Production of a new platform based on fumed and mesoporous silica nanoparticles for enhanced solubility and oral bioavailability of raloxifene HCl

Abstract

The purpose of the present study was to compare mesoporous and fumed silica nanoparticles (NPs) to enhance the aqueous solubility and oral bioavailability of raloxifene hydrochloride (RH). Mesoporous silica NPs (MSNs) and fumed silica NPs were used by freeze‐drying or spray‐drying methods. MSNs were obtained with different ratios of cetyltrimethylammonium bromide. Saturation solubility of the NPs was compared with the pure drug. The optimised formulation was characterised by scanning electron microscopy (SEM), X‐ray diffraction (XRD) and differential scanning calorimetry. The pharmacokinetic studies were done by oral administration of a single dose of 15 mg/kg of pure drug or fumed silica NPs of RH in Wistar rats. MSNs enhanced the solubility of RH from 19.88 ± 0.12 to 76.5 μg/ml. Freeze‐dried fumed silica increased the solubility of the drug more than MSNs (140.17 ± 0.45 μg/ml). However, the spray‐dried fumed silica caused about 26‐fold enhancement in its solubility (525.7 ± 93.5 μg/ml). Increasing the ratio of silica NPs enhanced the drug solubility. The results of XRD and SEM analyses displayed RH were in the amorphous state in the NPs. Oral bioavailability of NPs showed 3.5‐fold increase compared to the pure drug. The RH loaded fumed silica NPs prepared by spray‐drying technique could more enhance the solubility and oral bioavailability of RH.

1 Introduction

Osteoporosis is a disease of skeletal system and is associated with fragility fracture at the hip, spine and wrist. Raloxifene hydrochloride (RH) is a second‐generation non‐steroidal benzothiophene, selective estrogen receptor modulator approved by the Food and Drug Administration (FDA) in 1997 for the prevention and treatment of post‐menopausal bone loss at a dose of 60 mg/day. It acts as an oestrogens agonist in bone. RH inhibits vertebral bone loss by inhibiting the activity of cytokines, which stimulate bone re‐sorption [1]. RH possesses 2% absolute bioavailability orally. Although RH has low solubility and bioavailability, it is highly permeable [2]. RH is rapidly absorbed from the gastrointestinal tract and undergoes extensive first‐pass glucuronidation. It pertains to class II of the Biopharmaceutics Drug Disposition Classification System (BCS), where the drug is characterised by high permeability, poor solubility and high metabolism [2]. As a result, it is very important to improve its water solubility for gastrointestinal delivery and treatment of osteoporosis, which provides economic benefit to drug manufacturer and consumers. Reduction of the active pharmaceutical ingredient (API) particle size by micronisation or nanonisation was among the first techniques that could be applied in development, leading to the commercialisation of poorly water‐soluble compounds. There are some more recently developed series of technologies based on the principle of converting the API from its crystalline state into a high energy state (amorphous) using technologies such as spray drying or lyophilisation. Although these types of formulations often result in increased bioavailability, the potential intrinsic instability of these high energy systems may limit their attractiveness towards commercialisation due to limited shelf‐life claims and extended timelines. Given these potential limitations and the more than ever water‐insoluble nature of the compounds in development, there is a profound need for new formulation platforms. Nanosize drug delivery systems generally focus on formulating bioactive molecules in biocompatible nanosystems such as magnetic nanoparticles (NPs) [3, 4], dendrimers [5], albumin NPs [6, 7] and so on. One of these carriers is silica NP, which is intensively studied in drug delivery. Silica carrier system increases the dissolution rate of APIs to the highest possible level (i.e. molecular) without the stability‐related issues of amorphous dispersions or solid solutions [8, 9]. The principle of dissolution enhancement originates in the adsorption of an API onto the surface of the carrier material in a molecular manner [10]. Unlike any other carrier material, which might also have a high specific surface to allow adsorption, the silica material described in this study has a uniquely designed structure. In the first step, the API is loaded onto the silica via an ‘incipient wetness technique’ using the appropriate organic solvent [11]. Dissolution rate enhancement of the API is caused by the fact that the van der Waals forces and hydrogen bonds that keep the drug molecules adsorbed onto the surface and inside the pore system of the silica carrier are easily broken up from the contact with water. Consequently, detached drug molecules are released in a ‘dissolved’ state and are available for absorption in the gastrointestinal tract [12]. The characteristics of the pores in silica such as pore diameter, specific surface area and pore volume can be tailored to accommodate the ideal environment for the drug molecules, taking into account, for example, their molecular size. This technique has been used for enhanced water solubility and oral bioavailability of many drugs of class II of BCS like celecoxib [13], ezetimib [14], atorvastatin [15], fenofibrate [16], itraconazole [17], meloxicam [18], cyclosporin A [19] and so on.

