European Journal of Pharmaceutical Sciences 75 (2015) 101–113

Contents lists available at ScienceDirect

European Journal of Pharmaceutical Sciences journal homepage: www.elsevier.com/locate/ejps

Electrospun polycaprolactone nanofibers as a potential oromucosal delivery system for poorly water-soluble drugs Tanja Potrcˇ, Saša Baumgartner, Robert Roškar, Odon Planinšek, Zoran Lavricˇ, Julijana Kristl, Petra Kocbek ⇑ University of Ljubljana, Faculty of Pharmacy, Aškercˇeva 7, 1000 Ljubljana, Slovenia

a r t i c l e

i n f o

Article history: Received 4 February 2015 Received in revised form 3 April 2015 Accepted 5 April 2015 Available online 21 April 2015 Keywords: Electrospinning Nanofibers Polycaprolactone Poorly soluble drugs Oromucosal drug delivery

a b s t r a c t The number of poorly water-soluble drug candidates is rapidly increasing; this represents a major challenge for the pharmaceutical industry. As a consequence, novel formulation approaches are required. Furthermore, if such a drug candidate is intended for the therapy of a specific group of the population, such as geriatric or pediatric, the formulation challenge is even greater, with the need to produce a dosage form that is acceptable for specific patients. Therefore, the goal of our study was to explore electrospun polycaprolactone (PCL) nanofibers as a novel nanodelivery system adopted for the oromucosal administration of poorly water-soluble drugs. The nanofibers were evaluated in comparison with polymer films loaded with ibuprofen or carvedilol as the model drugs. Scanning electron microscopy revealed that the amount of incorporated drug affects the diameter and the morphology of the nanofibers. The average fiber diameter increased with a higher drug loading, whereas the morphology of the nanofibers was noticeably changed in the case of nanofibers with 50% and 60% ibuprofen. The incorporation of drugs into the electrospun PCL nanofibers was observed to reduce their crystallinity. Based on the morphology of the nanofibers and the films, and the differential scanning calorimetry results obtained in this study, it can be assumed that the drugs incorporated into the nanofibers were partially molecularly dispersed in the PCL matrix and partially in the form of dispersed nanocrystals. The incorporation of both model drugs into the PCL nanofibers significantly improved their dissolution rates. The PCL nanofibers released almost 100% of the incorporated ibuprofen in 4 h, whereas only up to 77% of the incorporated carvedilol was released during the same time period, indicating the influence of the drug’s properties, such as molecular weight and solubility, on its release from the PCL matrix. The obtained results clearly demonstrated the advantages of the new nanodelivery system compared to the drug-loaded polymer films that were used as the reference formulation. As a result, electrospinning was shown to be a very promising nanotechnology-based approach to the formulation of poorly water-soluble drugs in order to enhance their dissolution. In addition, the great potential of the produced drug-loaded PCL nanofiber mats for subsequent formulation as oromucosal drug delivery systems for children and the elderly was confirmed. Ó 2015 Elsevier B.V. All rights reserved.

1. Introduction The development of novel formulations and delivery approaches for poorly water-soluble drugs is a common challenge with modern pharmaceuticals. The main reason for this is the large proportion of drug candidates that are poorly water soluble. However, when these drugs need to be administered to specific populations, e.g., paediatric or geriatric, the technological ⇑ Corresponding author at: University of Ljubljana, Faculty of Pharmacy, Aškercˇeva cesta 7, 1000 Ljubljana, Slovenia. Tel.: +386 1 47 69 620; fax: +386 1 42 58 031. E-mail address: [email protected] (P. Kocbek). http://dx.doi.org/10.1016/j.ejps.2015.04.004 0928-0987/Ó 2015 Elsevier B.V. All rights reserved.

challenges associated with their poor water solubility are accompanied by additional challenges associated with the acceptability of the final dosage form with these patients. Nanofibers represent one of the newest and very promising nanomaterials for a number of applications. They have already demonstrated important applicability in biomedicine, where numerous studies have already described in the field of tissue engineering, wound healing as well as drug delivery (Rošic et al., 2013). Immediate or modified drug release can be achieved by the selection of a polymer for nanofiber production and the manner of the drug loading. The drug can be either incorporated into the polymer matrix of the nanofibers or bound to their surfaces (Huang et al., 2003; Meinel et al., 2012; Rošic et al., 2012). Both biodegradable

102

T. Potrcˇ et al. / European Journal of Pharmaceutical Sciences 75 (2015) 101–113

and non-degradable, either natural or synthetic, polymers can be used to control the drug release via diffusion alone or by a combination of diffusion and fiber degradation. Furthermore, the proper selection of the polymer can ensure the optimal combination of mechanical and biomimetic properties for the nanofibers (Pelipenko et al., 2013a; Sill and von Recum, 2008). The special characteristics of nanofibers, in addition to their very small diameters, include a high surface-to-volume ratio, a very high porosity, a small pore size, a good mechanical strength and a diversity in the surface functionalities (Huang et al., 2003; Pelipenko et al., 2013b). The drug release can be tuned for a specific application by changing the fiber’s composition, production technology or the process parameters, resulting in a different fiber diameter and/or porosity (Bertoncelj et al., 2014; Sill and von Recum, 2008). The application of nanofiber mats on mucosa results in fluid absorption, due to the presence of numerous nanometer-sized interfibrillar pores. This is, in addition to the large surface area of the nanofibers available to interact with the biosurface, the main mechanism for the adhesion of the nanofibers to the biosurfaces, making them one of the promising mucoadhesive drug delivery systems (Sill and von Recum, 2008). Electrospinning is the most common method for the production of fibers with diameters ranging from a few nanometers to a few micrometers from polymer solutions or melts (Frenot and Chronakis, 2003; Huang et al., 2003). This method is applicable to virtually any soluble or fusible polymer (Bhardwaj and Kundu, 2010; Sill and von Recum, 2008). Biodegradable and biocompatible polymers, such as polycaprolactone (PCL) (Fig. 1a), are good candidates for the preparation of nanofibers for applications in biomedicine. Such nanofibers represent a novel class of nanomaterials that is currently being investigated for drug-delivery applications (Dash and Konkimalla, 2012; Woodruff and Hutmacher, 2010). The oral mucosa is very interesting for the purposes of drug delivery. It provides a much more constant environment for drug absorption than the gastrointestinal environment, where the drug is not exposed to the harsh conditions in the gastrointestinal tract and the absorption across the oral mucosa can bypass the hepatic first-pass effect. Even though the permeability of the oral

