Optimization and bioavailability evaluation of self-microemulsifying drug delivery system of the daidzein–nicotinamide complex

Abstract

Daidzein (DDZ) is effective in the treatment of a variety of cardiovascular diseases, hypertension and atherosclerosis. However, its extremely low oral bioavailability that is mainly caused by the poor solubility either in the aqueous or oil phase restricts its oral applications. To improve the solubility, a daidzein–nicotinamide (DDZ–NCT) complex was prepared by a solvent evaporation method with DDZ and nicotinamide (NCT). Then self-microemulsifying drug delivery system (SMEDDS) formulation of the DDZ–NCT complex was optimized using central composite design-response surface methodology. The results of powder X-ray diffraction (PXRD), Fourier transform infrared spectroscopy (FT-IR) and Simultaneous DSC-TGA (SDT) indicated that we successfully prepared the DDZ–NCT complex. The saturated solubility of the DDZ–NCT complex was improved 3.49-fold compared to that of DDZ in distilled water. The optimized SMEDDS (DDZ–NCT complex) formulation consisted of 15% Castol oil, 60.7% Cremophor RH40, 24.3% PEG400 and 5% DDZ–NCT complex. The in vitro intestinal absorption in 2 h and the oral bioavailability of SMEDDS (DDZ–NCT complex) were 4.41-fold and 5.69-fold higher than that of pure DDZ, respectively. The data showed that SMEDDS (DDZ–NCT complex) significantly enhanced the solubility and bioavailability of DDZ. Therefore, SMEDDS (DDZ–NCT complex) might be a promising system for the oral administration of DDZ and would be a potential system to deliver poorly water-soluble drugs.

---

1. Introduction

Daidzein (DDZ), as a primary chemical substance in leguminous plants, was very effective in the treatment of a variety of cardiovascular diseases, hypertension and atherosclerosis^1,2 (Fig. 1a). However, DDZ possesses tremendously low solubility characteristics (2.30 ± 0.10 μg mL⁻¹) in aqueous systems at 25 °C. As the marketed form of DDZ has an inevitably high dose requirement (≥150 mg per day) for therapeutic effect, the low lipophilicity restricts its application in either oral or intravenous administration.³

Fig. 1 Chemical structure of DDZ (a) and NCT (b).

Nicotinamide (NCT) could easily dissolve in water and is applied to cosmetics due to its depigmenting and moisturizing properties⁴ (Fig. 1b). Recent years NCT has been reported as an appropriate substance to form complex with various APIs.⁵

In order to improve the pharmaceutical properties of an active pharmaceutical ingredient (API), the development of the pharmaceutical complex have been broadly studied in recent years.^6,8,13–15 The API and co-formers in the complex typically require hydrogen bonding or other directional non-covalent interactions.¹⁷ A variety of techniques can be used to prepare complex such as crystallization, grinding and recrystallization from solvents.^18,19 With these developed techniques, a number of complexes have been studied and successfully synthesized.^7,9–12,16 As reported, complexes alter the physicochemical properties of API such as solubility, dissolution rate, even physical stability. As yet, the complex of DDZ has not been reported, this article will describe the preparation, characterization, and pharmacokinetics of the DDZ complex.

Self-microemulsifying drug delivery system (SMEDDS) is another promising technology to improve the absorption of poorly water-soluble drugs. SMEDDS are comprised of isotropic mixtures of drug substance, oil, surfactant (S) and/or co-surfactant (Co–S) and water.^20,21 The fundamental theory of this delivery system is to form fine oil-in-water (o/w) microemulsions by gentle agitation, similar to those which would be encountered in the gastro-intestinal tract.^22–25 Due to the dissolved form of the drug in the formulation and the resulting small droplets size, as interfacial surface area become large, these systems can improve drug solubility and enhance bioavailability. In a previous study, our laboratory has successfully prepared a SMEDDS (DDZ) formulation.²⁶ The formulation consisted of Castol oil, Cremophor RH40 and PEG400. This system is proved to have significantly increased the bioavailability of DDZ, which provided us with a potential tactic in order to deliver DDZ through oral route. The qualities of SMEDDS are closely dependent on the ratio of surfactant to co-surfactant (Sur/Co–S ratio) as well as the oil percentage.²⁷ The previous work assessed the impact of every variable while maintaining others constant. Yet, it is unable to predict and analyse the interactions between those factors. Furthermore, the characterization of DDZ complex is dramatically different from DDZ. Therefore, an applicable method is designed not only to improve this deficiency but also optimize the preparation of DDZ complex SMEDDS.

Response surface methodology (RSM) serves as an experimental design strategy which could be appropriately used to achieve the goal of exploring the relationships between several selected variables and one or more response variables as well as optimizing the responses. Basically, central composite designs (CCD) is one of the most reliable RSM categories. The factorial experiment, centre point and axial points are the core components of CCD. In recent years, CCD has been proved to successfully optimize the preparation conditions of drug delivery systems like liposomes, nanoparticles, microspheres.^28–30

Recent years, there are some reported articles about nanoparticles of DDZ. Due to the fact that DDZ has poor hydrophilicity and lipophilicity, by improving one of the aspects, the oral availability of DDZ would change. Zhang³¹ has successfully developed daidzein–phospholipid complex loaded lipid nanocarriers to improve its lipophilicity. The oral bioavailability has improved 6.87-fold compared with that of free DDZ. However, we prepared DDZ–NCT complex to enhance its hydrophilicity.