Silica is ‘generally recognised as safe’ by the United States FDA. Recently, FDA has approved a special form of silica for human clinical trial for targeted molecular imaging [20, 21]. Studies on the systematic toxicity of MSNs after intravenous injection of single and repeated doses to mice indicated low in‐vivo toxicity of MSNs from different aspects of clinical features, pathological examinations, mortalities and blood biochemical indexes [22]. MSNs are mainly excreted through faeces and urine following different administration routes [23]. These NPs exhibit a three‐stage degradation behaviour in simulated body fluid [24], favourable for cargo release. Huang et al. [25] studied organ distribution of silica NPs. They showed that after intravenous administration of MSNs they were mainly presented in the liver, spleen and lung (>80%). In‐vivo behaviour of MSNs was greatly affected by their shapes. So that, the short‐rod MSNs were easily accumulated in the liver, while long‐rod MSNs distributed in the spleen. PEG modification of MSNs caused the higher content in the lung. MSNs were mainly excreted by urine and faeces, and the clearance rate of MSNs was primarily dependent on the particle shape, where short‐rod MSNs had a more rapid clearance rate than long‐rod MSNs in urine and faeces. Haematology, serum biochemistry and histopathology results indicated that MSNs would not cause significant toxicity in vivo [25].

To the best of our knowledge, there is no report on the comparison of different types of mesoporous silica with fumed silica in enhancement of RH solubility. The aim of the present study was to invent a new method for enhancing the raloxifene solubility and bioavailability by silica NPs. In this regard, mesoporous silica and fumed silica were compared and the effect of different process parameters on the water solubility of raloxifene and its pharmacokinetic parameters were evaluated.

2 Materials and methods

2.1 Materials

The RH powder was kindly gifted by Iran Hormone Research Laboratories (Tehran, Iran). Aerosil^® 200 the hydrophilic fumed silica NPs with a specific surface area of 200 m² /g were provided from EVONIC Industries (Germany). Tetraethyl orthosilicate (TEOS), sodium hydroxide, cetyltrimethylammonium bromide (CTAB) and hydrochloric acid were purchased from Sigma‐Aldrich (USA). Ethanol, sodium dihydrogen phosphate dehydrate and all other reagents and chemicals were of analytical grade and obtained from Merck Chemical Company (Germany).

2.2 Synthesis of mesoporous silica NPs (MSNs)

The mixture of CTAB (used as surfactant) (0.5 and 1.5 g), NaOH (2.0 M, 1.75 ml) and water (120 ml) was heated to 80°C for 30 min and the pH was adjusted to 12.3. TEOS (2.335 g) was added rapidly to the obtained clear solution by injection, which caused the production of a white precipitate rapidly. The reaction was continued for 2 h and then the product was centrifuged and washed with water and methanol for three times. An acid extraction was performed using mixture of methanol (100 ml), concentrated HCl (1 ml) and previously prepared sample (1.0 g) at 60°C for 6 h using hot plate [19]. The solid product was washed with methanol and water after centrifugation. The prepared MSNs were oven dried at 60°C and used for further characterisations.

2.3 Drug loading in MSNs

Drug loading in MSNs was carried out in two types of MSNs obtained from 0.5 or 1.5 g of CTAB in three ratios of MSNs/RH, i.e. 1:1, 3:1 and 5:1. For this purpose, 40 mg of RH was dissolved in 1 ml of methanol, then 40, 120 or 200 mg of each type of MSNs was added to the clear solution of the drug and stirred at 400 rpm for 24 h. At the end of the loading procedure, which lasted 24 h the drug loading was assessed by sampling 1 ml of the loading mixture. The sample was centrifuged in an Eppendorf Ultracentrifuge (Germany) at 7000 rpm for 30 min. The supernatant was collected and passed through Millipore filter (0.22 µm) followed by the determination of the drug content using ultraviolet spectrophotometer (UV mini 1240, Shimadzu, Japan) at 287 nm [26].

2.4 Preparation of freeze‐dried dispersions of fumed silica and RH

Two different mixtures of Aerosil^® and RH were prepared in water or DMSO. The ratio of Aerosil^® to RH was 5:1 and the dispersion was freeze dried (freeze drier Christ, α 2–4, Germany).

