Preparation, characterization and photostability assessment of curcumin microencapsulated within methacrylic copolymers

Journal of Drug Delivery Science and Technology 33 (2016) 88e97

Contents lists available at ScienceDirect

Journal of Drug Delivery Science and Technology journal homepage: www.elsevier.com/locate/jddst

Research paper

Preparation, characterization and photostability assessment of curcumin microencapsulated within methacrylic copolymers Tiziana M.G. Pecora a, Simona Cianciolo b, Alfio Catalfo c, Guido De Guidi c, Barbara Ruozi d, Maria Chiara Cristiano a, Donatella Paolino e, Adriana C.E. Graziano f, Massimo Fresta a, Rosario Pignatello b, * a

Department of Health Sciences, University “Magna Græcia”, Catanzaro, Italy Department of Drug Sciences, University of Catania, Catania, Italy Department of Chemical Sciences, University of Catania, Catania, Italy d Department of Pharmaceutical Sciences, University of Modena and Reggio Emilia, Modena, Italy e Department of Experimental and Clinical Medicine, University “Magna Græcia”, Catanzaro, Italy f Department of Biomedical and Biotechnological Sciences, University of Catania, Catania, Italy b c

a r t i c l e i n f o

a b s t r a c t

Article history: Received 13 January 2016 Received in revised form 27 March 2016 Accepted 28 March 2016 Available online 30 March 2016

Microencapsulated curcumin (CUR) was obtained by coevaporation with polymethacrylate polymers (blends at various percent ratios of Eudragit® RS100 and RL100 resins). The suspensions were freezedried to produce free flowing microparticles, which were sieved in the 420e90 mm range. They were characterized in the solid state for micromeritic properties and drug loading, and by FT-IR, powder X-ray diffractometry and differential scanning calorimetry for physical state. Encapsulation efficiency largely varied from 35 to 95%, mainly depending on the copolymer composition and to a less extent from drugto-polymer ratio. Solid-state characterization confirmed the chemical stability of CUR in microparticles, and suggested that the drug was in a microcrystalline form within the polymer matrix; microscopy analysis confirmed the latter statement. In vitro release and dissolution profile of neat and encapsulated CUR were assessed in simulated gastric and intestinal fluids: from these studies, it was however found that the microencapsulation of CUR in these polymers did not improve the solubility of this very poorly soluble compound in simulated gastric and intestinal aqueous media. Interestingly, photostability experiments showed that the dispersion of CUR in the polymer matrix effectively protects the drug from light-induced chemical degradation, with an effect dependent on the drug-to-polymer ratio. © 2016 Elsevier B.V. All rights reserved.

Keywords: Eudragit® Retard polymers Photodegradation Differential scanning calorimetry FT-IR spectrophotometry Powder X-ray diffractometry

1. Introduction Curcumin (diferuloymethane; CUR) is the most active component of Curcuma longa (or turmeric), belonging to the family Zingiberaceae. Apart its use as a food additive and pigment, centuries of traditional use in Chinese and Indian medicine has suggested a

Abbreviations: CUR, curcumin; DSC, differential scanning calorimetry; ERL, Eudragit® RL100; ERS, Eudragit® RS100; FT-IR, Fourier-transform infrared spectrophotometry; PDI, Polydispersity Index; PXRD, powder X-ray diffractometry; SGF, simulated gastric fluid; SIF, simulated intestinal fluid; UV, visible-ultraviolet spectrophotometry. * Corresponding author. Department of Drug Sciences, University of Catania, V.le A. Doria, 6, I-95125, Catania, Italy. E-mail address: [email protected] (R. Pignatello). http://dx.doi.org/10.1016/j.jddst.2016.03.013 1773-2247/© 2016 Elsevier B.V. All rights reserved.

plethora of potential therapeutic applications. In the last decades, many scientific studies have validated various pharmacological effects, such as antioxidant, anti-inflammatory, antibacterial, and, more recently, anticancer activity [1]. This opened the way to a wide therapeutic potential for this natural compound, and CUR has been recommended to various extent in arthritis, diabetes, cardiovascular diseases, liver fibrosis, gall stone formation, neurological diseases, tumors, and inflammatory bowel disease [2,3]. Such a plethora of apparently different clinical conditions is related to the range of molecular targets of CUR, that include transcription factors, inflammatory cytokines, enzymes and the epigenetic modulation which modulate histone deacetylases, histone acetyltransferases, DNA methyltransferase I and miRNAs [4,5]. Unfortunately, oral bioavailability of CUR is very limited (about 60%) [6], mainly because of its very low solubility in aqueous media