The objectives of this study were: (1) prepared and characterized the DDZ–NCT complex (2) optimized formulation of SMEDDS containing DDZ–NCT complex with RSM (3) assessed absorption in vitro intestinal and pharmacokinetics studies of DDZ–NCT complex and its SMEDDS.

Specifically, the DDZ–NCT complex was prepared with solvent evaporation method.^32–34 Solid-state characterizations of the DDZ–NCT complex were carried out by PXRD, FT-IR, DSC and TGA. Then, formulation optimization of SMEDDS (DDZ–NCT complex) was achieved by central composite designs (CCD). Finally, intestinal absorption study was confirmed by recording the accumulative amount of DDZ in 120 min and oral bioavailability was examined through pharmacokinetic study.

2. Materials and methods

2.1 Materials

DDZ was provided by Qingze Co., Ltd. (Nanjing, China, 99%). NCT and Castol oil were supplied by Sinopharm Chemical Reagent Co., Ltd (Shanghai China). Polyethylene glycol 400 (PEG400), dehydrated alcohol, HPLC-grade methanol and acetonitrile were purchased from Shanghai Chemical Reagents Institute (Shanghai, China). Cremophor RH40 was purchased from BASF (Germany). Heparin sodium was obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Deionized water was prepared by a Milli-Q purification system (Millipore, U.S.A.).

2.2 Preparation of DDZ–NCT complex

Based on the preliminary experiment and analysis, NCT, glycine, benzamide, citric acid, L-proline, L-cysteine, L-(+)-tartaric acid were formed with DDZ, respectively. As a result, NCT was chosen to be the best complex former.

Solvent evaporation method was used to construct DDZ–NCT complex and detailed below: DDZ and NCT (the molar ratio was 1 : 4) were dissolved in 5 mL methanol at 50 °C for 2 h under constant stirring (600 rpm). Then the solvent was evaporated slowly at room temperature. The DDZ–NCT complex was dried in a vacuum desiccator over silica gels at room temperature. In addition, the physical mixture of DDZ and NCT (DDZ–NCT-PM) DDZ–NCT-PM was obtained by mixing DDZ and NCT (the molar ratio was 1 : 4) in a vortex mixer for 10 min.

2.3 Solid-state properties

2.3.1 Powder X-ray diffraction (PXRD). Powder X-ray diffraction patterns of all the samples were obtained on an X-ray diffractometer (D₈ advance, BRUKER-AXS) with Cu Kα radiation. The following experimental parameters were used: tube voltage (40 kV), tube current (40 mA), scanning interval of 2θ (3–50°) and scan rate (2° min⁻¹).

2.3.2 Fourier transform infrared spectroscopy (FT-IR). Potassium bromide diffuse reflectance mode was adopted to collect the samples. FT-IR spectrometer (Nicolet 6700, Thermo Fisher, USA) and software (Spectrum 6.1 version) were used to characterize the IR spectra of all the samples.

2.3.3 Thermal properties. Differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA) were conducted using SDT Simultaneous DSC-TGA (Q600, TA Instruments Co., New Castle, DE, USA). Samples weighing 3 mg were heated from 20 °C to 600 °C at a heating rate of 10 °C min⁻¹ under nitrogen purge.⁴

2.4 Preparation of SMEDDS (DDZ–NCT complex)

DDZ–NCT complex was dissolved in PEG400 at 37 °C, followed by adding certain amount of Cremophor RH40 and Castol oil. Each formulation contains 5% (w/w) DDZ. Afterward, the compositions were mixed gently, then a transparent and homogeneous preparation was formed.

2.5 Formulation optimization of SMEDDS

The SMEDDS was optimized by central composite design-response surface methodology as the factors were designed in ESI Table 1.†

2.5.1 Central composite design. Previous experiments showed that the drug content, the oil percentage (oil%, w/w) as well as the Sur/Co–S ratio play an important role in affecting the properties of SMEDDS.³⁵ In this study, for the purpose of reaching the regulations of medical use, the drug content was kept as a fixed concentration. According to our previous study, the range of two variables were set as follows: oil percentage (X₁): 10–30%; Sur/Co–S ratio (X₂): 1–4 (ESI Table 1†). Three responses include emulsification time (Y₁), droplet size (Y₂) and equilibrium solubility (Y₃) are observed and analysed to evaluate the qualities of SMEDDS. The experimental design consists of 9 batches (1 centre point, 4 star points and 4 factorial points) and 4 replicated batches (centre points) arranging 13 experiments (ESI Table 2†). The experimental error was estimated by the replicated batches.

In each trial, the data acquired from the three responses was fitted to third-order quadratic model. The mathematical models were expressed as follows:

Y = b₀ + b₁X₁ + b₂X₂ + b₃X₁² + b₄X₂² + b₅X₁X₂ + b₆X₁²X₂ + b₇X₁X₂² (1)

where Y is the measured response, X₁ and X₂ correspond to the studied factors, b₀ is an intercept, and b₁–b₇ are the regression coefficients.