mucosa is lower than the intestinal mucosa, the rich blood supply enables an effective drug absorption (Hearnden et al., 2012; Shakya et al., 2011). Oromucosal formulations are, due to their simple use and non-invasive application, widely accepted by patients (Lam et al., 2014). They are especially suitable for pediatric (Lam et al., 2014; Sattar et al., 2014) and geriatric patients (Illangakoon et al., 2014; Sattar et al., 2014), as well as for any other patient with swallowing or digestion difficulties (Illangakoon et al., 2014). The buccal and sublingual dosage forms are among the most commonly used oromucosal delivery systems (Lam et al., 2014; Patel et al., 2011). However, for the drug to be a suitable candidate for the formulation of an oromucosal delivery system it has to satisfy some key criteria that relate to the drug’s molecular weight, potency and water solubility. The molecular weight of drug candidates that are appropriate for oromucosal delivery should not exceed 800 Da (Lam et al., 2014) and the drug potency should be relatively high due the limited surface area that is available for the drug absorption. Usually, only doses up to a few milligrams can be efficiently absorbed through the oral mucosa. In addition, the dissolved drug might be washed away with saliva before it permeates through the mucosal membrane (Lam et al., 2014; Patel et al., 2011); therefore, an intense contact between the dosage form and the mucosa is highly advantageous for reducing the possibility of drug washout by the saliva. A drug delivered via the oromucosal route should not cause any local irritation at the application site, i.e., the oral mucosa (Lam et al., 2014). Furthermore, a sufficient aqueous solubility is necessary to allow the drug to diffuse through the mucus layer. The number of poorly water-soluble drug candidates has been increasing rapidly over recent years (Kawabata et al., 2011; Pouton, 2006); however, they are not usually suitable for the formulation of oromucosal drug delivery systems. The preparation of salts usually improves their solubility; however, drug molecules are better absorbed through the oral mucosa if they are in unionized form (Lam et al., 2014). Therefore, a nanotechnology-based approach, which would improve the dissolution rate and the solubility of such drugs, and would not affect their chemical properties, would be advantageous. The formulation of nanofibers represents one possible route to achieving these goals. During the electrospinning of a polymer solution the rapid evaporation of the solvent results in the instant formation of nanofibers and the entrapment of the drug in the polymer matrix or its deposition onto the nanofiber surface (Seif et al., 2015; Yu et al., 2010), resulting in a decreased drug mobility (Yu et al., 2010). The drugs are usually randomly encapsulated within ultrathin and flexible polymer nanofibers with a very high surface area available for contact with the application site. All these characteristics of nanofibers have a significant influence on the drug’s bioavailability (Yu et al., 2010). The aim of our research was to explore electrospun PCL nanofibers as a novel nanodelivery system intended for the oromucosal administration of poorly water-soluble drugs and to determine their potential advantages over polymer films. We chose two model drugs with similar hydro-lipophilic properties, but differing in their molecular weights, and investigated their influence on the nanofiber’s physical properties and drug release profiles, aiming to establish a correlation between the drug’s properties and the release characteristics of a PCL nanofiber-based delivery system.

2. Materials and methods 2.1. Materials

Fig. 1. Chemical structure of polycaprolactone (a), ibuprofen (b) and carvedilol (c).

The polycaprolactone (PCL) Mw 70,000–90,000 g/mol was purchased from Sigma–Aldrich, Germany. The sodium iodide

T. Potrcˇ et al. / European Journal of Pharmaceutical Sciences 75 (2015) 101–113 Table 1 The process parameters optimized for preparation of electrospun drug loaded PCL nanofibers. Polymer solution 10% 10% 10% 10% 10% 10% 10% 10% 10% 10% 10% 10% a

PCL/10%a PCL/15%a PCL/30%a PCL/40%a PCL/50%a PCL/60%a PCL/10%a PCL/15%a PCL/30%a PCL/40%a PCL/50%a PCL/60%a

ibuprofen ibuprofen ibuprofen ibuprofen ibuprofen ibuprofen carvedilol carvedilol carvedilol carvedilol carvedilol carvedilol

Flow rate (ml/h)

Tip to collector distance (cm)

Voltage (kV)

1.63 1.98 1.98 1.98 1.63 1.63 1.63 1.63 1.63 1.63 1.63 1.63

15 15 15 15 15 15 15 15 15 15 15 15

15.0 15.0 15.0 17.5 17.5 17.5 15.0 15.0 17.5 17.5 17.5 17.5

Based on the dry weight of polymer.

(P99.5%), chloroform, acetone (P99.5%), methanol (P99.8%), dichloromethane (P99.8%), sodium chloride (P99.5%), sodium hydroxide (P99.5%), potassium chloride (P99.5%), potassium dihydrogen phosphate (P99.5%), di-sodium hydrogen phosphate (P99.0%) and ortho-phosphoric acid (98%) were all obtained from Merck, Germany. The acetonitrile (P99.5%) was purchased from J.T. Baker, Netherlands. The ibuprofen and carvedilol were obtained from Sigma–Aldrich, Germany. The water was purified by reverse osmosis. 2.2. Methods 2.2.1. Preparations of polymer solutions The PCL solutions (10%, w/w) were prepared by dissolving the PCL in a mixture of chloroform and acetone in the weight ratio 75:25, then 0.03% (w/w) of sodium iodide was added and stirred with a magnetic stirrer for 3–4 h at room temperature. The ibuprofen and carvedilol were then added (10%, 15%, 30%, 40%, 50% or 60% w/w, based on the dry weight of the polymer) and the solutions were stirred for an additional 2–3 h prior to the electrospinning or film-casting process. 2.2.2. Electrospinning process The solution was placed in a plastic syringe fitted with a metal needle that had an inner diameter of 0.8 mm. A high voltage was applied between the needle and the collector using a high-voltage generator (model HVG-P60-R-EU, Linari Engineering s.r.l., Italy) to initiate the jet of the polymer solution, which was fed with a controlled rate by the syringe pump (model R-99E, Razel Scientific, USA). A grounded, aluminum-foil-covered screen was used as the collector. The process parameters used in the current study are shown in Table 1. 2.2.3. Film casting The film casting was performed using polymer solutions prepared according to the method described in Section 2.2.1. Each polymer solution (4 ml) was pipetted into a Petri dish and left in a fume hood at room temperature for at least 24 h, allowing the solvents to evaporate and the polymer film to dry. 2.2.4. Scanning electron microscopy (SEM) The electrospun product or polymer film was fixed onto metallic studs with double-sided conductive tape (diameter 12 mm, Oxford instruments, Oxon, UK). The morphology of the sample was observed with a scanning electron microscope (Supra35 VP, Carl Zeiss, Oberkochen, Germany), using an accelerating voltage of 1 kV and a secondary-electron detector. SEM images were used

