 0.5 h after oral administration and had a higher concentration than the microparticles. For the pure trans-resveratrol particles lacking solubilizing additives, the rapid dissolution generated by the 0.17 µm nanoparticles led to rapid oral absorption, resulting in enhanced C max . However, the degree of enhancement via particle size reduction was relatively small for the AUC values, which may be due to the rapid metabolism of resveratrol [48]. Nonetheless, the absolute bioavailability of pure trans-resveratrol nanoparticles prepared using the SAS process was 25.2%. Previously, the absolute bioavailability of trans-resveratrol was reported to exist between 2.6% and 46.4%, thereby varying according to the sample preparation (the type and amount of suspending agent and surfactant) employed for oral administration in rats [38,39,[49][50][51].  After oral administration, dissolution in the physiological condition of the stomach is the first process in the oral absorption of trans-resveratrol. Thus, the dissolution data obtained in FaSSGF were employed to establish the relationship between in vitro dissolution rate and in vivo pharmacokinetic parameters ( Figure 6). Cmax values were well correlated with the in vitro dissolution rate compared with the AUC0→8 h values. In fact, the in vitro dissolution rate controlled by particle size could predict the in vivo pharmacokinetic parameters for trans-resveratrol.  After oral administration, dissolution in the physiological condition of the stomach is the first process in the oral absorption of trans-resveratrol. Thus, the dissolution data obtained in FaSSGF were employed to establish the relationship between in vitro dissolution rate and in vivo pharmacokinetic parameters ( Figure 6). C max values were well correlated with the in vitro dissolution rate compared with the AUC 0→8 h values. In fact, the in vitro dissolution rate controlled by particle size could predict the in vivo pharmacokinetic parameters for trans-resveratrol.

Conclusions
Antioxidants 2020, 9, x FOR PEER REVIEW 11 of 17 The absolute oral bioavailability (F%) was determined by dividing the mean value of orally administered AUC0→8 h by the mean value of intravenously administered AUC0→8 h with dose normalization. a p < 0.05 vs. 18.75 μm trans-resveratrol; b p < 0.05 vs. 1.94 μm trans-resveratrol. Data are expressed as mean ± standard deviation (n = 6). AUC0→8 h, the area under the plasma concentration versus time curve; Cmax, the maximum plasma concentration of trans-resveratrol; Tmax, the time required to reach Cmax. After oral administration, dissolution in the physiological condition of the stomach is the first process in the oral absorption of trans-resveratrol. Thus, the dissolution data obtained in FaSSGF were employed to establish the relationship between in vitro dissolution rate and in vivo pharmacokinetic parameters ( Figure 6). Cmax values were well correlated with the in vitro dissolution rate compared with the AUC0 → 8 h values. In fact, the in vitro dissolution rate controlled by particle size could predict the in vivo pharmacokinetic parameters for trans-resveratrol.

Conclusions
Pure trans-resveratrol nanoparticles can be successfully prepared using the SAS process with alcohol (methanol or ethanol) and dichloromethane. We demonstrated that their particle size is controlled by solvent composition. Further, we revealed that the in vitro dissolution rate of trans-resveratrol is significantly improved by a reduction in particle size, thereby enhancing the oral bioavailability. These findings clearly indicate that the oral bioavailability of trans-resveratrol can be enhanced by preparing pure trans-resveratrol nanoparticles without additives (surfactants, polymers, and sugars) by the SAS process. This pure trans-resveratrol nanoparticle can be used as an active ingredient in the development of health supplements, pharmaceutical products, and cosmetic products.