Analysis of data is accomplished with nonlinear estimation using STATISTICA 6.0 software.

2.6 Intestinal absorption of SMEDDS (DDZ–NCT complex)

Male Sprague-Dawley (SD) rats (250 ± 20 g) were provided by Shanghai Jiao Tong University for in vitro intestinal absorption experiment. All animals used in the whole study were conducted in accordance with the internationally accepted principles for laboratory animal use and care in the European Community guidelines (EEC Directive of 1986; 86/609/EEC) and approved by the Experimental Animal Ethics Committee of Shanghai Jiao Tong University. Rat gut sacs model was used to study the intestinal absorption of insoluble drugs which was described previously.³⁶ Briefly, after being fasted but free access to water for 24 h, the rats were anesthetized by injecting 3% sodium pentobarbital (30 mg kg⁻¹). 10 cm small intestine was excised quickly and rinsed with Krebs–Ringer's (K–R) solution. Pure DDZ, DDZ–NCT-PM, DDZ–NCT complex and SMEDDS (DDZ–NCT complex), containing 20 μg DDZ respectively, were dispersed in 1 mL K–R solution and were added into the gut sacs separately. Then the gut sacs were tied by suture before they were kept in the K–R buffer medium at 37 °C with continuous ventilation (5% CO₂ and 95% O₂). Samples (5 mL) were taken at 0, 5, 15, 30, 45, 60, 90 and 120 min, and the equal volume (5 mL) of K–R solution was added to the medium. Then the sample solution was centrifuged at 10 000 rpm for 10 min and filtered through a 0.45 μm filter. The drug concentration in the sample was analysed by HPLC and the apparent permeability coefficients (P_app) and absorption promoting ratio (R) were calculated as follows:³⁷

P_app = dQtAC₀/dt (2)

R = P_app(sample)/P_app(control) (3)

where dQt/dt was the drug permeation rate, A was the transport area and C₀ was the initial DDZ concentration in the donor solution.

2.7 Pharmacokinetics study

12 male SD rats, weighting 250 ± 10 g, were fasted for 24 h but with freely water. They were randomly divided into four groups to decrease individual variance and separately treated with 4 different formulations. They were administered orally at a single dose of DDZ 10 mg kg⁻¹. Blood samples were drawn from orbit at the following points: 0.25, 0.5, 0.75, 1, 1.5, 2, 3, 4, 6, 8 and 12 h. After collecting, blood samples were transferred into the heparinized tubes and centrifuged at 3000 rpm for 10 min. The 100 μL of supernatant was drawn and mixed with 200 μL of acetonitrile by a vortex mixer for 5 min. Subsequently, the mixture was centrifuged at 12 000 rpm for 10 min and the supernatant of blood samples was analysed by HPLC.

2.8 HPLC analysis

In this study, the concentration of DDZ was recorded by the HPLC equipment (LC-20A system, Shimadzu, Japan). The column is Dikma Dimonsil C₁₈ (150 mm × 4.6 mm, 5 μm, Dikma, Technologies). The mobile phase consists of methanol and ultrapure water with a ratio of 50 : 50. 20 μL of each sample was monitored at 249 nm.

3. Results and discussion

3.1 Solid-state properties

3.1.1 Powder X-ray diffraction (PXRD). PXRD is an effective method in the identification of the solid-state characters and structure at the crystalline lattice level. The diffraction data in PXRD characterization can confirm whether the complex was formed and whether it was a simple physical mixture. The PXRD patterns of DDZ, NCT, DDZ–NCT-PM and DDZ–NCT complex were displayed in Fig. 2.

Fig. 2 PXRD patterns of DDZ (a), NCT (b), DDZ–NCT-PM (c) and DDZ–NCT complex (d).

The PXRD patterns of DDZ had peaks at 2θ values of 6.9°, 8.48°, 10.38°, 12.94°, 15.90°, 16.98°, 24.48°, 25.16°, 25.60°, 26.44°, 28.00° and 28.72° (Fig. 2a). The PXRD patterns of NCT were similar to that reported paper which had peaks at 2θ values of 14.80°, 19.50°, 22.68°, 23.12°, 23.36°, 25.36°, 25.82°, 27.30°, 28.40°, 32.54° and 36.94° (Fig. 2b).³⁸ Compared with the diffractogram of DDZ and NCT, the PXRD patterns of DDZ–NCT-PM (Fig. 2c), showed peaks at 2θ values of 10.36°, 14.74°, 15.88°, 16.98°, 24.50°, 25.28°, 25.78° and 27.26°. Apart from the diffractogram of the physical mixture was largely dominated by reflections from DDZ, which was owning to the small crystallinity of NCT itself, there was no new peak. That is to say, the curve of DDZ–NCT-PM did not change basically. However, the diffractogram of the DDZ–NCT complex (Fig. 2d) was obviously distinguishable from DDZ, NCT and DDZ–NCT-PM, and showed new peaks at 2θ values of 8.54°, 19.42°, 19.56°, 19.90°, 22.74° and 39.54° whose diffraction intensity range from 606 to 1026. The different pattern of the complex in the PXRD could confirm the change of crystalline lattice, implying the DDZ–NCT complex was a new compound which was not a simple physical mixture of DDZ and NCT. Moreover, the sharp peaks were contrasting from steamed buns peak which was the peculiar peaks of amorphous state, demonstrating the complex was crystalline polymer.³⁹