103

to determine the average diameter of the nanofibers. Between thirty and forty, randomly selected, nanofibers were measured using ImageJ 1.44p software (NIH, USA), and based on the measurements the average fiber diameter and its standard deviation (SD) were determined. 2.2.5. Preparation of buffer solutions A phosphate buffer with a pH of 7.4 was prepared by dissolving 8.00 g of sodium chloride, 0.20 g of potassium chloride, 1.44 g of disodium hydrogen phosphate and 0.24 g of potassium dihydrogen phosphate in distilled water, after which the pH was adjusted to pH 7.4 with 0.1 M sodium hydroxide and the distilled water was added up to 1000 ml. A phosphate buffer of pH 7.6 was prepared by mixing 250 ml of 0.2 M potassium dihydrogen phosphate and 200 ml of 0.2 M NaOH. The pH was adjusted to pH 7.6 with 1 M NaOH, after which distilled water was added up to 1000 ml. 2.2.6. Differential scanning calorimetry (DSC) The crystallinity of the drug in the nanofibers and the polymer films was assessed by DSC. Pure ibuprofen, pure carvedilol, pure PCL, a physical mixture of PCL and ibuprofen (weight ratio 5:2), ibuprofen-loaded nanofibers and films, a physical mixture of PCL and carvedilol (weight ratio 5:2), carvedilol-loaded nanofibers and films were analyzed. The DSC analyses were performed using a Mettler Toledo, STARe System. Accurately weighed samples (3– 5 mg) were sealed in 40 ll, non-hermetically sealed, aluminum pans. The heating rate was 5 K/min and the nitrogen purge rate was 50 ml/min. All the samples were analyzed in the temperature range from 50 to 130 °C. All the DSC curves were normalized to the sample mass. 2.2.7. Determination of the drug content in the nanofibers and polymer films An Agilent 1290 Infinity liquid chromatograph coupled to an Agilent 6460 triple quadrupole mass spectrometer (Agilent Technologies, USA) was used for a determination of the total carvedilol content in the fibers and polymer films. A known amount of sample (about 13 mg) was dissolved in 5 ml of dichloromethane (DCM). The obtained solution was diluted 250 times with methanol and then analyzed. The sample (0.1 ll) was injected onto a 100  3.0 mm, 2.7 lm Poroshell EC-C18 column (Agilent Technologies, USA) at 50 °C using 0.1% formic acid and acetonitrile (50:50, v/v) as the mobile phase at a flow rate of 0.4 ml/min. The retention time of the carvedilol was approximately 1.2 min and the run time was 1.7 min. After each injection, the sampling needle was washed with a washing solvent composed of methanol and water in the ratio 80:20 (v/v). The detection was performed using a JetStreamÒ electrospray source operated in the positive mode. The mass spectrometer parameters were set as follows: drying gas temperature, 275 °C; drying gas flow, 5 l/min; nebulizer, 45 psi; sheath gas temperature, 320 °C; sheath gas flow, 11 l/min; capillary entrance voltage, 4000 V; and nozzle voltage, 1000 V. Both quadrupoles Q1 and Q3 were set at a unit mass resolution. The MRM m/z transitions and collision-energy characteristics for optimal quantification were 407 ? 100, 28 eV (quantifier ion) and 407 ? 222, 16 eV (qualifier ion), respectively. The instrument control, data acquisition and quantification were performed with MassHunter Workstation software. All the analyses were performed at least in triplicate. The ibuprofen content in the nanofibers and polymer films was determined based on the results of the dissolution studies described in Section 2.2.8. The drug concentration in the medium reached a plateau during the time period of the dissolution study, and this maximum concentration was used to calculate the drug content in the sample.

T. Potrcˇ et al. / European Journal of Pharmaceutical Sciences 75 (2015) 101–113

104

The drug entrapment efficiency (EE) was calculated according to the ratio of the actual to the theoretical drug content in the nanofibers (Eq. (1)).

EE ½% ¼

wd  100 wt

ð1Þ

where wd represents the amount of drug, which was determined in 100 mg of prepared nanofibers, and wt is the weight of the drug used for the preparation of 100 mg of nanofibers. 2.2.8. Dissolution studies The PCL nanofibers (80 mg of ibuprofen-loaded nanofibers or 10 mg of carvedilol-loaded nanofibers) were gently rolled on a glass carrier and placed in 250 ml of phosphate buffer, pH 7.4, used as a dissolution medium. The sample was stirred on a magnetic stirrer at room temperature. At the predetermined time points 1 ml of ibuprofen- or 2 ml of carvedilol-containing sample was withdrawn and replaced with fresh phosphate buffer. The sample was then filtered through a 0.45-lm filter (Agilent Technologies, Germany) and analyzed by HPLC. The dissolution studies were performed in triplicate. The ibuprofen and carvedilol were quantified using an HPLC analysis (Agilent 1100 Series, Hewlett Packard, Waldbron, Germany) using the drug-adjusted conditions, as follows. The ibuprofen was analyzed using an ODS Nucleosil C18 column (5 lm, 250 mm  4 mm; Thermo Scientific) at 35 °C. The mobile phase consisted of a mixture of acetonitrile and phosphate buffer (pH 7.6) in the ratio 28.5:71.5 (v/v). A constant flow rate of 1.2 ml/min was used and the drug was monitored using a diode array detector at 225 nm. The carvedilol was determined using a Beta Basic C8 column (3 lm, 150 mm  4.6 mm; Thermo Scientific) at 35 °C. The mobile phase was a mixture of 0.02 M potassium dihydrogen phosphate and acetonitrile in the ratio 6.5:3.5 (v/v) adjusted to pH 2.0 with phosphoric acid R. The flow rate of the mobile phase was 1.0 ml/ min and the drug was monitored using a diode array detector at 241 nm. 3. Results and discussion 3.1. Preparation of drug-loaded PCL nanofiber mats A polymer solution suitable for the preparation of electrospun nanofibers should exhibit good spinnability; therefore, the selection of the right combination of polymer and solvent is crucial. In the present study PCL was chosen for the preparation of drugloaded nanofiber mats, since it exhibits good compatibility with many drugs and can be used in various formulations intended for controlled drug delivery or tissue engineering (Dash and Konkimalla, 2012; Woodruff and Hutmacher, 2010). PCL has previously been reported as a main building block of different PCLbased formulations, e.g., micro and nanospheres (Sinha et al., 2004), films, micelles, scaffolds, and fibers (Dash and Konkimalla, 2012; Woodruff and Hutmacher, 2010). It has also been successfully electrospun in previous studies (Canbolat et al., 2014; Fujihara et al., 2005; Karuppuswamy et al., 2015; Kenawy et al., 2009; Kim et al., 2012; Kouhi et al., 2013; Li et al., 2005; LuongVan et al., 2006; Madhaiyan et al., 2013; Reneker et al., 2002; Seif et al., 2015; Shin et al., 2004; Srikar et al., 2007; Valarezo et al., 2013; Vrbata et al., 2013; Yoshimoto et al., 2003; Zamani et al., 2010). The solvent mixture composed of 75% (w/w) chloroform, being a very good solvent for PCL, and 25% (w/w) acetone, being less favorable for PCL dissolution (Sinha et al., 2004; Woodruff and Hutmacher, 2010), and optimized regarding the processability and morphology of the obtained electrospun product, was selected