T.M.G. Pecora et al. / Journal of Drug Delivery Science and Technology 33 (2016) 88e97

and incomplete absorption in the gastro-intestinal tract [7]. To overcome such drawbacks, different technological approaches have been explored, including encapsulation of CUR in colloidal polymeric and lipid-based delivery systems, such as liposomes, nanoparticles and microspheres [8e12]. Many of these strategies have shown to improve the solubility and bioavailability of CUR, although its biological activity has not always been positively affected, thus raising an intriguing hypothesis about the effects of encapsulation inside delivery systems on the metabolism and pharmacodynamics of CUR and its active metabolites [13]. However, dispersion of CUR in polymeric or lipid micro- or nanomatrices remains a potentially valid strategy to improve the solubility and absorption of this compound after oral administration [8,14,15]. We are conducting a wide research project aimed at producing microcarriers, made using Eudragit® Retard polymers, for the oral delivery of naturally occurring active compounds [16]. Eudragit® RS100 (ERS) and RL100 (ERL) resins are copolymers of poly(ethylacrylate, methyl-methacrylate and chlorotrimethylammonioethyl methacrylate), containing an amount of quaternary ammonium groups between 4.5e6.8% and 8.8e12% for ERS and ERL, respectively. Both are insoluble at physiological pH values and capable of swelling [17], thus being valid candidates for the dispersions of bioactive compounds [18e20]. These polymers are usually employed for the coating of solid oral dosage forms; however, recent studies supported the use of these polymers for producing controlled release micro- and nanosystems of pharmaceutical interest [21e27]. In this paper, the preparation and characterization of CURloaded ERL/ERS microparticles are reported. Microparticles were produced by a solvent displacement technique, starting from a cosolution of CUR and polymer(s) in organic solvents. The micromeritics and drug loading of microparticles were analyzed, as well as the chemical interactions of CUR with Eudragit® polymers in the solid-state. The dissolution profile and in vitro release pattern of neat and encapsulated CUR were assessed in simulated gastric fluid and in phosphate buffer (pH 6.8). Furthermore, the photostability of CUR loaded in the microparticles was assessed in different media both under UV and visible irradiation. It is in fact known that several degradation products are formed upon light exposure of this compound, and CUR itself was found to act as photosensitizer via a singlet oxygen mechanism [28,29]. 2. Materials and methods 2.1. Chemicals CUR (Curcuma longa rhizome dry extract; total curcuminoids 95% min.) was produced by Vivatis Pharma GmbH (Hamburg, Germany) and was kindly gifted by Labomar srl (Istrana, Italy); ERL and ERS resins (Evonik Rohm GmbH) were kindly provided by Rofarma Italia srl (Gaggiano, Italy). HPLC solvents were from VWR and Labscan (Milan, Italy); the other reagents and solvents were purchased from Sigma-Aldrich Chimica srl (Milan, Italy). All of them were used as received. 2.2. Preparation of blank microparticles To optimize the composition and properties of Eudragit microparticles, unloaded systems were initially produced using different variables. Concentrated solutions with the copolymer composition reported in Table 1 were prepared by dissolving 4 g total of each Eudragit® resin blend in 40 ml of ethanol, by overnight mechanical stirring at room temperature. An exact volume of these solutions

89

Table 1 Polymer composition of blank microparticles. Code

ERL (%)

ERS (%)

A B C D E

100 70 50 30 10

e 30 50 70 90

(i.e., 5, 10 or 20 ml, corresponding to a final copolymer concentration of 1, 2 or 4%, w/v, respectively) (Table 2) was slowly dropped within 60 min into 50 ml distilled water containing 0.02% (w/v) Tween® 80, under mechanical stirring at 150 rpm and room temperature, to produce the microparticles. The mixture was then left under stirring for about 24 h to allow the evaporation of the solvent, after which it was frozen and lyophilized for 24 h (Edward Modulyo). 2.2.1. Sieving and size distribution The produced microparticle batches were passed through European Pharmacopoeia standard metallic sieves (841, 420, 210, 125, and 90 mm; corresponding to 20, 40, 70, 120, and 170 mesh, respectively) using a vibratory apparatus (Giuliani Tecnologie srl, Turin, Italy). Each portion was collected and weighed, to calculate the size distribution data gathered in Table 2, where d10, d50 and d90 are the mean particle size (in mm) determined at the 10th, 50th and 90th percentiles of undersized microparticles, respectively. 2.3. Preparation of CUR-loaded microparticles Based on the behaviour of blank microparticles, selected compositions were individuated to be loaded with CUR (Table 3). The amount of drug needed to produce each chosen drug-to-polymer weight ratio (DPR) (namely 1:1, 1:5 and 1:10) was dissolved in 10 ml acetone and added to the volume of polymer ethanol solution required for each batch, as above described. The preparation, collection and sieving of CUR-loaded microparticles was carried out as described for the blank systems. 2.4. Preparation of the physical mixtures For the sake of comparison of physico-chemical properties, physical mixtures having the same composition of microparticles were produced by simple mixing CUR and the required polymers in a porcelain mortar for 10 min. To attain an average diameter

Table 2 Cumulative size distribution of sieved blank microparticles (see Table 1 for the composition of copolymer blends). Code

Polymer concentration (%, w/v)

Yielda

Cumulative size distribution (mm) d10

d50

d90

A1 B1 B2 B4 C1 C4 D4 E1 E2 E4

1 1 2 4 1 4 4 1 2 4

72.5 90.2 81.5 74.8 95.6 60.0 79.8 36.0 32.8 73.8

<90 100 115 <90 95 <90 110 <90 <90 90

330 115 335 120 235 100 195 <90 <90 160

675 285 705 525 565 145 440 120 320 >841

a Percent of recovered microparticles in the pooled 841e90 mm sieve fractions, compared to the initial amount of polymers.