3.1.2 Fourier transform infrared spectroscopy (FT-IR). IR is a very powerful spectroscopic technique in testing the chemical conformation of compounds. It also can be a very effective tool in probing solid-state at the intermolecular and intramolecular levels. In this experiment, IR characterization was used to identify the most possible mode of combination between DDZ and NCT. The FT-IR spectra for DDZ, NCT, and the DDZ–NCT complex were presented in Fig. 3. The IR spectrum of DDZ (Fig. 3a) showed absorption bands at 3440 cm⁻¹ and 3223 cm⁻¹ which were assigned to the –OH group stretching vibration, band at 1631 cm⁻¹ was assigned to the C=O group stretching vibration.⁴⁰ After the construction of DDZ with NCT, these bands were shift to 3367 cm⁻¹, 3167 cm⁻¹ and 1631 cm⁻¹ respectively (Fig. 3c). The two characteristic shift towards lower frequencies revealed that the OH group and C=O group might take part in hydrogen bonds. The IR spectrum of NCT (Fig. 3b) had peaks at 1618 cm⁻¹ and 1394 cm⁻¹ which corresponded to the C=N group and C–N group stretching vibrations.⁴² After the construction of DDZ with NCT, the 1618 cm⁻¹ band shift to 1594 cm⁻¹ (Fig. 3c) which might be caused by hydrogen bond association of pyridine structure of NCT. The change demonstrates the guanidine structure of NCT formation of the hydrogen bonds. The change of IR spectrum implied the successful construction of the DDZ–NCT complex once again and the changes were predicted due to the formation of hydrogen bonds between DDZ and NCT.

Fig. 3 IR spectra for DDZ (a), NCT (b) and DDZ–NCT complex (c).

3.1.3 Thermal properties. SDT Simultaneous DSC-TGA is used to gain the information of DSC and TGA with the same sample at the same time to eliminate the influence resulted from weight, sample homogeneity and heating rate consistency. In this experiment, SDT Simultaneous DSC-TGA characterization was carried out to investigate the relationship between melting point and decomposition, in order to further confirm whether the complex was solvate or hydrate.

DSC was performed to determine the melting point which was a basic thermodynamic property for a specific compound. The DSC traces for DDZ, NCT, and the DDZ–NCT complex were shown in Fig. 4. DDZ exhibited the melting peak at 333.08 °C (Fig. 4b) and NCT showed a melting peak at 126.56 °C (Fig. 4c) which were accord (Fig. 4b) and NCT showed a melting peak at 126.56 °C (Fig. 4c) which were accord with the report, thermal behaviour had confirmed the reliability of the results of DSC.⁴¹ By contrast, the complex displayed a distinguishable melting peak at 123.41 °C (Fig. 4a) which was lower than DDZ and NCT.

Fig. 4 DSC curve of DDZ–NCT complex (a), DDZ (b) and NCT (c).

TGA was carried out to monitor the decomposition temperature of samples. The TGA curves for DDZ, NCT, and the DDZ–NCT complex were shown in Fig. 5. The decomposition temperature of DDZ and NCT was 338.53 °C (Fig. 5a), 207.41 °C (Fig. 5b) respectively, which were accord with the reported thermal behaviour and confirmed the reliability of the results of TGA. By contrast, the complex displayed a distinguishable decomposition temperature at 189.64 °C (Fig. 5c).

Fig. 5 TGA curve of DDZ (a), NCT (b) and DDZ–NCT complex (c).

The result was not only confirmed the melting point and decomposition temperature of the complex but also further demonstrated the complex showed a different thermal properties compared with DDZ and NCT. The distinguishable thermal properties of the DDZ–NCT complex indicated the existence of different forms. Besides, from the curves of DSC and TGA, it was clear that the decomposition temperature of DDZ–NCT complex (189.64 °C) was behind its melting point (123.41 °C). The SDT information of the complex indicated no weight loss before their melting point, proving that there have no residual solvent and the new compound was neither a solvate nor a hydrate.

3.2 Central composite design

TATISTICA 6.0 software was used to analyse all the data. The tested parameters of all the different SMEDDS formulations are shown in Table 1. The polynomial fitting correlation coefficient (r) of emulsification time, droplet size and solubility for third order polynomial was 0.970, 0.956 and 0.978, respectively. The fitting significance value P was less than 0.05. The response surface was drawn in accordance with the fitted third order eqn (4)–(6) by software Statistics.