for the preparation of the PCL nanofibers in the current study. If the percentage of acetone in the solvent mixture was above 50%, the drying of the polymer solution was very fast, causing needle clogging and hindering the electrospinning process. Increasing the percentage of chloroform in the solvent mixture reduced the volatility and prolonged the drying time of the polymer solution, enabling the continuous production of nanofibers. Furthermore, the mixture of chloroform and acetone was chosen to balance the low dielectric constant of the chloroform (e = 4.75) (Horváth and Szalai, 2014) and the high dielectric constant of the acetone (e = 19.10) (Akerlof, 1932). It is well known that the high electrical conductivity of the solution generally results in thinner, smoother nanofibers with fewer beads (Bhardwaj and Kundu, 2010). Although the obtained PCL solution in the mixture of chloroform and acetone was spinnabile, 0.03% (w/w) sodium iodide was additionally added to further increase the solution’s conductivity and enable the formation of fibers in the nanometer size range (Lalia et al., 2013; Zong et al., 2002). The polymer concentration has a significant influence on the fiber diameter. The starting concentration of the PCL solution used in our research was chosen on the basis of the literature data. It was later optimized experimentally, depending on the solvent mixture being used. The optimal concentration of PCL solution for electrospinning was shown to be 10% (w/w). Higher polymer concentrations caused problems with needle clogging, whereas lower concentrations resulted in the solution dripping. Similar observations were reported by other authors (Frenot and Chronakis, 2003; Zong et al., 2002). The concentration of PCL solution used in this study is thus similar to previously published data, reporting the electrospinning of PCL solutions with concentrations from 7.5% (w/w) in a mixture of chloroform and methanol (1:1, 2:1, 3:1 and 4:1 (v/v) (Kouhi et al., 2013), 3:1 (w/w) (Fujihara et al., 2005) up to 18% (w/v) in pure acetone (Reneker et al., 2002). The starting point for the optimization of the process parameters in our study was the results of the previous studies on the electrospinning of PCL solutions. The optimal distance between the needle and the collector was shown to be 15 cm, the flow rate 1.63–1.98 ml/h, and the voltage necessary to obtain smooth, beadless nanofibers from all the investigated polymer solutions was set to 15.0–17.5 kV (Table 1). Two poorly water-soluble model drugs were incorporated into the PCL nanofibers in the scope of the present study, i.e., ibuprofen (Fig. 1b) and carvedilol (Fig. 1c). The aqueous solubilities of the ibuprofen and carvedilol at room temperature are 0.056 mg/ml (Kocbek et al., 2006) and 0.020 mg/ml (Loftsson et al., 2008), respectively. Both drugs belong to the class II drugs, according to the biopharmaceutical classification system (BCS) (Lindenberg et al., 2004; Loftsson et al., 2008). They are soluble in the solvent mixture used for the preparation of the PCL solution; therefore, they readily dissolved in the PCL solution prior to the electrospinning. 3.2. Morphology of the electrospun products and polymer films The electrospinning of PCL solutions under optimal process parameters (Table 1) enabled the preparation of beadless, randomly oriented, continuous ibuprofen- and carvedilol-loaded nanofibers (Figs. 2 and 3). In this study nanofibers with a very high drug loading were successfully prepared. The theoretical drug loading in the nanofibers reached up to 37.5% (w/w) (Table 2) in the case of nanofibers with 60% of drug based on the dry weight of the polymer. The achieved drug loading was higher than all the available previously published data, where the drug loading did not exceed about 30% (w/w) (based on the dry weight of the polymer) (Vrbata et al., 2013). A high drug loading can significantly affect the electrospun product’s morphology, as shown in Fig. 2 and

T. Potrcˇ et al. / European Journal of Pharmaceutical Sciences 75 (2015) 101–113

105

Fig. 2. SEM images of electrospun PCL nanofibers loaded with 10% (A and B), 15% (C and D), 30% (E and F), 40% (G and H), 50% (I and J) and 60% (K and L) ibuprofen (based on the weight of dry polymer). The nanofibers were visualized at low (A, C, E, G, I and K) and high (B, D, F, H, J and L) magnification.

Fig. 3. SEM images of electrospun PCL nanofibers loaded with 10% (A and B), 15% (C and D), 30% (E and F), 40% (G and H), 50% (I and J) and 60% (K and L) carvedilol (based on the weight of dry polymer). The nanofibers were visualized at low (A, C, E, G, I and K) and high (B, D, F, H, J and L) magnification.

reported also by Karuppuswamy et al. (2015) and Shen et al. (2011). SEM imaging of the prepared electrospun products with 10%, 15%, 30% and 40% of incorporated drug revealed nanofibers with a smooth surface, i.e., without any visible drug crystals (Figs. 2 and 3). This observation indicates that the ibuprofen and carvedilol were fully embedded in the nanofibers. In the case of ibuprofen-loaded electrospun products with 50% and 60% of incorporated drug the morphology is distinctly different (Fig. 2J and L),

whereas the SEM images of the electrospun product with a similar loading of carvedilol do not show any evident changes in the fiber morphology (Fig. 3J and L). The individual fibers loaded with larger amounts of ibuprofen are flattened, fused together and highly perforated, in contrast to the smooth, rounded, individual nanofibers loaded with smaller amounts of the drug. Their average diameters range from 465 ± 88 nm (10% ibuprofen) to 686 ± 196 nm (60% ibuprofen), and in the case of the carvedilol-loaded PCL nanofibers

T. Potrcˇ et al. / European Journal of Pharmaceutical Sciences 75 (2015) 101–113

106

Table 2 Comparison of the average diameters of electrospun ibuprofen and carvedilol loaded PCL nanofibers.

a

Nanofiber sample

Drug loading (%, w/w)

Average diameter (nm)

Nanofiber sample

Drug loading (%, w/w)

Average diameter (nm)

10%a 15%a 30%a 40%a 50%a 60%a

9.1 13.0 23.1 28.6 33.3 37.5

465 ± 88 454 ± 83 593 ± 105 568 ± 97 582 ± 109 686 ± 196

10%a 15%a 30%a 40%a 50%a 60%a

9.1 13.0 23.1 28.6 33.3 37.5

418 ± 76 530 ± 102 539 ± 80 603 ± 113 639 ± 121 756 ± 139

ibuprofen ibuprofen ibuprofen ibuprofen ibuprofen ibuprofen

carvedilol carvedilol carvedilol carvedilol carvedilol carvedilol

Based on the dry weight of polymer.

from 418 ± 76 nm (10% carvedilol) to 756 ± 139 nm (60% carvedilol) (Table 2). The average fiber diameter increased with the amount of incorporated drug, being in line with the previous studies (Karuppuswamy et al., 2015; Shen et al., 2011). To evaluate the advantages of nanofibers, polymer films with the same qualitative and quantitative composition were prepared as a reference material. The polymer films were prepared from the same solutions as the nanofibers by drying the solution at room temperature in a fume hood. The SEM images of the ibuprofenloaded films captured at a lower magnification revealed the presence of some rounded pores in the films with 10% to 40% of drug, whereas a higher magnification also showed the presence of smaller pores in the films with 50% and 60% of ibuprofen (Fig. 4). Some films were composed of differently sized, irregular, pentagonal segments, most probably indicating the crystallization of the polymer, while the others did not show any regular pattern. A significant difference in the morphology is visible in the case of films with 50% and 60% ibuprofen, which is in good correlation with the changes in the morphology of the electrospun nanofibers. The carvedilol-containing polymer films, except the film with 10% carvedilol, exhibited a patterned structure that was clearly visible at higher magnification. These films seem denser than the ibuprofen-loaded films, but without the presence of rounded pores in their structure (Fig. 5). The presence of pores most probably results from the solvent evaporation during the drying. None of the films exhibited visible drug crystals on their surface, indicating the drug’s incorporation into the film. However, the crystallinity of the drugs was further investigated using DSC studies, as described in the following section. 3.3. Evaluation of the drug’s crystallinity The preparation of drug-loaded nanofibers involves drug dissolution and a subsequent solidification, which can cause changes in its crystal structure. The DSC studies were therefore performed to evaluate any changes in the drug crystallinity during the preparation of the nanofibers or polymer films. Pure PCL, pure ibuprofen, pure carvedilol, physical mixtures of the drugs and PCL, nanofibers and films containing different amounts of drug were analyzed and the results are shown in Figs. 6–8. The DSC curves of the pure ibuprofen and carvedilol show endothermic peaks at 79.3 °C and 118.1 °C (Fig. 6A and B), indicating the melting points of the ibuprofen and carvedilol, according to the literature data (Planinšek et al., 2011; Xu et al., 2004). The endothermic peak observed on the DSC curve of the pure PCL at 59.8 °C indicates its melting (Fig. 6C). The determined melting point is in line with the literature data, reporting that the PCL’s melting point is in the range from 59 to 64 °C, (Sinha et al., 2004; Woodruff and Hutmacher, 2010). The endothermic peak at a temperature of around 60 °C is visible on the DSC curves of all the samples containing PCL (Figs. 6–8), indicating its crystalline structure. The DSC curves of the physical mixtures show two endothermic peaks, corresponding to the melting points of both components in the mixture (Fig. 6D and E). In the case of the PCL/ibuprofen physical