2.5 Preparation of spray‐dried dispersions of fumed silica and RH

Aerosil^® and RH mixtures in water were prepared in different mass ratios of 1:1, 3:1 and 5:1. RH‐loaded fumed silica solid dispersions were produced by Büchi Mini Spray Dryer B290 (Büchi Labortechnik AG, Flawil, Switzerland) under the following set of conditions; inlet temperature of 80°C, outlet temperature of 50°C, the pump flow rate was 5% (2.5 ml/min) and aspirator was set at 85%. The spray‐dried dispersions were kept in a desiccator cabinet until used for further studies.

2.6 Particle size analysis

The mean particle size and particle size distribution of the NPs were measured by photon correlation spectroscopy (Zetasizer, ZEN 3600, Malvern Instrument, UK). The results were achieved via a He‐Ne laser beam at 658 nm at a fixed angle of 90°.

2.7 Saturated solubility

Saturated solubility studies were performed on the pure drug and NPs. About 500 mg of NPs were dispersed in 2 ml of de‐ionised water, which could no longer dissolve it and the mixture was stirred at 400 rpm for 24 h. The resulting suspensions were centrifuged (Eppendorf, Germany) at 10,000 rpm for 10 min. Then the supernatants were filtered through a 0.22 µm filter membrane and the saturated solubility of each formulation was determined spectrophotometrically (UV‐mini 1240, Shimadzu, Kyoto, Japan) at λ _max  = 286 nm [27].

2.8 Drug content determination of the formulations

In order to calculate the entrapment efficiency (EE) of RH within the RH loaded NPs, appropriate weight of each formulation which had theoretically 1 mg of drug was dissolved in 5 ml of methanol and the absorption of the samples was determined at 287 nm by using UV–visible spectrophotometer. EE and loading percent of RH‐loaded NPs were calculated by means of the following equations:

$$\mathrm{EE}\left(\text{\%}\right)=\frac{\mathrm{analysed}\mathrm{weight}\mathrm{of}\mathrm{entrapped}\mathrm{drug}}{\mathrm{theoritical}\mathrm{weight}\mathrm{of}\mathrm{drug}\mathrm{loaded}\mathrm{in}\mathrm{the}\mathrm{system}}\times 100$$

$$\mathrm{Loading}\left(\text{\%}\right)=\frac{\mathrm{entrapped}\mathrm{drug}\mathrm{in}\mathrm{dispersion}}{\mathrm{total}\mathrm{amount}\mathrm{of}\mathrm{polymer}\mathrm{and}\mathrm{drug}}\times 100$$

The samples were analysed in triplicate.

2.9 Dissolution rate studies

Suitable amounts of drug loaded silica NPs dispersion containing <15% of the saturated solubility of RH were dispersed in 1 ml of water and transferred to the dialysis bag (molecular weight cut off 12000 Da) then placed into 24 ml of de‐ionised water while stirring at 400 rpm on the magnetic stirrer (IKA‐WERKE, Model RT 10 power, Japan). The samples were withdrawn at certain time intervals, analysed spectrophotometrically at 286 nm and returned back to the medium to keep the volume of the dissolution test constant. In addition, the dissolution efficiency up to 3 h (DE) was calculated according to the following equation:

$$\mathrm{D}{\mathrm{E}}_t\frac{{\int }_0^t{y}_t\cdot \mathrm{d}t}{y_{100}\cdot t}$$

In this equation, ${\int }_0^t{y}_t\cdot \mathrm{d}t$ is the area under the curve of drug dissolved percent (y_t ) in an interval of time t, expressed in percentage value and $y_{100}\cdot t$ is the rectangle area considering 100% dissolution in the same time t.

2.10 Scanning electron microscopy (SEM)

Morphological analysis was performed on pure RH, RH loaded in MSNs and fumed silica NPs prepared by freeze drying or spray drying. Initially, a double‐sided adhesive tape was used to mount the samples on an aluminium stub. Then to make electrically conductive the sputter‐coating procedure by a Hitachi Ion Sputter (E‐1030) was carried out under vacuum with gold in an argon atmosphere. The morphology of the samples was examined by an SEM (Hitachi F41100, Japan).

2.11 Solid‐state characterisation of optimum formulation of RH loaded NPs

As the spray‐dried fumed silica NPs showed the best results in the enhancement of solubility and dissolution rate of RH, further studies were performed on these NPs.

2.12 X‐ray powder diffraction (XRPD)

To evaluate the crystallinity of the samples, the XRPD test was executed. Chosen samples were gently packed into the sample holder of a Bruker, X‐ray diffractometer (D8Advance, Germany). The test was done for pure drug, fumed silica and RH loaded in fumed silica NPs using monochromatic Cu Kα‐radiation at 40 kV and 30 mA over a range of (2θ) between 5° and 60° with an angular increment of 0.02°/s.