Y_t = 0.023X₁²X₂ + 0.433X₂²X₁ − 0.039X₁² − 6.037X₂² − 2.829X₁X₂ + 3.756X₁ + 35.826X₂ − 41.626 (4)

Y_DS = −0.055X₁²X₂ − 0.409X₂²X₁ + 0.18X₁² + 8.387X₂² + 4.21X₁X₂ − 8.652X₁ − 62.899X₂ + 125.46 (5)

Y_S = 0.016X₁²X₂ − 0.05X₂²X₁ − 0.044X₁² + 1.126X₂² − 0.36X₁X₂ + 1.089X₁ + 0.207X₂ + 5.611 (6)

Table 1 Experimental responses and result of central composite design (n = 3)

No.	X1 (oil%, w/w)	X2 (Km)	Y1 (emulsification time, min)	Y2 (droplet size, nm)	Y3 (solubility, mg mL^−1)
1	12.93	1.44	5	24.05 ± 0.01	10.62 ± 0.18
2	12.93	3.56	7	19.39 ± 0.01	12.06 ± 0.07
3	27.07	2.50	10	32.28 ± 0.05	5.47 ± 0.02
4	27.07	2.50	20	26.14 ± 0.02	8.26 ± 0.03
5	10.00	1.00	5	18.54 ± 0.02	11.76 ± 0.08
6	30.00	4.00	8	38.30 ± 0.05	6.55 ± 0.01
7	20.00	2.50	8	24.40 ± 0.07	9.50 ± 0.17
8	20.00	2.50	13	25.04 ± 0.04	10.04 ± 0.05
9	20.00	2.50	6	25.51 ± 0.06	10.14 ± 0.04
10	20.00	2.50	4	25.19 ± 0.05	9.55 ± 0.15
11	20.00	2.50	7	19.16 ± 0.01	9.32 ± 0.13
12	20.00	2.50	6	24.18 ± 0.01	8.43 ± 0.12
13	20.00	2.50	5	24.46 ± 0.04	9.11 ± 0.03

---

3.2.1 Influence of formulation composition factors on the emulsification time. Emulsification time plays an important role in assessing SMEDDS. Fig. 6A illustrates the relationship between the emulsification time and the two factors (oil percentage and Sur/Co–S ratio). Seeing from the response surface and contour plot, when the Sur/Co–S ratio varies from 1 to 3 and the oil percentage is no more than 20%, the emulsification time can be approximately 4 min or less.

Fig. 6 (A) The response surface of emulsification time (Y₁). (B) The contour plot of emulsification time (Y₁). (C) The response surface of droplet size (Y₂). (D) The contour plot of droplet size (Y₂). (E) The response surface of solubility (Y₃). (F) The contour plot of solubility (Y₃).

3.2.2 Influence of formulation composition factors on droplet size. Droplet size is another crucial factor to assess SMEDDS. Previous study showed that smaller droplet size enhances the release rate of the formulation and increases the absorption of drugs in oral administration.^43,44 Fig. 6C presents the response surface for droplet size and demonstrated the relationship between the droplet size and the two factors (oil percentage and Sur/Co–S ratio). When the Sur/Co–S ratio varies from 1.5 to 4.0 and oil percentage is less than 20%, the droplet size is nearly 20 nm.

3.2.3 Influence of formulation composition factors on equilibrium solubility. In order to load more drugs, the SMEDDS formulation was further studied. It was previously showed that oil percentage and Sur/Co–S ratio are able to change drug solubility in SMEDDS. Fig. 6E shows that when Sur/Co–S ratio kept constant, solubility was decreased with the enhancement of the oil percentage. According to the preliminary research, it was demonstrated that the sequence of DDZ–NCT solubility was Castol oil < Cremophor RH40 < PEG400. Hence, the ratio of Castol oil may be the primary factor that influences the solubilizing capability of the SMEDDS. As a result, decreasing the oil percentage improves solubility of the drug.

3.2.4 Evaluation of the optimized formulation. The oil percentage (X₁) and Sur/Co–S ratio (X₂) were optimized to be 15% and 2.5, respectively. In detail, the final optimized formulation of SMEDDS (5% of DDZ complex) was: Castol oil of 15.0%, Cremophor RH40 of 60.7% and PEG400 of 24.3%. The emulsification time, droplet size and solubility of this optimal formulation (experimental value) compared with the results calculated from the fitted third order equation above (predicted value), all bias are lower than 5%, which indicated an exactly predictability of the model (ESI Table 3†).

3.3 In vitro intestinal absorption

In vitro intestinal absorption was performed to determine the characteristics of DDZ, DDZ–NCT complex, SMEDDS (DDZ) and the SMEDDS (DDZ–NCT complex) through rat small intestine. The cumulative amount–time line chart was displayed in Fig. 7. The ultimate cumulative amount of the DDZ in pure DDZ, DDZ–NCT complex, SMEDDS (DDZ) and the SMEDDS (DDZ–NCT complex) in 2 hours were 3.22 ± 0.75 μg, 6.18 ± 1.12 μg, 10.31 ± 0.88 μg and 14.21 ± 1.02 μg respectively (SMEDDS (DDZ–NCT complex) > SMEDDS (DDZ) > DDZ–NCT complex > pure DDZ). The apparent permeability coefficient (P_app) and the absorption enhancement ratios (R) of the pure DDZ, DDZ–NCT complex, SMEDDS (DDZ) and the SMEDDS (DDZ–NCT complex) were summarized in Table 2. The P_app of DDZ–NCT complex (16.19 ± 2.75 nm s⁻¹) and SMEDDS (DDZ) (29.72 ± 5.05 nm s⁻¹) were both higher than that of the pure DDZ (7.87 ± 1.34 nm s⁻¹), but were both lower than that of the SMEDDS (DDZ–NCT complex) (36.80 ± 4.95 nm s⁻¹). The results indicated that DDZ–NCT complex promoted DDZ transport through rat small intestine in 2 h, which might be attributed to the successful construction of the complex enhanced the water solubility and dissolution rate of DDZ.