90

T.M.G. Pecora et al. / Journal of Drug Delivery Science and Technology 33 (2016) 88e97

Table 3 Composition and properties of CUR-loaded Eudragit® microparticles. Code Percent copolymer composition (ERL/ ERS)

Final polymer concentration (%, w/ DPR Percent encapsulation v) efficiencya

Theoretical drug contentb

Actual drug contentb

B41 B12 B13 B42 B43 D12 D13 D42 D43

4 1 1 4 4 1 1 4 4

e 20 10 20 10 20 10 20 10

e 13.14 5.23 7.00 3.46 19.34 6.96 14.08 8.34

a b

70:30 70:30 70:30 70:30 70:30 30:70 30:70 30:70 30:70

1:1 1:5 1:10 1:5 1:10 1:5 1:10 1:5 1:10

e 65.70 52.30 35.01 34.60 96.70 69.60 70.42 83.40

Percent of theoretical CUR found in the pooled 841e90 mm microparticle sieve fractions. mg CUR/100 mg microparticles; the actual drug content values refer to the 841e90 mm sieve fractions.

comparable to that of the microparticles, the polymers were powdered in advance by triturating in a mortar the commercial granules and selecting the 420e125 mm sieve fraction.

values are also reported in Table 3, as mg of CUR per 100 mg of each microparticle batch. 2.7. Solubility and in vitro dissolution tests

2.5. Microparticle characterization FT-IR spectra were recorded in potassium bromide on a PerkinElmer 1600 spectrophotometer, in the range 4000e500 cm
1. DSC experiments were performed using a DSC 12E calorimeter (MettlerToledo SpA, Novate M., Italy), connected to a Haake D8-G thermocryostat. The temperature and enthalpy changes were calibrated using a pure indium sample. The detection system consisted of a Mettler Pt100 sensor, with a thermometric sensitivity of 56 nV/ C, a calorimetry sensitivity of about 3 nV/mW and a background noise of 60 nV (<1 mV). Each scan had an accuracy of ±0.4  C, and a reproducibility and a resolution of 0.1  C. To determine the thermotropic behaviour, each compound (8e10 mg) was sealed in a 100 ml aluminium pan. An empty pan was used as reference. Samples were submitted to a heating cycle between 40 and 240  C, at a scan rate of 2  C/min. PXRD data were collected with a Philips X'Pert Pro X-ray diffraction system (Eindhoven, The Netherlands) (available at the Centro Interdipartimentale Grandi Strumenti di Modena e Reggio Emilia d CIGS) equipped with a PANanalytical solid state detector, operating in reflection mode, with a CuKa radiation (without monochromator: l is a “mixing” between lKa1 ¼ 1.540598 and lKa2 ¼ 1.544426; Kb radiation was removed by Ni foil). X-ray data were collected over a range 10 < 2q < 30 at room temperature; the scan-rate was set at 0.007 /s. Microscopy analysis was performed on a Leica DM IRB brightfield microscope equipped with a computer-assisted Leica DFC 550 camera (Leica Microsystems Srl, Milan, Italy). The lyophilized microparticle samples (420e90 mm sieve fraction) were resuspended in water containing 0.5% (w/v) Tween 80 immediately before the analysis and placed in 24-well plastic plate. 2.6. Determination of encapsulation efficiency The 841e125 mm and 420e90 mm sieve fractions (approximately 25 mg each, exactly weighed) were separately dispersed in 10 ml of UV-grade methanol and stirred for 2 h at room temperature, to promote the complete dissolution of the drug. After filtration through a 0.45 mm pore size hydrophilic polypropylene (GHP) membrane filter (Acrodisc PSF, (Pall Italia srl, Buccinasco, Italy), the solutions were analyzed by an UVevisible spectrophotometer (UV1601, Shimadzu Italia srl, Milan, Italy) at 420 nm, with reference to a fresh calibration curve of the drug in methanol (linear in the range 0.5e20 mg/ml; r2 ¼ 0.9993). The encapsulation efficiency (Table 3) was expressed as the percentage of theoretical CUR analytically found in the two microparticle sieve fractions. The drug content