Fig. 4. SEM images of PCL films loaded with 10% (A and B), 15% (C and D), 30% (E and F), 40% (G and H), 50% (I and J) 60% (K and L) ibuprofen (based on the weight of dry polymer). The films were visualized at low (A, C, E, G, I and K) and high (B, D, F, H, J and L) magnification.

mixture the peaks are at 59.6 °C and 76.1 °C (Fig 6D), whereas the PCL/carvedilol physical mixture shows peaks at 59.7 °C and 117.3 °C (Fig. 6E). These results revealed that both drugs and the

T. Potrcˇ et al. / European Journal of Pharmaceutical Sciences 75 (2015) 101–113

Fig. 5. SEM images of PCL films loaded with 10% (A and B), 15% (C and D), 30% (E and F), 40% (G and H), 50% (I and J) 60% (K and L) carvedilol (based on the weight of dry polymer). The films were visualized at low (A, C, E, G, I and K) and high (B, D, F, H, J and L) magnification.

polymer were crystalline in the studied PCL/drug physical mixtures. In contrast to the DSC curves of the physical mixtures, DSC curves of prepared nanofibers show only a single endothermic peak in the temperature range from 53.5 to 57.5 °C, indicating melting of the PCL (Fig. 7). The peak is slightly shifted to lower temperatures, compared to the pure PCL, indicating the reduction in the polymer’s crystallinity and the size of its crystals. The shift of the PCL melting peaks in the DSC curves of the carvedilol

107

nanofibers with up to 40% of carvedilol correlates well with the increasing amount of drug in the nanofibers (Fig. 7b). The melting point of PCL in the form of plain PCL nanofibers (Fig. 7) is the same as the melting point of pure PCL, i.e., 59.8 °C (Fig. 6). In the DSC curves of all the drug-loaded nanofibers the endothermic peak, which would indicate melting of the drug, is absent. This finding suggests that the ibuprofen and carvedilol were molecularly dispersed in the polymer matrix, indicating the formation of a solid solution (Bodmeier and Chen, 1989; Kouhi et al., 2013; Seif et al., 2015; Vrbata et al., 2013; Zamani et al., 2010) or the drugs were amorphous (Kouhi et al., 2013; Vrbata et al., 2013). However, the glass transition temperatures were not observed for any sample of drug-loaded nanofibers or films in the scope of our study. The magnitude of the PCL melting enthalpies (which is equal to the area under the peak, or in other words, to the size of the melting peak) revealed an unexpectedly minor decrease, when the amount of carvedilol in the sample was increased, whereas in the case of the samples with ibuprofen the values were increased even above the value that is characteristic for the melting of the plain PCL nanofibers (Fig. 7). This observation suggests that a part of the drug in both types of nanofibers was present in nanocrystalline form and thus the melting or dissolution (in PCL melt) of such a nanosized crystal nuclei overlaps with the melting of the PCL, thereby increasing the total melting enthalpy in this temperature interval. It needs to be pointed out that the DSC measurements were very challenging, because PCL has a lower melting temperature than either of the two drugs, thus it melts first and the drug nanocrystals can dissolve in the PCL melt during the DSC analysis, which could also contribute to the absence of the drug’s melting peaks on the DSC curves, as previously reported also by Bodmeier and Chen (1989). The SEM images did not show any visible crystals on the surface of the nanofibers, but they can still be present in the polymer matrix. In the scope of this research the polymer films were prepared as a reference for the polymer nanofibers. The results of the DSC analysis of the films (Fig. 8) are very similar to the results of the nanofiber samples (Fig. 7). The endothermic peaks representing the melting of the PCL are visible, whereas the melting peaks of the ibuprofen and carvedilol are absent. The assessment of the melting enthalpies of the films also revealed increased values compared to the pure PCL. The comparison between the melting enthalpies of the nanofiber and the film samples of both investigated drugs suggests a decreased crystallinity of the drugs in the nanofibers, with 10–15% of ibuprofen and 10–40% of carvedilol. The changes in the drug’s crystallinity due to the electrospinning of the PCL solution were already reported in previous studies (Kouhi et al., 2013; Seif et al., 2015; Valarezo et al., 2013; Vrbata et al., 2013; Zamani et al., 2010). The group of Vrbata confirmed the changed crystalline structure of naproxen and sumatriptan succinate when incorporated into PCL nanofibers (Vrbata et al., 2013). Based on the DSC measurements they concluded that the produced fibers contained the drug in a non-crystalline form, either in the amorphous state or, which is more likely, homogenously dispersed in the polymer matrix of the fibers. The reduction in the PCL’s crystallinity with an increasing amount of loaded metronidazole benzoate and the absence of a crystalline form of the drug in nanofibers was reported by Zamani et al. (2010). They assumed that the drug was molecularly dispersed in a polymeric matrix. Similar results were also reported by Kouhi et al., who investigated the incorporation of simvastatin in PCL nanofibers (Kouhi et al., 2013). To sum up, the incorporation of drugs in electrospun PCL nanofibers reduced their crystallinity. Based on the morphology of the nanofibers and the films, and the DSC results obtained in the present study, it can be assumed that the drugs incorporated into the

108

T. Potrcˇ et al. / European Journal of Pharmaceutical Sciences 75 (2015) 101–113

Fig. 6. DSC curves of pure ibuprofen (A), pure carvedilol (B), pure PCL (C), PCL/ibuprofen (D) and PCL/carvedilol physical mixture (E).