Fig. 7 Cumulative amount–time profiles of DDZ in different formulations in the transport across rats' small intestine in vitro (n = 4).

Table 2 Transport properties of DDZ in different formulations through rates small intestine in vitro (n = 4)

Formulations	Papp (×10^−6 cm s^−1)	Absorption promoting ratio (R)
a Significant difference compared with DDZ (p < 0.01).
DDZ (control)	7.87 ± 1.34	1.00
DDZ–NCT complex	16.19 ± 2.75^a	2.06
SMEDDS (DDZ)	29.72 ± 5.05^a	3.78
SMEDDS (DDZ–NCT complex)	36.80 ± 4.95^a	4.68

---

Since the low surface tension of the microemulsion after self-microemulsified, it is easier for the SMEDDS go through the hydration layer of the gastrointestinal wall, so that the drug is directly in contact with gastrointestinal epithelium and is easily absorbed.^44,45 Hence, the SMEDDS (DDZ) could increase the drug transportation in the rat small intestine and enhance the bioavailability. With the combination of complex and SMEDDS, we predicted that complex SMEDDS could maximum improve the bioavailability of DDZ. The result also confirmed this prediction.

3.4 Bioavailability study

An in vivo absorption study was undertaken to determine whether DDZ–NCT complex, SMEDDS (DDZ) and SMEDDS (DDZ–NCT complex) could increase the gastrointestinal absorption compared with pure drug after oral administration. Fig. 8 showed that plasma concentration profiles of DDZ in SMEDDS (DDZ–NCT complex) group showed significantly improvement of drug absorption than the complex group and SMEDDS group, let alone the control group. Pharmacokinetic parameters of the maximum plasma concentration (C_max) and the corresponding time (T_max) for DDZ following oral administration were shown in Table 3. The area under the concentration–time curve (AUC_{0→12 h}) was estimated according to the linear trapezoidal rule.

Fig. 8 Plasma concentration–time profiles of different formulations after oral administration to rats (n = 3, 10 mg kg⁻¹).

Table 3 Pharmacokinetic parameters of different formulations after oral administration to the rats (n = 3)^a

Parameters	DDZ	DDZ–NCT complex	SMEDDS (DDZ)	SMEDDS (DDZ–NCT complex)
a Significant difference compared with DDZ (p ≤ 0.05).
AUC0→12 h (ng h mL^−1)	321.9 ± 23.6	674.2 ± 48.7^a	1199.1 ± 96.3^a	1830.7 ± 118.8^a
Cmax (ng mL^−1)	165.2 ± 10.7	230.4 ± 21.1^a	472.0 ± 47.6^a	828.8 ± 50.9^a
Tmax (h)	0.33 ± 0.14	0.28 ± 0.21	0.19 ± 0.096	0.14 ± 0.096
CL (L h^−1)	2.43 ± 0.24	2.31 ± 0.33	2.10 ± 0.36	2.00 ± 0.37

---

Compared with the pure DDZ, the increased values of the AUC_{0→12 h} of DDZ–NCT complex were summarized as they could improve the amount of DDZ absorbed by the rat small intestine with the conducive effect of NCT. The existence of NCT modulated the intermolecular interactions between DDZ and water, and resulted in the complex showing improved water solubility. On the other hand, the formation of DDZ–NCT complex could improve its solubility and dissolution rate of DDZ. The advantage of pharmaceutical complex was that it could increase its bioavailability reported previously.^46–48 That is to say, the bioavailability study, for one thing, confirming the superiority of DDZ–NCT complex was identical with the advantage of pharmaceutical complex, for another, it could provide more possibility for the optimization of other formulations.

Compared with the pure DDZ, the increased values of the AUC_{0→12 h} of SMEDDS (DDZ) were summarized as they could improve the amount of DDZ absorbed by the rat small intestine with the help of microemulsion. The possible reason is that after oral administration, SMEDDS spontaneously forms plenty of 40 nm particles by the slightly moving of the gastrointestinal tract. These microemulsion droplets have a larger surface area and easily pass through the hydration layer on the mucosal side of the intestine, the drug can be directly contacted with the gastrointestinal epithelial cells.⁴⁹ On the other hand, after being digested by acid and trypsin, emulsion droplets of smaller particle size are formed.⁵⁰ To a large extent, the solubility and transmembrane capacity of DDZ in the digestive fluids are all largely increased. Thereby SMEDDS also can enhance the oral bioavailability of drugs.