The solubility of pure CUR or associated to the Eudragit® resins as either microparticles or physical mixtures was assessed in simulated gastric fluid without pepsin (SGF, pH 1.2, made up by 7 ml of 0.02% HCl, 2 g NaCl and 0.2 g Tween® 40 in 1 L of HPLC-grade water) or simulated intestinal fluid without enzymes (SIF; Int. Pharm. 3rd Ed.) (0.0753 M phosphate buffer solution, pH 6.8, containing 0.02% w/v Tween® 40) [30]. An excess amount of each sample (15e20 mg) was added to screw-capped glass vials containing 20 ml of each medium. The vials were shaken in a thermostatically water bath maintained at 37 ± 0.1  C for either 6, 24 or 48 h and the supernatants were filtered through a 1.0 mm pore size glass fiber filter (Pall Life Sciences; VWR International PBI, Milan, Italy). The filtrates were suitably diluted with water and assayed spectrophotometrically for the concentration of CUR. Experimental results indicated that a 24-h period was sufficient to reach the equilibrium solubility of the drug in both media, since no increase in CUR concentration was registered after a 48-h incubation (data not shown). A calibration curve of the drug in SGF/methanol (1:1, v/ v) was used for the experiments in SGF (linear in the range 0.2e10 mg/ml; r2 ¼ 0.9987; lmax: 423.5 nm); for tests in SIF, the calibration curve of CUR in methanol at 420 nm was used (see above), since CUR is in fact practically insoluble in the medium. All tests were made at least in triplicate. For the drug dissolution assay, duplicate samples of approximately 2 mg of CUR or 15 mg of lyophilized CUR-loaded microparticles or physical mixtures were placed in 20 ml of either SGF for 2 h or SIF for 4 h, at 37  C under slow mechanical stirring (150 rpm). At predetermined times, 1-ml aliquots of each medium was withdrawn and immediately replaced with the same volume of the respective pre-warmed buffer solution to keep the volume constant. Each sample was filtered through a 1.0 mm glass fiber membrane filter (Pall Life Sciences) and analyzed by UV spectrophotometry for CUR concentration (see above). 2.8. Ageing studies Selected CUR-loaded lyophilized microparticles were stored in screw-capped amber glass bottles at room temperature or in a static oven at 40 ± 1  C. After 3 and 6 months, weighed aliquots of the aged samples were dissolved in methanol and the amount of CUR was determined by UV analysis, as described above. Each analysis was made in duplicate. Methanol was preferred for these experiments since it dissolved more quickly the polymer matrices, besides than CUR.

T.M.G. Pecora et al. / Journal of Drug Delivery Science and Technology 33 (2016) 88e97

2.9. Photostability studies The protective effect of the polymers on the photo-degradation of CUR was assayed by exposition of samples of neat and microencapsulated drug to UV light. Test specimens were freshly made by diluting the microparticles with HPLC-grade water and neat CUR with water/methanol (80:20, v/v). Based on the drug content value (Table 3), each microparticle specimen was weighed so that to contain approximately 3.5 mg/ml CUR. The solubility at equilibrium of neat CUR in different solvents was verified by stirring each sample solution for about 1 h at room temperature and determining the molar absorption coefficient, using an UVeVis diode array spectrophotometer HP 8452. Samples were stored in the dark at room temperature until irradiation experiments. Samples were irradiated at room temperature in a time range of 10e30 min with the Xenon lamp of a Fluorolog® spectrofluorometer (Horiba Jobin Yvon, Milan, Italy), with a fluence rate of about 1000 mW/cm2 emitting in a 15 nm pass band with a peak centred at 422 nm. The incident photon flux on the irradiated solution in the cuvette (3 ml) was in the range of 1  1016 quanta s
1. Immediately after irradiation, samples were frozen by liquid nitrogen and lyophilized for the further HPLC analysis. The HPLC analysis was performed according to literature on a Hewlett-Packard 1100 chromatograph equipped with on-line diode array detector (DAD) and a Kontron SFM fluorescence detector (FLD) [28]. The freeze-dried samples were redissolved in 1 ml HPLC-grade Methanol and were injected into a 20 ml loop combined with an analytical C18 Chrompack Inertsil ODS-2 column (Varian, 4.6  250 mm, 5 mm) shielded by a pre-column (4.6  4.5 mm). Chromatograms were obtained using DAD (absorbance centred at 420 nm) and FLD (lex 422, lem 557). An isocratic flow rate of 1 ml/ min of methanol/water (96:4, v/v) was used. A freshly prepared CUR standard in methanol (0.1 mg/ml) was used to obtain a calibration curve (linear in the range 1e8 mg/ml; r2 ¼ 0.988).

91

within the lower size classes was measured (Table 2). 3.2. CUR-loaded microparticles Based on the above size analysis, some polymer compositions and operation variables were selected to produce the drug-loaded systems. In particular, the polymer compositions B and D were chosen (cf. Table 1), using a 1 or 4% (w/v) copolymer concentration and three DPR values, namely 1:1, 1:5 and 1:10. To produce the drug solution, acetone was preferred to ethanol, because of a higher solubility of CUR in this solvent. Table 3 resumes the properties of the produced samples. After the first attempts, the 1:1 DPR value (batch B41) was casted-off from further studies, since the amount of drug, and the subsequent volume of acetone to dissolve it, appeared to be excessive compared to the required volumes of ethanol polymer solutions. 3.3. Drug-polymer interaction studies The physico-chemical and technological evaluation of the lyophilized microparticles was carried out by conventional techniques, using as references the starting polymer(s), pure CUR, blank microparticles and the corresponding PhM. FT-IR was used to assess any interaction occurring in the solid coevaporates between the polymeric matrix and the active compound. DSC and XRPD experiments were made to assess the degree of crystallinity or amorphization of CUR in the various polymer matrices. The FT-IR spectrum of pure CUR (Fig. 1a) showed characteristic