2.13 Thermal analysis (DSC)

Thermal analysis of the drug, the fumed silica and RH loaded in fumed silica NPs were performed by differential scanning calorimetry (DSC) test to describe potential mismatches. The samples (3–6 mg) were placed in an aluminium pan. In order to have good sealing, an aluminium lid was crimped. A DSC device (822e Mettler‐Toledo, Switzerland) equipped with a refrigerated cooling system for recording the thermograms was employed. The DSC was calibrated for temperature and enthalpy with indium standard. The aluminium pan was heated in nitrogen atmosphere between 20 and 300°C at a heating rate of 10°C/min. A similar empty pan was used as the reference. The results were analysed by the software (STARe Ver. 10.00 Mettler Toledo, Switzerland) of the device to record the melting point of the sample.

2.14 Animal studies

2.14.1 Pharmacokinetic studies

Male albino Wistar rats weighing 180–220 g were obtained from animal house of Isfahan University of Medical Sciences. Animals were randomly divided into two groups of six animals each. They were kept under the standard conditions with the temperature of 18–22°C and 55–65% relative humidity and 12 h light/dark cycles. The animals were allowed enough access to food and water ad libitum during the test except on the evening before dosing, when all food was removed and withheld until 4 h after dosing. The experiments were performed according to the ethical guidelines of Institutional Animal Ethics Committee guidelines provided by the Committee for the Purpose of Control and Supervision of Experiments on Animals (CPCSEA) of the Health Ministry (ethical review number: 943752).

2.14.2 Drug administration and plasma sampling

The rats were kept fasted 12 h before and 4 h after drug administration and just had free access to the water. Then they were divided into two groups and were orally administered a single dose of 15 mg/kg of the pure drug or RH loaded in spray‐dried fumed silica NPs. Then the rats were anaesthetised with ether and blood sampling was done by inserting of a heparinised capillary into the retro‐orbital vein to get 0.5 ml of blood at a time interval of 0.083, 0.25, 0.5, 1.0, 2, 4, 8, 12 and 24 h. The samples were centrifuged at 5,000 rpm for 15 min and the plasma was separated and frozen immediately at −20°C until analysis by the high‐performance liquid chromatography (HPLC) method.

2.14.3 Preparation of standard solution

A series of standard solutions of RH were prepared in methanol in the range of 0.1–100 μg/ml. Standard calibration samples were prepared by adding 100 μl standard solution of drug and 300 μl of acetonitrile to 100 μl of blank plasma. Final standard RH concentrations in plasma were 0.02–20.0 μg/ml.

2.14.4 Plasma sample preparation

To 100 μl of plasma sample, 100 μl of methanol and 300 μl of acetonitrile were added and vortex‐mixed for 20 s. The samples were centrifuged at 15,000 rpm for 10 min to separate the denatured proteins. About 20 μl of the supernatant was injected into the HPLC for analysis. The pharmacokinetic parameters including C _max, T _max, AUC_0–24 and k _el were calculated and compared to the pure drug and the NPs.

2.14.5 5. HPLC analysis method

A reversed‐phase HPLC method was used to measure the plasma concentration of the drug. HPLC (Waters, 5.5, USA) device with two pump and UV detector were used. Mobile phase consisted of acetonitrile, ammonium acetate (pH 4.0, 0.05 M) (50: 50%v/v) with a flow rate of 0.8 ml/min to elute the drug. The mobile phase was filtered through 0.45 μm nylon filters (Millipore, USA) and the samples were analysed at λ _max  = 289 nm.

2.15 Data analysis and statistical evaluation

All analyses were done at least three times and the values were reported as the mean ± SD. p  < 0.05 was assessed as statistical significant and the statistical analysis was done by SPSS software (SPSS, Inc., Chicago, IL, version 20).

3 Results and discussion

3.1 Saturation solubility and drug loading percent

RH loaded‐MSNs were prepared by an inactive method. Loading of RH in MSNs was studied using formulation variables including changes in CTAB concentration (during the MSNs synthesis) and MSNs to RH ratio (during drug loading in MSNs). The results of the incorporation of RH in MSNs are shown in Table 1.

Table 1.