In pharmacokinetic parameters of SMEDDS (DDZ–NCT complex), AUC_{0→12 h} and C_max were 1830.7 ± 118.86 ng h mL⁻¹ and 828.8 ± 50.98 ng mL⁻¹, respectively, compared with the DDZ–NCT complex group which were 674.2 ± 48.7 ng h mL⁻¹ and 230.4 ± 21.1 ng mL⁻¹, respectively. The values of AUC_{0→12 h} and C_max have enhanced 3-fold. This depends on the advantage of SMEDDS which significantly enhances the lipophilicity of the formulation. Compared with the SMEDDS group which were 1199.1 ± 96.3 ng h mL⁻¹ and 472.0 ± 47.6 ng mL⁻¹, respectively. The values of AUC_{0→12 h} and C_max have enhanced nearly 2-fold. Drugs have released from SMEDDS, the complex form of drug enhances the hydrophilicity of the formulation during absorption in intestinal. Because the complex SMEDDS is comprised of DDZ–NCT and SMEDDS, it can obviously further improve the bioavailability of drugs than any of its single part. Compared with pure drug which were 321.9 ± 23.63 ng h mL⁻¹ and 165.2 ± 10.65 ng mL⁻¹, respectively. SMEDDS (DDZ–NCT complex) significantly enhanced the values of AUC_{0→12 h} and C_max of drug more than 6-fold, which indicate that SMEDDS (DDZ–NCT complex) can maximize the bioavailability of drugs in the rat small intestine.

4. Conclusions

In this study, a novel SMEDDS (DDZ–NCT complex) was successfully constructed. The DDZ–NCT complex was characterized by PXRD, IR, DSC and TGA. The results of solid-state characterization imply the existence of hydrogen bonding between DDZ and NCT. Solubility studies revealed the complex dramatically increased the solubility and dissolution rate of DDZ. Central composite design-response surface methodology was applied to optimize the SMEDDS formulation. In the meantime, the effects of oil percentage and Sur/Co–S ratio on the emulsification time, droplet size and equilibrium solubility were investigated. The predicted values were in close agreement with the observed responses for the optimum formulation, indicating the convincingly predictability of the optimization procedure. Investigation of the small intestinal absorption and pharmacokinetic study confirmed that the combination of complex and SMEDDS could significantly enhance rat small intestinal absorption in vitro and improve the oral bioavailability of DDZ respectively. According to all discussed above, SMEDDS (DDZ–NCT complex) might be a positive and potential system to delivery drugs with poorly water solubility as DDZ and so forth.