3. Results 3.1. Preliminary studies of blank microparticles Eudragit® microparticles were produced by a solvent displacement technique, starting from ethanol solutions of pure ERL or ERL/ ERS blends (Table 1). The two polymer matrices are known to mainly differ for their permeability to water and gastro-intestinal juices [31]; in particular, a progressive increase of ERS percentage is expected to reduce the permeability of microparticles and diffusion of encapsulated drugs [32]. A second formulation variable was the amount of polymer solution added to the aqueous phase (0.02%, w/v Tween® 80 in distilled water), leading to a final polymer concentration ranging from 1 to 4% (w/v). Table 2 gathers the cumulative size distribution of blank microparticles. As a general trend, using ERL only (A1 batch) larger microparticles were obtained, with a relatively high percentage of aggregates (>841 mm). Also the ERL/ERS 1:1 blend (C batches) produced quite large microparticles, especially when a lower polymer concentration was used (C1 batch). The formulation containing prevalently ERS (batches E) produced very small particles and/or powders (<90 mm sieve fraction), giving a low recovery yield in the useful size window. For such reasons, these compositions were not further considered for drug loading. Among the other polymer blends (B and D), the particle distribution seemed to be strictly related to the variable expressing the initial polymers concentration. In particular, a 2% w/v copolymer concentration (B2 batch) was associated with the outcome of larger microparticles, while using the highest polymer blend concentration (4%, batches B4 and D4) a more homogeneous distribution

Fig. 1. FT-IR spectrum of a) pure CUR, b) ERS, c) B12 microparticles, and d) CUR-ERS 5:1 physical mixture.

92

T.M.G. Pecora et al. / Journal of Drug Delivery Science and Technology 33 (2016) 88e97

signals at 3510 cm
1 (stretching of phenolic OH groups), 1684 cm
1 (carbonyl C]O stretching), 1627.6 cm
1 (enol C]O stretching), 1508 cm
1 (aromatic CeC stretching), 1458 and 1430 cm
1 (alkane CeH groups scissoring and bending), 1282 cm
1 (alcohol CO stretching), two series of peaks at 1233e1153 cm
1 and 1027e963 cm
1 due to the stretching of alcohol, ether, carboxylic and ester CeO groups, and peaks at 856e814 cm
1 (out-of-plane bending of meta- and para-substitutions) [33]. The two polymers gave an identical FT-IR spectrum (Fig. 1b shows the one of ERS) characterized by the C]O ester vibration at 1731 cm
1 and bands of ester groups at 1150e1195 and 1240e1270 cm
1. CHx vibrations can be detected at 1385, 1450, 1485 and 2950e3000 cm
1 [31]. In the corresponding FT-IR spectra of CUR-ERS physical mixture (Fig. 1d) (using ERL identical results were however obtained), both the sets of signals of the starting components are visible without shifts of resonance frequencies, showing that CUR maintained its chemical structure when physically dispersed within the polymeric network. Conversely, in the IR spectra of microparticles (Fig. 1c reports as an example that one of batch B12), the peaks of the polymer structure are more visible, due to its higher weight ratio in the coevaporate; certain peaks proper of neat CUR are not more visible, especially the ones around 1685 and 1282 cm
1, suggesting the occurrence of some electrostatic interactions between the drug and the copolymer. The DSC curves of CUR, ERS and CUR-loaded microparticles are shown in Fig. 2. Pure CUR showed a sharp endothermic peak corresponding to melting at 178.5  C; both ERS (Fig. 2) and ERL resins (not shown) showed a small broad peak at 75  C, related to the relaxation peak that follows the glass transition. No net melting point was observed, due to the amorphous nature of these copolymers. A thermal decomposition of the polymers occurred only at higher temperatures (around 360  C, not shown). Upon microencapsulation, the peak corresponding to the melting point of neat CUR disappeared in the DSC scans of both

Fig. 2. DSC thermograms of neat CUR, ERS, some microparticle batches and CUR-ERS 1:5 and 1:10 physical mixtures (PM2 and PM3, respectively). See Table 3 for the exact microparticle composition.