Particle size, loading efficiency and saturated solubility of RH loaded in MSNs in comparison with pure, untreated drug

Formulation code	CTAB concentration, %	MSNs to RH ratio	Particle size (nm) ± SD	Loading efficiency (%) ± SD	Saturated solubility of RH, µg/ml
CTAB0.5 MSN1	0.5	1	210 ± 40	75.0 ± 2.0	39.6
CTAB0.5 MSN3	0.5	3	245 ± 35	77.9 ± 1.0	44.5
CTAB0.5 MSN5	0.5	5	230 ± 70	86.3 ± 6.2	73.8
CTAB1.5 MSN1	1.5	1	310 ± 25	79.9 ± 3.3	31.5
CTAB1.5 MSN3	1.5	3	290 ± 60	75.8 ± 3.1	45.4
CTAB1.5 MSN5	1.5	5	340 ± 20	77.7 ± 0.9	76.5
pure RH	—	—	6100 ± 170	—	19.88

As shown in Table 1, changes in CTAB concentration and MSNs to RH ratio did not have meaningful effect on drug loading efficiency in MSNs (p  > 0.05). Although the loading efficiency of RH in MSNs was not changed significantly with changes in CTAB content (p  > 0.05) (Table 1), saturated solubility was affected by the ratio of MSNs to drug. Increasing in MSNs content increased saturated solubility of RH (p  < 0.05). In the best condition when the CTAB concentration was 1.5% and the ratio of MSNs to RH was 5, the saturated solubility was changed from 19.88 to about 76.5 µg/ml, i.e. 3.8‐fold enhancement in RH solubility.

When using fumed silica (Aerosil) the results were different. The freeze‐dried RH‐Aerosil mixture could enhance saturation concentration significantly (p  < 0.05) compared with pure RH especially when the mixture was prepared in DMSO (Table 2). In the case of using DMSO for soaking the drug and Aerosil, saturation solubility of the drug was increased to 140.17 ± 0.45 µg/ml but when water was used for preparation of mixture of the drug and Aerosil, the solubility of the drug was almost half of the situations using DMSO and was about 54.36 ± 0.33 µg/ml (Table 2). Due to the inherent nature of this method in which the drug is not wasted, the loading efficiency is 100% and the entire drug is trapped in Aerosil structure during the freeze‐drying process (Table 2).

Table 2.

Particle size, loading efficiency and saturated solubility of RH loaded in fumed silica NPs (Aerosil) in 1:5 ratio by freeze‐drying method in comparison with pure drug

Sample	Particle size (nm) ± SD	Saturated solubility of RH (µg/ml) ± SD	Loading efficiency (%) ± SD
untreated RH	6100 ± 170	19.88 ± 0.12	—
Aerosil‐RH‐Water	1950 ± 190	54.36 ± 0.33	99.8 ± 0.1
Aerosil‐RH‐DMSO	1600 ± 75	140.17 ± 0.45	100.0 ± 0.0

The results of saturation solubility and drug loading EE of NPs obtained from spray drying method with different ratios of Aerosil/RH are shown in Table 3.

Table 3.

Particle size, water solubility, dissolution efficiency (DE), EE and drug loading of RH/Aerosil NPs achieved by spray drying technique

Aerosil to RH ratio	Particle size (nm) ± SD	Water solubility (µg/ml) ± SD	DE3 (%) ± SD	Entrapment efficiency (%) ± SD	Drug loading (%) ± SD
1	509.8 ± 43.6	331.5 ± 10.9	34.39 ± 3.1	35.4 ± 1.3	17.7 ± 0.6
3	267.2 ± 16.5	486.8 ± 26.3	44.5 ± 3.9	45.4 ± 0.6	18.9 ± 6.6
5	330.0 ± 64.9	525.7 ± 93.5	52.3 ± 2.3	70.3 ± 6.5	11.7 ± 1.1

Table 3 shows data pertaining to solubility of RH in water, suggesting enhancement of solubility of the drug in aqueous medium. The results of this study revealed poor water solubility of pure RH (19.88 ± 0.12 µg/ml). All the formulations showed noticeable increasing of the solubility of the drug by Aerosil in spray drying technique in comparison to the pure RH (p  < 0.05) (Table 3). The results of Table 3 shows with increasing in carrier/drug ratio from 1 to 5, extensive improvement of water solubility of RH was observed (p  < 0.05).

Mesoporous silica with low density and high specific area seems to be an attractive drug delivery system with high loading capacity. MSNs have both internal and external surfaces cause selective functionalisation, high pore volume and good thermal stability [19, 28]. Drug amorphisation can occur by interacting with porous media and drug molecules, resulting in very different behaviour than those in bulk phase [29]. Incorporation of the drugs in mesoporous silica can lead to the enhancement of drug solubility and dissolution rate in comparison to the pure drug [30].