Fig. 7. DSC curves (a) of pure ibuprofen (A), plain PCL nanofibers (B) and PCL nanofibers with 10% (C), 15% (D), 30% (E), 40% (F), 50% (G) and 60% (H) of ibuprofen and DSC curves and (b) of pure carvedilol (A), plain PCL nanofibers (B) and PCL nanofibers with 10% (C), 15% (D), 30% (E), 40% (F), 50% (G) and 60% (H) of carvedilol.

nanofibers were partially molecularly dispersed in the polymer matrix and partially in the form of dispersed nanocrystals. 3.4. Entrapment efficacy and dissolution studies The drug entrapment efficiency can usually be determined on the basis of results from a dissolution study, which was also the

case in the analysis of the ibuprofen-loaded samples. The determination of the entrapment efficacy was much more challenging in the case of the carvedilol-loaded samples, where the amount of drug released from some samples differed a great deal from the theoretical drug loading. Therefore, a completely new method was developed. Based on the obtained results the entrapment efficiency of carvedilol in the electrospun PCL nanofibers was

T. Potrcˇ et al. / European Journal of Pharmaceutical Sciences 75 (2015) 101–113

109

Fig. 8. DSC curves (a) of pure ibuprofen (A), PCL films (B) and PCL films with 10% (C), 15% (D), 30% (E), 40% (F), 50% (G) and 60% (H) of ibuprofen and DSC curves and (b) of pure carvedilol (A), PCL films (B) and PCL films with 10% (C), 15% (D), 30% (E), 40% (F), 50% (G) and 60% (H) of carvedilol.

between 91% and 100% for the nanofibers with 10%, 15%, 30% and 40% of carvedilol, and about 86% for the samples with 50% and 60% of carvedilol. This high entrapment efficacy in the nanofibers was expected, since the solution of the drug and the PCL is solidified during the electrospinning and the drug is entrapped in the polymer matrix. The risk of drug loss during this process is low. However, when the amount of drug in the polymer solution was high, the entrapment efficiency decreased. It needs to be pointed out that the entrapment efficiency of all the drug-loaded PCL nanofibers was significantly higher than the drug-loaded PCL films (Fig. 9), where the entrapment efficiency was lower than 87%, but still above 74%. The most probable reason for the determined lower drug entrapment efficiency was the drug’s inhomogeneity in the film, resulting from the procedure of the film’s preparation. Some parts of the film were firmly adsorbed to the bottom of the Petri dish and could not be removed from it. Since the drug could have been predominantly deposited on the surface of the Petri dish, it remained there after the film’s removal and the determined drug content was, consequently, much lower than expected. Based on the experimental results it can be concluded that the drug entrapment efficiency was higher than 86% for the nanofibers and generally lower for the drug-loaded films. The entrapment efficiency of ibuprofen in the nanofibers and the films was determined on the basis of the dissolution studies. The comparison of the total amount of the drug released and the theoretical drug loading showed that the entrapment efficiency of the ibuprofen was

100%. The overall results demonstrated that the efficiency of the drug loading achieved by the electrospinning is very high. When developing a dosage form for oromucosal drug delivery, the drug release profile from the formulation is of crucial importance. In our study the drug release from nanofiber mats was determined in vitro and compared to the drug release from polymer films with the same qualitative and quantitative composition. The influences of the drug loading and the molecular weight of the incorporated drug on the drug release profiles were also evaluated. The release profiles revealed that the ibuprofen-loaded nanofiber mats released the drug faster than the ibuprofen-loaded PCL films. This makes them a very promising system for the oromucosal delivery of poorly water-soluble drugs. The release of ibuprofen from all the nanofiber mats was fast (Fig. 10a), reaching about 96% of the total drug release in the first 4 h from the PCL nanofibers. Similar results were already reported previously (Karavasili et al., 2014). The drug release rates from the fibers loaded with different amounts of ibuprofen were not significantly different, indicating that the differences in the fiber diameters and the surface morphology did not influence the release of the ibuprofen. In contrast to the release of the ibuprofen from the nanofibers, the release of the carvedilol from the PCL nanofibers was slower (Fig. 10b). In the first 4 h some 77% of the drug was released from the nanofibers with 30% of carvedilol, whereas only 56–62% of the drug was released from other samples of nanofibers. Furthermore, in 120 h the PCL nanofibers released only 89% of the carvedilol in the case

110

T. Potrcˇ et al. / European Journal of Pharmaceutical Sciences 75 (2015) 101–113

Fig. 9. Carvedilol entrapment efficiency in PCL nanofibers (a) and films (b).

Fig. 10. Release profiles of ibuprofen (a) and carvedilol (b) from electrospun PCL nanofibers. The data points represent the means ± SD of three measurements.

of the nanofibers with 30% of the drug and between 73% and 79% of carvedilol in the case of the nanofibers with 10%, 15% and 40% of the drug. The slowest drug release was observed in the case of the nanofibers with 50% and 60% of carvedilol, where only 49– 56% of the drug was released in 120 h. The observed differences in the release rates of the ibuprofen and carvedilol are not the result of different drug crystallinity or different nanofiber morphologies, but can be attributed to the differences in the drug solubility and/or molecular weight. The ibuprofen has an almost two times smaller molecular weight than the carvedilol and its release is faster compared with the carvedilol. It is known from the literature that the drug can be released from the PCL matrix by a diffusion process (Dash and Konkimalla, 2012; Kenawy et al., 2009), where the molecular weight is a limiting factor. Since the nanofibers are characterized by their very large surface-to-volume ratio and the short diffusion distance for the incorporated drug, the drug release profiles of the electrospun

nanofibers are very different to those of the polymer films. The results showed that 63–83% of the ibuprofen was released in the first 4 h from the PCL films with 10–50% of drug and only about 38% from the PCL film with 60% of ibuprofen (Fig. 11a). The total drug release in the first 4 h from the PCL films was about 13–58% smaller than the ibuprofen release from the drug-loaded PCL nanofibers (Fig. 10a). After 20 h the PCL films with 10–50% of ibuprofen released about 100% and the PCL film with 60% of drug released only about 74% of the theoretical amount of ibuprofen in the film. The comparison between the release profiles of the carvedilolloaded PCL films and the nanofibers showed a similar relationship to that observed for the ibuprofen samples. The carvedilol release was much faster from the PCL nanofibers (Fig. 10b) than from the PCL films (Fig. 11b). The PCL films released only from 10% (PCL film with 60% of carvedilol) to 43% (PCL film with 10% of carvedilol) of the carvedilol in the first 4 h and from 52% (PCL film with 60% of carvedilol) to 96% (PCL film with 10% of carvedilol)

T. Potrcˇ et al. / European Journal of Pharmaceutical Sciences 75 (2015) 101–113

111

Fig. 11. Release profiles of ibuprofen (a) and carvedilol (b) from PCL films. The data points represent the means ± SD of three measurements.

of the incorporated carvedilol in the first 120 h (Fig. 11b). The results are in line with the literature data reporting an initially faster release of the incorporated drug within the first few hours, followed by the second slow release phase from the electrospun mat and the thin film (Kenawy et al., 2009). As was already observed, the drug’s properties, especially its molecular weight and water solubility, could also explain the differences in the obtained release rates. After the dissolution tests were finished it was observed that the amount of released drug from the carvedilol-loaded nanofibers differed from the total drug loading, since the determined drug concentration in the dissolution medium was lower than expected, which should be close to the theoretical. This observation was much more pronounced in the case of the carvedilol-loaded nanofibers than the ibuprofen-loaded nanofibers, where only 52–89% of the drug was detected. Therefore, additional experiments to determine the carvedilol loading were performed.