coevaporates and 1:10 physical mixture (PM3). Only the physical mixture obtained at a DPR of 1:5 (PM2) showed a residual sign of CUR melting peak. To confirm the above hypothesis, solid X-ray analysis was performed on the same specimens (Fig. 3). Neat CUR showed peaks at distinct values of  2q, at 8.93, 12.21, 14.55, 17.29, 18.19, 23.35, 24.60, and 25.61. A series of smaller peaks at 7.93, 14.99, 15.16, 15.70, 15.89, 16.34, 18.60, 18.85, 19.47, 21.12, 21.27, 22.86, 23.75, 26.13, 26.74, 27.39, and 29,31 was also present. The pure ERS copolymer showed a PXR diffractogram avoid of any peak, proper of an amorphous material. The dispersion of CUR in the polymer blend (ERL/ERS, 30:70, in Fig. 3) did not abolish its crystallinity, since the diffraction peaks typical of the pure drug are still detectable at identical 2q values in the D13 sample, although attenuated by the dilution effect of the polymer matrix. This feature is expected to affect CUR solubility and dissolution patterns from the microparticles, besides of its in vivo pharmacokinetic profile. By examining the corresponding coevaporate with a 1:10 DPR value (batch D43), the dilution effect due to the polymer network on CUR diffraction peaks was further enhanced. In the 1:5 physical mixture (PM2), the diffraction profile of CUR crystals was more evident, suggesting that its simple dry mixing with the polymer did not induce significant changes in the physical state of the drug. Thus, the cross-comparison of PXRD and DSC data would suggest that CUR did not become totally amorphous in the Eudragit® matrix, but, upon evaporation of their organic co-solution, formed a solid solution or a homogenous microcrystalline dispersion. 3.4. Microparticle analysis 3.4.1. Size distribution Sieving of CUR-loaded microparticles gave the distribution pattern shown in Fig. 4. In general, it was difficult to find correlations between the formulation variables and particle size distribution. Apart the batch B13 (cf. Table 3), the copolymer concentration variable seemed to affect only marginally the particle size distribution; the most evident differences between each respective pair of microparticles, and especially for batches of the B series, was that a lower (1%) polymer concentration produced a higher percentage of particles with a size over 841 mm, probably indicating some aggregates, a behaviour not much dissimilar from what observed for the corresponding blank microparticles. For this reason, in the further studies the batches produced with a 4% polymer concentration were chosen. Also changes in the DPR value did not seem to cause determining differences in the size of microparticle populations. 3.4.2. Encapsulation efficiency The encapsulation efficiency and drug content values gathered in Table 3 would suggest some effects of the operational variables used. The actual amount of CUR was measured in the pooled sieve fraction between 841 and 90 mm, thus excluding large aggregates and finer powders. The free drug was removed almost completely within the <90 mm sieve fraction, as suggested both by dosing CUR concentration in this fraction (that resulted to be almost completely made by drug substance) and by the various characterization methods, which suggested the absence of large amounts of CUR on the particle surface or in the lyophilized powders. Moreover, we tried to treat neat CUR under the same protocol used for the microparticle production and we observed that, after the precipitation and lyophilization of the aqueous suspension, all the drug remained below 90-mm sieve fraction. Although the DPR values did not seem to exert effects on the

T.M.G. Pecora et al. / Journal of Drug Delivery Science and Technology 33 (2016) 88e97

93

Fig. 3. Overlapping of PXRD spectra of CUR-loaded microparticles with the corresponding spectra of native CUR, pure ERS and a 1:5 physical mixture (PM2).

drug loading results, the mean value was instead generally lower for the microparticles based on a 70:30 ERL/ERS blend (batches B) than for the ones produced with 30:70 ERL/ERS (batches D). Since ERL and ERS only show differences in terms of permeability [31] but not of solubility in ethanol and insolubility in water, these findings can be explained by supposing that CUR has a lower affinity for the ERL polymer in solution (maybe because of the higher concentration of chloride counter-ions present), and thus precipitates separately as a fine powder, discarded during the sieving process. Furthermore, between the pairs of microparticles produced using a 1 or 4% polymer concentration in the ethanol solution, the latter batches showed a lower encapsulation efficiency (Table 3). A possible explanation is that a higher polymer concentration in the organic solution produced, upon contact with the aqueous phase and precipitation of the insoluble material, larger microparticles or aggregates of drug and polymer, that were discarded by sieving as the >841 mm fraction. 3.4.3. Stability The physical stability of CUR-loaded microparticles was evaluated at either room temperature (25e30  C) or at 40  C. Results are reported in Table 4. 3.4.4. Solubility studies and in vitro CUR release Nanotechnological approaches have demonstrated to be a valid tool to improve the aqueous solubility of this very poorly soluble compound [11]. To assess whether the dispersion of CUR in Eudragit matrices would also exert a positive effect, pure CUR or microparticle batches D13 and D43 (Table 3) were dispersed in different media, simulating respectively the gastric environment (SGF) and the intestinal lumen (SIF). Fig. 5 shows the solubility at equilibrium of CUR from the various samples after incubation at 37  C for 24 h. CUR is an almost insoluble drug in aqueous media: the microencapsulation appeared to enhance the amount of CUR dissolved in SGF, up to 5-fold in the case of the D13 sample;