In the best condition when the CTAB concentration was 1.5% and the ratio of MSNs to RH was 5, the saturated solubility of RH changed from 19.88 to about 76.5 µg/ml, i.e. 3.8‐fold by MSNs (Table 1). Recently, mesoporous silica has been used frequently for drug delivery applications [31]. The drug in mesoporous silica system is in the amorphous state because of space confinement. In this system the drug molecules are prevented by the pores to fold together so they cannot arrange themselves in a critical nucleation size [31, 32, 33]. Here, the ratio of drug/carrier can be regarded as a practical approach [32]. As can be expected, change in crystalline habit to amorphous can increase the solubility of the drug [34]. Our results were in accordance to He et al. [35] studies who used MSNs for enhanced solubility of paclitaxel.

By using fumed silica (Aerosil) the freeze‐dried RH‐Aerosil prepared in DMSO, saturation solubility of the drug was increased to 140.17 ± 0.45 µg/ml but when water was used for preparation of mixture of the drug and Aerosil the solubility of the drug was almost half of the situations using DMSO and was about 54.36 ± 0.33 µg/ml (Table 2).

The results revealed that all formulations showed noticeable increasing of the solubility of the drug by Aerosil in spray drying technique in comparison to the pure RH (p  < 0.05) (Table 3). Aerosil has shown promising results in enhancement of the solubility of poorly water‐soluble drugs like CoQ‐10 [36] and Repaglinide [37].

3.2 Particle size, polydispersity index and morphology of NPs

The particle size and morphology of MSNs were studied by DLS (Malvern Nano Zetasizer) and SEM methods. SEM images of pure RH, RH‐MSNs sample (CTAB_1.5 MSN₅), freeze‐dried Aerosil‐RH NPs prepared by DMSO (Aerosil‐RH‐DMSO) and spray‐dried Aerosil‐RH in 5:1 ratio are shown in Fig. 1. As shown in Fig. 1 a, SEM images revealed that RH had cubic crystalline morphology. The micrographs of RH‐MSNs showed the particle size of nearly 300 nm and a cylindrical shape for drug‐loaded NPs (Fig. 1 b). SEM graphs presented MSNs had a nearly mono‐dispersed cylindrical structure. MSNs size detected by SEM had good correlation with DLS results (Table 1). In DLS results the particle size of these NPs was determined to be around 300–350 nm with the PDI of 0.5. Zeta potential of these NPs was around 40 mV (Table 1).

Fig. 1.

SEM micrographs of

(a) RH crystals

(b) Mesoporous silica NPs (CTAB_1.5 MSN₅)

(c) Freeze‐dried Aerosil‐RH NPs prepared by DMSO (Aerosil‐RH‐DMSO)

(d) Spray‐dried Aerosil‐RH in 5:1

Fig. 1 c shows the SEM micrographs of freeze‐dried Aerosil‐RH‐DMSO particles. As this figure indicates, the particles are quite agglomerated with high particle size of about 1600 ± 75 nm. Although this method could reduce the particle size of the untreated drug (6100 ± 170 nm) significantly, but freeze drying the Aerosil NPs inherently caused higher particle size than the spray drying technique (Table 3). The freeze drying of Aerosil‐RH using DMSO caused significant smaller particles than water (Table 2) which related to the high solubility of RH in DMSO while water could just disperse the drug without its solubility. As it is shown in Fig. 1 d, the spray‐dried NPs of Aerosil‐RH are in spherical shape with wrinkled surface, which is due to the spray drying technique and suggesting the total conversion of the crystalline state of RH into amorphous form. The results of particle size determination of different ratios of Aerosil‐RH spray‐dried NPs analysed by DLS method are seen in Table 3. The ratios of 3:1 and 5:1 exhibited a desirable particle size with no significant difference (p  > 0.05). However, the particle size of 1:1 ratio was significantly higher than the other two ratios (p  < 0.05).

3.3 Dissolution studies

Dissolution profiles in all studied formulations compared with the pure RH are shown in Fig. 2. In‐vitro drug release profiles from RH loaded‐MSNs are shown in Fig. 2 a. The release of RH from MSNs in most formulations was without initial burst release and was in a sustained release pattern during 120 h of the study. As it is seen in Fig. 2 a, the CTAB_0.5 MSN₃ had the fastest release rate among the studied formulations of MSNs. However, even this formulation could not release more than 10% of the loaded drug after 120 h. In other words, while the MSNs could significantly enhance saturated solubility of RH (Table 1) but drug dissolution rate from these NPs was very slow and the worse disadvantage of these formulations was their release efficacy. The poor solubility of RH as the rate‐limiting step of drug dissolution and entrapment of the drug in the porous structure of MSNs may explain this result. On the other hand, RH loaded freeze‐dried Aerosil prepared in DMSO with the ratio of 5:1 (Aerosil/drug) could better enhance the dissolution rate of the drug than those obtained by water and pure drug (Fig. 2 b). However, the drug release rate at comparable times (120 h) was not significantly enhanced by this method in comparison to the MSNs (the best formulation, i.e. CTAB_0.5 MSN₃) and in both methods drug release was about 10% after 120 h (Figs. 2 a and b).