Because the release profiles reached the plateau in a time period of 120 h, it was concluded that the drug cannot be completely released from the nondegraded PCL matrix, which starts to degrade, according to the literature, after 9 weeks (Dash and Konkimalla, 2012) and can to run for more than one year, depending on its molecular weight (Sinha et al., 2004; Woodruff and Hutmacher, 2010). The amount of drug remaining entrapped in the polymer increased with the drug loading, reaching 27% in the case of the nanofibers with 60% of carvedilol. Similar results were reported by Srikar et al. (2007) and Vrbata et al. (2013). Both groups assumed that only the compound localized on the surface of the fibers and within their pores can be released, whereas the drug encapsulated in the bulk of the polymer matrix cannot be released (Srikar et al., 2007). In our study it was observed that the amount of drug remaining in the nanofibers also depends on its molecular weight. Smaller drug molecules, e.g., ibuprofen, could be released more easily than the higher-molecular-weight drug

Fig. 12. The amount of the drug ((a) ibuprofen and (b) carvedilol) that remained in nanofibers after the dissolution test was finished.

112

T. Potrcˇ et al. / European Journal of Pharmaceutical Sciences 75 (2015) 101–113

molecules, e.g., carvedilol. Besides the molecular weight, the interactions between the drug and the polymer can also influence the amount of unreleased drug. Since the amount of carvedilol remaining unreleased was much higher than the amount of ibuprofen (Fig. 12), stronger hydrophobic–hydrophobic interactions between the polymer and the drug exist (Tarafder et al., 2013). The interactions between the PCL and the drug also influence the drug release rates; therefore, the ibuprofen release was faster than the carvedilol release from the nanofibers. As expected, the incorporation of the poorly soluble drugs in the PCL nanofibers improved their release rates, nevertheless, the release rate also depends on the drug’s properties, most importantly the drug’s molecular weight. 4. Conclusion The results obtained in this study demonstrated that electrospinning can be used for the preparation of highly drug-loaded PCL nanofibers. The incorporation of poorly water-soluble drugs in such a nanomaterial significantly enhances their dissolution rate. However, the release rate from the nanofibers was shown to be drug dependent. The release of ibuprofen, having a smaller molecular weight and a better aqueous solubility, was shown to be faster than the carvedilol. In the first 4 h up to 96% of ibuprofen and only up to 77% of carvedilol was released during the same time period. Furthermore, electrospinning decreased the drug’s crystallinity, and thus the drugs in the nanofibers were partially molecularly dispersed or present in an amorphous state and partially in the form of nanocrystals. To conclude, electrospinning was shown to be very promising nanotechnology-based approach to the formulation of poorly water-soluble drugs in order to enhance their release. Specifically, the drug-loaded PCL nanofiber mats show great potential for further formulation as oromucosal drug delivery systems. Acknowledgements The authors gratefully acknowledge the financial support provided by the Slovenian Research Agency (Programme Code P10189 and Research Projects J1-4236, J1-6746). References Akerlof, G., 1932. Dielectric constants of some organic solvent-water mixtures at various temperatures. J. Am. Chem. Soc. 54, 4125–4139. Bertoncelj, V., Pelipenko, J., Kristl, J., Jeras, M., Cukjati, M., Kocbek, P., 2014. Development and bioevaluation of nanofibers with blood-derived growth factors for dermal wound healing. Eur. J. Pharm. Biopharm. 88, 64–74. Bhardwaj, N., Kundu, S.C., 2010. Electrospinning: a fascinating fiber fabrication technique. Biotechnol. Adv. 28, 325–347. Bodmeier, R., Chen, H., 1989. Preparation and characterization of microspheres containing the anti-inflammatory agents, indomethacin, ibuprofen, and ketoprofen. J. Control. Release 10, 167–175. Canbolat, M.F., Celebioglu, A., Uyar, T., 2014. Drug delivery system based on cyclodextrin-naproxen inclusion complex incorporated in electrospun polycaprolactone nanofibers. Colloids Surf., B 115, 15–21. Dash, T.K., Konkimalla, V.B., 2012. Poly-e-caprolactone based formulations for drug delivery and tissue engineering: a review. J. Control. Release 158, 15–33. Frenot, A., Chronakis, I.S., 2003. Polymer nanofibers assembled by electrospinning. Curr. Opin. Colloid Interface Sci. 8, 64–75. Fujihara, K., Kotaki, M., Ramakrishna, S., 2005. Guided bone regeneration membrane made of polycaprolactone/calcium carbonate composite nano-fibers. Biomaterials 26, 4139–4147. Hearnden, V., Sankar, V., Hull, K., Juras, D.V., Greenberg, M., Kerr, A.R., Lockhart, P.B., Patton, L.L., Porter, S., Thornhill, M.H., 2012. New developments and opportunities in oral mucosal drug delivery for local and systemic disease. Adv. Drug Deliv. Rev. 64, 16–28. Horváth, B., Szalai, I., 2014. Linear and nonlinear dielectric properties of chloroform–bromoform and chloroform–dichloromethane liquid mixtures. J. Mol. Liq. 189, 81–84. Huang, Z.-M., Zhang, Y.Z., Kotaki, M., Ramakrishna, S., 2003. A review on polymer nanofibers by electrospinning and their applications in nanocomposites. Compos. Sci. Technol. 63, 2223–2253.