however, in absolute values the drug remained scarcely soluble. At pH 6.8, the enhancement of solubility was even less marked. These results can be correlated with the solid-state analysis of these microparticles (see above), which have shown that CUR maintained a micro-crystalline nature in the ERS/ERL matrix; as a consequence, no benefit deriving from its amorphization was gained. The release profile of CUR from Eudragit® microparticles was studied in the same media, by incubating free CUR or microparticle samples for 2 h in SGF or for 4 h in SIF (Fig. 6). In both cases, the microencapsulation did not improve the dissolution rate of the drug, that remained at very low levels (below 5% of the loaded amount) and superimposable to that of pure CUR, at both acidic and intestinal pH values. 3.5. Photosensitivity assay CUR is known to be particularly sensitive to sun and UV light, as most natural pigments [28,29]. These authors found several degradation products depending on the irradiation wavelength. In particular, at wavelengths longer than 400 nm a breakdown of the chain connecting the two aromatic side rings in CUR molecule occurs, producing smaller phenolic compounds rich in carboxylate groups and then a polymerization of the resulting phenolic products, that gives rise to phenolic polymeric products. CUR was also found to act as photosensitizer via a singlet oxygen mechanism. In our experiments, the irradiation of pure CUR specimens at lexc ¼ 422 nm resulted in a progressive decomposition of the drug, with the loss of FLD signal (Fig. 7). Therefore, the possible protective effect exerted by the polymeric matrix on CUR photosensitivity was evaluated by exposing to controlled light conditions the microparticles batches D13 and D43 (Table 3), chosen as representative of different drug-to-polymer ratios. The time-course degradation profile of encapsulated CUR was compared to that of the pure drug. As Fig. 8 shows, the Eudragit® matrix was able to protect the

94

T.M.G. Pecora et al. / Journal of Drug Delivery Science and Technology 33 (2016) 88e97

Fig. 4. Cumulative and class size distribution of some CUR-Eudragit® microparticle batches. See Table 3 for their detailed composition.

T.M.G. Pecora et al. / Journal of Drug Delivery Science and Technology 33 (2016) 88e97

95

Table 4 Percent residual CUR in D13 and D43 microparticles upon different storage conditions. Values are the mean of duplicate experiments. Batch

D13 D43

24 h

70.4% 83.4%

3 months

6 months

R.T.

40  C

R.T.

40  C

70.0% 94.1%

89.2% 84.0%

61.2% 87.1%

65.3% 74.1%

Fig. 7. CUR photo-degradation HPLC-FLD profile at various irradiation times (0, 10, 30, and 120 min) (l ¼ 422 nm).

Fig. 5. Solubility at equilibrium of free CUR or encapsulated in ERS/ERL (30:70) microparticles. Each specimen was kept at 37  C for 24 h in either SGF or SIF before UV determination.

drug from the photochemical degradation. In fact, while pure CUR degraded by 50% after 30 min and was almost completely decomposed after 2 h, the microparticles containing the drug at a 1:5 CURpolymer ratio (sample D13) still contained 70% of CUR after 1 h and 40% of intact CUR after 2 h of exposition to direct UV light. Moreover, a higher ratio between the polymer and the drug (sample D43) resulted in an even lower photodegradation rate, with more than 70% intact drug still detected after 2 h. The values differ from other data reported in a recent related study [15], essentially due to the different experimental set-up of the assay and the use of nanoparticles; however, the general trend that micro- or nanoencapsulation of CUR in a polymeric network would protect the drug from photo-degradation is confirmed. 4. Discussion

Fig. 8. Photochemical degradation profiles of neat or microencapsulated CUR upon exposure to UV light (lexc ¼ 422 nm).

CUR has been used as an either a model or active compound, until recent, for encapsulation in various micro- and nanosystems based on polymers and lipid materials. In particular, nanocarriers have been produced with the aim at assessing the effects on cancer,

either in vitro or in vivo. Among the drug delivery approaches useful for the oral administration of antioxidants, polymeric microparticles often remain a useful strategy, although less ‘fashionable’ than the more

Fig. 6. In vitro dissolution profile of free CUR or encapsulated in ERS/ERL (30:70) microparticles after 2 h incubation in SGF (left) or 4 h incubation in SIF (right). Standard error bars were within ±4% of each experimental value and were omitted for clarity.

96

T.M.G. Pecora et al. / Journal of Drug Delivery Science and Technology 33 (2016) 88e97

Fig. 9. Light microscopy pictures (20) of blank and CUR-loaded microparticles: a) and c): blank B and D microparticles (Table 1); b), B13 batch, and d) D12 batch (Table 3). Lyophilized samples were resuspended in water containing Tween 80 (0.5%, w/v) immediately before the analysis. Bars correspond to 100 nm.