Fig. 2.

RH release profiles from

(a) Mesoporous silica NPs

(b) Freeze‐dried Aerosil‐RH NPs prepared by DMSO (Aerosil‐RH‐DMSO)

(c) Spray‐dried Aerosil‐RH in 5:1 (n  = 3)

As shown in Fig. 2, the pure RH was dissolved only about 8% after 120 h. All studied spray‐dried Aerosil formulations enhanced the dissolution rate of the drug significantly in all studied time points (p  < 0.05). The ratio of 5:1 showed better dissolution rate than others and there was a significant increase in the rate of the dissolution between the formulations by the increase in drug to carrier ratio (p  < 0.05).

Table 3 displays that the DE₃ % (dissolution efficiency at 3 h) of the spray‐dried Aerosil‐RH formulations were within the range of 34.39 ± 3.1–52.3 ± 2.3 and increasing the amount of Aerosil^® led to significant enhancement of dissolution efficiency (p  < 0.05), while the DE of pure RH was only 5.28 ± 2.6%. It has been postulated that increasing the quantity of Aerosil^® improved drug dissolution rate (p  < 0.05). There was a significant enhancement of DE% in spray‐dried formulations in comparison to the pure drug (p  < 0.05).

However, the results of Fig. 2 showed that the RH loaded freeze‐dried Aerosil NPs could not increase the dissolution rate of the drug more than MSNs, while the spray‐dried Aerosil‐RH in 5:1 ratio of Aerosil to drug could completely dissolve the drug within 2 h (Fig. 2 c).

The results of Table 1 also showed almost 2‐fold increase in dissolution rate of RH loaded in CTAB_0.5 MSN₃ compared to the pure drug. Mesoporous silica as a vehicle supplies a stable hydrophilic matrix with a nano‐porous structure so it can improve the drug dissolution rate by hindering drug crystallisation [38, 39]. On the other hand, drug release can be impressed greatly by pore morphology [40]. So, interconnected pore structures cause faster drug release than those with unconnected pore networks [41], which entrap the drug in mesoporous channels and retard the drug release rate [40].

Another reason for enhanced dissolution rate of RH from MSNs is higher surface area of NPs [42, 43, 44].

On the other hand, it is expected that the hydrophilic characteristics of the Aerosil^® 200 cause better wetting of the drug particles due to diminishing the interfacial tension between the NPs and dissolution medium. However, the results of Fig. 2 showed that the RH loaded freeze‐dried Aerosil NPs could not increase the dissolution rate of the drug more than MSNs while, the spray‐dried Aerosil‐RH in ratio of 5:1 could completely dissolve the drug within 2 h (Fig. 2 c). Shen et al. [45] also reported that co‐spray drying of ibuprofen with MSNs resulted in enhanced dissolution rate of this drug.

3.4 Solid‐state characterisation of the spray‐dried aerosil‐RH

From the results obtained, the spray‐dried fumed silica NPs in the ratio of 5:1 of Aerosil^® to RH was selected as the optimum formulation which showed the highest enhancement of solubility (Table 3) and dissolution rate of RH (Fig. 2). Therefore, further studies including XRPD and DSC were performed on these NPs.

3.4.1 X‐ray powder diffraction

The XRPDs of RH, spray‐dried Aerosil‐RH in 5:1 ratio and Aerosil^® are shown in Fig. 3. The X‐ray diffractometry pattern of RH suggests the numerous distinctive peaks at 2θ values of 15.78, 22.68, 25.87 and 46.1° corresponding to its crystalline nature (Fig. 3 a). On the contrary to untreated RH, spray‐dried Aerosil‐RH and Aerosil^® showed very low intrinsic peaks in their X‐ray diffractometry patterns, which indicates the transformation of the drug from the completely crystalline form to almost amorphous form. This results is in the agreement of other studies in which spray‐drying technique was used to produce NPs [45, 46, 47] and it is in coincidence of DSC thermographs (Fig. 4). Consequently, it can be determined that fumed silica NPs were in charge of non‐crystalline state of the drug [48].

Fig. 3.

X‐ray diffraction diffractograms of

(a) Pure RH

(b) Spray‐dried Aerosil‐RH in 5:1 ratio

(c) Pure Aerosil^®

Fig. 4.