Illangakoon, U.E., Gill, H., Shearman, G.C., Parhizkar, M., Mahalingam, S., Chatterton, N.P., Williams, G.R., 2014. Fast dissolving paracetamol/caffeine nanofibers prepared by electrospinning. Int. J. Pharm. 477, 369–379. Karavasili, C., Bouropoulos, N., Kontopoulou, I., Smith, A., van der Merwe, S.M., Rehman, I.U.R., Ahmad, Z., Fatouros, D.G., 2014. Preparation and characterization of multiactive electrospun fibers: poly-caprolactone fibers loaded with hydroxyapatite and selected NSAIDs. J. Biomed. Mater. Res., Part A 2014 (102A), 2583–2589. Karuppuswamy, P., Reddy Venugopal, J., Navaneethan, B., Luwang Laiva, A., Ramakrishna, S., 2015. Polycaprolactone nanofibers for the controlled release of tetracycline hydrochloride. Mater. Lett. 141, 180–186. Kawabata, Y., Wada, K., Nakatani, M., Yamada, S., Onoue, S., 2011. Formulation design for poorly water-soluble drugs based on biopharmaceutics classification system: basic approaches and practical applications. Int. J. Pharm. 420, 1–10. Kenawy, E.-R., Abdel-Hay, F.I., El-Newehy, M.H., Wnek, G.E., 2009. Processing of polymer nanofibers through electrospinning as drug delivery systems. Mater. Chem. Phys. 113, 296–302. Kim, Y.-J., Park, M.R., Kim, M.S., Kwon, O.H., 2012. Polyphenol-loaded polycaprolactone nanofibers for effective growth inhibition of human cancer cells. Mater. Chem. Phys. 133, 674–680. Kocbek, P., Baumgartner, S., Kristl, J., 2006. Preparation and evaluation of nanosuspensions for enhancing the dissolution of poorly soluble drugs. Int. J. Pharm. 312, 179–186. Kouhi, M., Morshed, M., Varshosaz, J., Fathi, M.H., 2013. Poly (e-caprolactone) incorporated bioactive glass nanoparticles and simvastatin nanocomposite nanofibers: preparation, characterization and in vitro drug release for bone regeneration applications. Chem. Eng. J. 228, 1057–1065. Lalia, B.S., Kochkodan, V., Hashaikeh, R., Hilal, N., 2013. A review on membrane fabrication: structure, properties and performance relationship. Desalination 326, 77–95. Lam, J.K.W., Xu, Y., Worsley, A., Wong, I.C.K., 2014. Oral transmucosal drug delivery for pediatric use. Adv. Drug Deliv. Rev. 73, 50–62. Li, W.-J., Tuli, R., Huang, X., Laquerriere, P., Tuan, R.S., 2005. Multilineage differentiation of human mesenchymal stem cells in a three-dimensional nanofibrous scaffold. Biomaterials 26, 5158–5166. Lindenberg, M., Kopp, S., Dressman, J.B., 2004. Classification of orally administered drugs on the World Health Organization model list of essential medicines according to the biopharmaceutics classification system. Eur. J. Pharm. Biopharm. 58, 265–278. Loftsson, T., Vogensen, S.B., Desbos, C., Jansook, P., 2008. Carvedilol: solubilization and cyclodextrin complexation: a technical note. AAPS PharmSciTech 9, 425– 430. Luong-Van, E., Grøndahl, L., Chua, K.N., Leong, K.W., Nurcombe, V., Cool, S.M., 2006. Controlled release of heparin from poly(e-caprolactone) electrospun fibers. Biomaterials 27, 2042–2050. Madhaiyan, K., Sridhar, R., Sundarrajan, S., Venugopal, J.R., Ramakrishna, S., 2013. Vitamin B12 loaded polycaprolactone nanofibers: a novel transdermal route for the water soluble energy supplement delivery. Int. J. Pharm. 444, 70–76. Meinel, A.J., Germershaus, O., Luhmann, T., Merkle, H.P., Meinel, L., 2012. Electrospun matrices for localized drug delivery: current technologies and selected biomedical applications. Eur. J. Pharm. Biopharm. 81, 1–13. Patel, V.F., Liu, F., Brown, M.B., 2011. Advances in oral transmucosal drug delivery. J. Control. Release 153, 106–116. Pelipenko, J., Kocbek, P., Govedarica, B., Rošic, R., Baumgartner, S., Kristl, J., 2013a. The topography of electrospun nanofibers and its impact on the growth and mobility of keratinocytes. Eur. J. Pharm. Biopharm. 84, 401–411. Pelipenko, J., Kristl, J., Jankovic´, B., Baumgartner, S., Kocbek, P., 2013b. The impact of relative humidity during electrospinning on the morphology and mechanical properties of nanofibers. Int. J. Pharm. 456, 125–134. Planinšek, O., Kovacˇicˇ, B., Vrecˇer, F., 2011. Carvedilol dissolution improvement by preparation of solid dispersions with porous silica. Int. J. Pharm. 406, 41–48. Pouton, C.W., 2006. Formulation of poorly water-soluble drugs for oral administration: physicochemical and physiological issues and the lipid formulation classification system. Eur. J. Pharm. Sci. 29, 278–287. Reneker, D.H., Kataphinan, W., Theron, A., Zussman, E., Yarin, A.L., 2002. Nanofiber garlands of polycaprolactone by electrospinning. Polymer 43, 6785–6794. Rošic, R., Pelipenko, J., Kristl, J., Kocbek, P., Baumgartner, S., 2012. Properties, engineering and applications of polymeric nanofibers: current research and future advances. Chem. Biochem. Eng. Q. 26, 417–425. Rošic, R., Kocbek, P., Pelipenko, J., Kristl, J., Baumgartner, S., 2013. Nanofibers and their biomedical use. Acta Pharm. 63, 295–304. Sattar, M., Sayed, O.M., Lane, M.E., 2014. Oral transmucosal drug delivery – current status and future prospects. Int. J. Pharm. 471, 498–506. Seif, S., Franzen, L., Windbergs, M., 2015. Overcoming drug crystallization in electrospun fibers – elucidating key parameters and developing strategies for drug delivery. Int. J. Pharm. 478, 390–397. Shakya, P., Madhav, N.V.S., Shakya, A.K., Singh, K., 2011. Palatal mucosa as a route for systemic drug delivery: a review. J. Control. Release 151, 2–9. Shen, X., Yu, D., Zhu, L., Branford-White, C., White, K., Chatterton, N.P., 2011. Electrospun diclofenac sodium loaded EudragitÒ L 100-55 nanofibers for colontargeted drug delivery. Int. J. Pharm. 408, 200–207. Shin, M., Ishii, O., Sueda, T., Vacanti, J.P., 2004. Contractile cardiac grafts using a novel nanofibrous mesh. Biomaterials 25, 3717–3723. Sill, T.J., von Recum, H.A., 2008. Electrospinning: applications in drug delivery and tissue engineering. Biomaterials 29, 1989–2006.

T. Potrcˇ et al. / European Journal of Pharmaceutical Sciences 75 (2015) 101–113 Sinha, V.R., Bansal, K., Kaushik, R., Kumria, R., Trehan, A., 2004. Poly-e-caprolactone microspheres and nanospheres: an overview. Int. J. Pharm. 278, 1–23. Srikar, R., Yarin, A.L., Megaridis, C.M., Bazilevsky, A.V., Kelley, E., 2007. Desorptionlimited mechanism of release from polymer nanofibers. Langmuir 24, 965–974. Tarafder, S., Nansen, K., Bose, S., 2013. Lovastatin release from polycaprolactone coated b-tricalcium phosphate: effects of pH, concentration and drug–polymer interactions. Mater. Sci. Eng. C 33, 3121–3128. Valarezo, E., Tammaro, L., González, S., Malagón, O., Vittoria, V., 2013. Fabrication and sustained release properties of poly(e-caprolactone) electrospun fibers loaded with layered double hydroxide nanoparticles intercalated with amoxicillin. Appl. Clay Sci. 72, 104–109. ˇ izˇinská, L., 2013. Vrbata, P., Berka, P., Stránská, D., Dolezˇal, P., Musilová, M., C Electrospun drug loaded membranes for sublingual administration of sumatriptan and naproxen. Int. J. Pharm. 457, 168–176. Woodruff, M.A., Hutmacher, D.W., 2010. The return of a forgotten polymer— polycaprolactone in the 21st century. Prog. Polym. Sci. 35, 1217–1256.

113

Xu, F., Sun, L.-X., Tan, Z.-C., Liang, J.-G., Li, R.-L., 2004. Thermodynamic study of ibuprofen by adiabatic calorimetry and thermal analysis. Thermochim. Acta 412, 33–57. Yoshimoto, H., Shin, Y.M., Terai, H., Vacanti, J.P., 2003. A biodegradable nanofiber scaffold by electrospinning and its potential for bone tissue engineering. Biomaterials 24, 2077–2082. Yu, D.-G., Yang, J.-M., Branford-White, C., Lu, P., Zhang, L., Zhu, L.-M., 2010. Third generation solid dispersions of ferulic acid in electrospun composite nanofibers. Int. J. Pharm. 400, 158–164. Zamani, M., Morshed, M., Varshosaz, J., Jannesari, M., 2010. Controlled release of metronidazole benzoate from poly e-caprolactone electrospun nanofibers for periodontal diseases. Eur. J. Pharm. Biopharm. 75, 179–185. Zong, X., Kim, K., Fang, D., Ran, S., Hsiao, B.S., Chu, B., 2002. Structure and process relationship of electrospun bioabsorbable nanofiber membranes. Polymer 43, 4403–4412.