References

1. X. Zhao, Q. Shen and Y. Ma, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2011, 879, 113–116.
2. F. J. Möller, P. Diel, O. Zierau, T. Hertrampf, J. Maaß and G. Vollmer, Toxicol. Lett., 2010, 196, 142–153.
3. L. Coward, N. C. Barnes, K. D. Setchell and S. Barnes, J. Agric. Food Chem., 1993, 41, 1961–1967.
4. X. Fan, J. Chen and Q. Shen, Int. J. Pharm., 2013, 458, 296–304.
5. J. Lu and S. Rohani, Org. Process Res. Dev., 2009, 13, 1269–1275.
6. N. Blagden, M. De Matas, P. Gavan and P. York, Adv. Drug Delivery Rev., 2007, 59, 617–630.
7. A. V. Trask, W. S. Motherwell and W. Jones, Cryst. Growth Des., 2005, 5, 1013–1021.
8. P. Vishweshwar, J. A. McMahon, M. L. Peterson, M. B. Hickey, T. R. Shattock and M. J. Zaworotko, Chem. Commun., 2005, 4601–4603.
9. S. L. Morissette, Ö. Almarsson, M. L. Peterson, J. F. Remenar, M. J. Read, A. V. Lemmo, S. Ellis, M. J. Cima and C. R. Gardner, Adv. Drug Delivery Rev., 2004, 56, 275–300.
10. S. L. Childs, L. J. Chyall, J. T. Dunlap, V. N. Smolenskaya, B. C. Stahly and G. P. Stahly, J. Am. Chem. Soc., 2004, 126, 13335–13342.
11. G. Bettinetti, M. R. Caira, A. Callegari, M. Merli, M. Sorrenti and C. Tadini, J. Pharm. Sci., 2000, 89, 478–489.
12. J. A. Zerkowski, J. C. MacDonald and G. M. Whitesides, Chem. Mater., 1997, 9, 1933–1941.
13. Ö. Almarsson, M. B. Hickey, M. L. Peterson, S. L. Morissette, S. Soukasene, C. McNulty, M. Tawa, J. M. MacPhee and J. F. Remenar, Cryst. Growth Des., 2003, 3, 927–933.
14. S. Datta and D. J. Grant, Nat. Rev. Drug Discovery, 2004, 3, 42–57.
15. I. D. Oswald, D. R. Allan, P. A. McGregor, W. S. Motherwell, S. Parsons and C. R. Pulham, Acta Crystallogr., Sect. B: Struct. Sci., 2002, 58, 1057–1066.
16. P. Vishweshwar, J. A. McMahon, J. A. Bis and M. J. Zaworotko, J. Pharm. Sci., 2006, 95, 499–516.
17. Ö. Almarsson and M. J. Zaworotko, Chem. Commun., 2004, 1889–1896.
18. K. Tanaka and F. Toda, Chem. Rev., 2000, 100, 1025–1074.
19. A. V. Trask, W. S. Motherwell and W. Jones, Chem. Commun., 2004, 890–891.
20. K. Kohli, S. Chopra, D. Dhar, S. Arora and R. K. Khar, Drug Discovery Today, 2010, 15, 958–965.
21. Y. Lu, J. Qi and W. Wu, Curr. Drug Metab., 2012, 13, 396–417.
22. P. P. Constantinides, Pharm. Res., 1995, 12, 1561–1572.
23. C. W. Pouton, Adv. Drug Delivery Rev., 1997, 25, 47–58.
24. N. Shah, M. Carvajal, C. Patel, M. Infeld and A. Malick, Int. J. Pharm., 1994, 106, 15–23.
25. D. Hauss, Am. Pharm. Rev., 2002, 5, 22–28.
26. Q. Shen, X. Li, D. Yuan and W. Jia, Chem. Pharm. Bull., 2010, 58, 639–643.
27. A. S. Zidan, O. A. Sammour, M. A. Hammad, N. A. Megrab, M. J. Habib and M. A. Khan, J. Pharm. Sci., 2007, 96, 2409–2423.
28. A. Billon, B. Bataille, G. Cassanas and M. Jacob, Int. J. Pharm., 2000, 203, 159–168.
29. C. Gløgård, G. Stensrud, R. Hovland, S. L. Fossheim and J. Klaveness, Int. J. Pharm., 2002, 233, 131–140.
30. E. C. Gil, A. I. Colarte, B. Bataille, J. L. Pedraz, F. Rodríguez and J. Heinämäki, Int. J. Pharm., 2006, 317, 32–39.
31. Z. Zhang, Y. Huang, F. Gao, H. Bu, W. Gu and Y. Li, Nanoscale, 2011, 3, 1780–1787.
32. W. Wu, Y. Wang and L. Que, Eur. J. Pharm. Biopharm., 2006, 63, 288–294.
33. X. G. Wu, G. Li and Y. L. Gao, Chem. Pharm. Bull., 2006, 54, 977–981.
34. N. Qiao, M. Li, W. Schlindwein, N. Malek, A. Davies and G. Trappitt, Int. J. Pharm., 2011, 419, 1–11.
35. A. S. Zidan, O. A. Sammour, M. A. Hammad, N. A. Megrab, M. J. Habib and M. A. Khan, J. Pharm. Sci., 2007, 96, 2409–2423.
36. J. Wu, Q. Shen and L. Fang, Drug Dev. Ind. Pharm., 2013, 39, 1010–1019.
37. M. Agüeros, L. Ruiz-Gatón, C. Vauthier, K. Bouchemal, S. Espuelas, G. Ponchel and J. Irache, Eur. J. Pharm. Sci., 2009, 38, 405–413.
38. D. J. Berry, C. C. Seaton, W. Clegg, R. W. Harrington, S. J. Coles, P. N. Horton, M. B. Hursthouse, R. Storey, W. Jones and T. Friscic, Cryst. Growth Des., 2008, 8, 1697–1712.
39. A. Jayasankar, D. J. Good and N. Rodríguez-Hornedo, Mol. Pharmaceutics, 2007, 4, 360–372.
40. H. Cheng, H. Liu, Y. Zhang and G. Zou, J. Lumin., 2009, 129, 1196–1203.
41. A. Tudor, M. Davies, C. Melia, D. Lee, R. Mitchell, P. Hendra and S. Church, Spectrochim. Acta, Part A, 1991, 47, 1389–1393.
42. J. Lu and S. Rohani, Org. Process Res. Dev., 2009, 13, 1269–1275.
43. B. K. Kang, J. S. Lee, S. K. Chon, S. Y. Jeong, S. H. Yuk, G. Khang, H. B. Lee and S. H. Cho, Int. J. Pharm., 2004, 274, 65–73.
44. M. Grove, G. P. Pedersen, J. L. Nielsen and A. Müllertz, J. Pharm. Sci., 2005, 94, 1830–1838.
45. E. A. Mueller, J. M. Kovarik, J. B. van Bree, J. Grevel, P. W. Lücker and K. Kutz, Pharm. Res., 1994, 11, 151–155.
46. N. Schultheiss and A. Newman, Cryst. Growth Des., 2009, 9, 2950–2967.
47. R. Chadha, A. Saini, P. Arora and S. Bhandari, Crit. Rev. Ther. Drug Carrier Syst., 2012, 29, 183–218.
48. N. Shan and M. J. Zaworotko, Drug Discovery Today, 2008, 13, 440–446.
49. A. J. Humberstone and W. N. Charman, Adv. Drug Delivery Rev., 1997, 25, 103–128.
50. P. C. de Smidt, M. A. Campanero and I. F. Trocóniz, Int. J. Pharm., 2004, 270, 109–118.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra22767h

---