DSC thermograms of

(a) Pure RH

(b) Spray‐dried Aerosil‐RH (5:1) NPs

(c) Pure Aerosil^®

3.4.2 Differential scanning calorimetry

To further confirm the non‐crystalline state of the drug in spray‐dried Aerosil^® NPs, DSC measurements were done. The DSC thermographs of RH powder, Aerosil^® and spray‐dried NPs are shown in Fig. 4. RH showed a sharp single endothermic peak which could be ascribed to melting point of the drug at 263.7°C (Fig. 4 a) and for the NPs four peaks were observed at 186.9, 235.0, 246.8 and 311.3°C. However, no melting endothermic peak that was seen in pure RH was seen obviously in NPs thermogram (Fig. 4 b). For pure Aerosil^® also no considerable endothermic peak was seen (Fig. 4 c).

No melting endothermic peak that was seen in pure RH was seen in NPs thermogram (Fig. 4 b). This indicates that the drug was effectively changed into the amorphous state in NPs with no trace of crystallinity when RH and Aerosil^® were co‐spray dried. For pure Aerosil^® also no considerable endothermic peak was seen (Fig. 4 c) as its melting point is reported to be 1610°C [49]. The small peak seen at 232°C may be related to the impurities. It is believed that when the drug is dispersed in an amorphous carrier, the melting point of the drug cannot be detected afterwards [50]. Furthermore, spray‐drying technique allows the solvent to evaporate slowly so the solute molecules can rearrange into an amorphous state [51] like azithromycin solid dispersions with fumed Aerosil silica NPs [52].

3.5 Pharmacokinetic studies

The plasma concentration profiles of the drug administered orally in the form of pure or as spray‐dried solid dispersion in fumed silica NPs with the ratio of 5:1 are shown in Fig. 5. In addition, the pharmacokinetic parameters are seen in Table 4. As this table shows the C _max of the pure drug enhanced to about 8‐fold in fumed silica NPs although the T _max remained unchanged. In addition, the AUC_0–24 of the fumed silica NPs showed about 3.5‐fold increase in comparison to the pure drug.

Fig. 5.

Plasma concentration–time profiles of RH after oral administration of 15 mg/kg RH powder and equivalent amounts of spray‐dried Aerosil‐RH 5:1 solid dispersion in rats (n = 6)

Table 4.

Pharmacokinetic parameters of pure RH in comparison with the optimum spray‐dried Aerosil‐RH 5:1 solid dispersion

RH type	C max, µg/ml	T max, min	K el, h−1	AUC0–24, µg.h/ml
Pure RH	0.30 ± 0.07	240.0 ± 0.0	0.004 ± 0.004	2.50 ± 0.35
Aerosil‐RH	2.42 ± 0.84a	240.0 ± 0.0a	0.012 ± 0.007a	8.76 ± 5.51a

Each value represents the mean ± SD (n  = 3).

^a p <0.05 shows significant difference compared to the pure RH powder.

Even though several research works have been done to enhance the solubility and bioavailability of RH by formulating it as tablets, microemulsions, microspheres, transdermal patches and polymeric NPs, still it requires to design the formulation having combined advantages like sustained release and avoiding first pass metabolism of RH. With 26‐fold increment in solubility of RH by fumed silica NPs with the ratio of 5:1 in spray drying method (Table 3), it is expected that the enhanced solubility causes significant increase in the bioavailability of this drug. Yang et al. [53] also reported that after oral administration of 30 mg/kg of the pure RH in rats the C _max of 3.23 ± 0.15 μg/ml and an AUC of 18.10 ± 2.63 μg.h.ml⁻¹ were obtained.

In another study, self‐emulsifying drug delivery system of RH was developed and oral bioavailability of RH in rats showed 4‐fold and 2.5‐fold higher AUC_0‐∞ than RH suspension (control) and marketed product, respectively [54]. Also by solvent diffusion method, nanostructured lipid carriers of RH were produced, in vivo pharmacokinetic study showed 3.75‐fold enhancement in its bioavailability [55].

4 Conclusion

Fumed silica could enhance RH solubility and dissolution rate better than MSNs. Solid dispersion of RH in fumed silica as a carrier by spray‐drying technique had more tremendous potential for improving the drug solubility than the freeze‐drying method. Besides, it can be considered as a prospective method to enhance the dissolution rate of the poorly water‐soluble drugs. In this study, the ratio of 5:1 of Aerosil NPs to RH presented the fastest dissolution rate with 26‐fold enhancement in solubility of the pure drug, showed the smallest particle size and the highest EE between the other studied ratios. The designed silica NPs loaded with RH enhanced oral bioavailability of this drug to about 3.5‐fold. Finally, it may be concluded that spray drying technique using fumed Aerosil, colloidal silicon dioxide, can help and support formulators to bring more API innovations to the market.