recent nanotechnologies. Literature presents some contributions in which CUR has gained physico-chemical benefits from microencapsulation, such as an improved solubility in aqueous media [11,34,35] and protection from photodegradation [15,36]. In this study, microparticles of CUR were produced by evaporation of organic co-solutions of the drug and ERL/ERS, a simple and scalable technique already used by some us to obtain pharmaceutically valid microparticles [19]. The present study actually belongs to a larger project, whose scope is to obtain polymeric microcarriers based on Eudragit® Retard polymers for the oral delivery of naturally occurring active compounds. Apart some favorable physicochemical properties, the interest of drug delivery technology toward ERS and ERL is also due to fact that they can be blended at various ratios, to modulate the permeability of the microparticle matrix and thus the release profile of the encapsulated drugs [21,22,32]. Microparticulate-based drug formulations possess a number of advantage over single-unit medicines, such as a slower gastro-intestinal transit and a greater superficial area, associated to an improved absorption, less inter-subject variability, constant and homogeneous drug delivery and a decreased local irritation of the intestinal mucosa [37]. Moreover, we have planned in a following study to add to these polymers different percentages of Eudragit® S100 or L100, which are anionic acrylic copolymers showing a pH-dependent solubility and that can ensure an intestinal or even a colon-specific release of CUR [38,39]. After a preliminary analysis of blank (unloaded) microparticles, we selected some polymer compositions to encapsulate CUR: microparticles were therefore prepared using a 70:30 ERL/ERS blend (batches B) and a 30:70 ERL/ERS blend (batches D) (Table 3). Such a choice was mainly linked to the satisfactory size distribution pattern of the corresponding void microparticles, that could avoid to obtain excessively fine powders or large aggregates after the addition of the drug. The experimental findings showed that these polymer blends are able to produce homogenous microparticle

populations, that can be also lyophilized without the need of cryoprotector agents: in fact, and differently from what we observed, for instance, with analogous systems containing lipoic acid [16], the lyophilized microparticles were free-flowing and did not show the formation of large or sticky aggregates. The produced systems were stable when stored at room temperature, far from direct light; a residual amount of CUR around 70e94% was still measured after 3 months and around 85% after 6 months in these conditions. As expected, a lower DPR value (i.e., 1:10 vs 1:5) resulted in a better chemical stability of the drug, due to the protective effect of the polymeric network. Upon storage at 40  C analogous values were registered, suggesting that temperature is not a direct cause of chemical degradation of microencapsulated CUR in the mid-term. A morphological investigation of the produced microparticles is shown in Fig. 9. Under light microscope, the lyophilized blank microparticles appeared as quadrangular or rectangular transparent structures. The presence of CUR was instead clearly visible as thin needles and as a diffuse orange color inside the microparticle matrix; in particular, the density of these crystals appeared greater for the D12 batch specimen, i.e., the ERL/ERS 30:70 microparticles which showed a very high value of encapsulation efficiency (see Table 3). When submitted to a specific photodegradation assay, the encapsulation of CUR in ERS/ERS appeared effective in protecting the drug from UV-light induced chemical damage, thus adding a potential technological value to these carrier systems. Also in this case, the 1:10 DPR value appeared to be more effective than the higher one. Unfortunately, the experimental data suggested that ERL and ERS were not very effective in improving the solubility of CUR in gastric and intestinal media, a very problematic hint that limits the bioavailability and clinical efficacy of this compound. As a possible explanation, the solid state characterization of the microparticles suggested that the addition to water of CUR and polymers cosolutions did not led to a complete amorphization of the drug,

T.M.G. Pecora et al. / Journal of Drug Delivery Science and Technology 33 (2016) 88e97

which maintained a microcrystalline nature within the polymeric network. As a consequence, the potential benefits, in terms of the higher solubility of the amorphous form, have not been attained. On the basis of these initial results, in vivo studies are in course with selected microparticle batches on a murine model of inflammatory bowel disease (IBD), to assess the effect of microencapsulation upon the oral availability and the local anti-inflammatory activity of CUR, that is among the natural remedies proposed for this clinical situation [40]. Declaration of interest The authors report no conflicts of interest. Acknowledgements All Authors declare no conflict of interest in the present study. Labomar srl is greatly acknowledged for providing the active compound. References [1] C.S. Beevers, S. Huang, Pharmacological and clinical properties of curcumin, Bot. Targets Ther. 1 (2011) 5e18. [2] H. Zhou, C.S. Beevers, S. Huang, The targets of curcumin, Curr. Drug Targets 12 (2011) 332e347. [3] S.C. Gupta, S. Patchva, B.B. Aggarwal, Therapeutic roles of curcumin: lessons learned from clinical trials, AAPS J. 15 (2013) 195e218. [4] B. Joe, M. Vijaykumar, B.R. Lokesh, Biological properties of curcumin-cellular and molecular mechanisms of action, Crit. Rev. Food Sci. Nutr. 44 (2004) 97e111. [5] S. Reuter, S.C. Gupta, B. Park, A. Goel, B.B. Aggarwal, Epigenetic changes induced by curcumin and other natural compounds, Genes Nutr. 6 (2011) 93e108. [6] V. Ravindranath, N. Chandrasekhara, Absorption and tissue distribution of curcumin in rats, Toxicol 16 (1980) 259e265. [7] R.A. Sharma, W.P. Steward, A.J. Gescher, Pharmacokinetics and pharmacodynamics of curcumin, Adv. Exp. Med. Biol. 595 (2007) 453e470. [8] A. Kumar, A. Ahuja, J. Ali, S. Baboota, Conundrum and therapeutic potential of curcumin in drug delivery, Crit. Rev. Ther. Drug Carr. Syst. 27 (2010) 279e312. [9] K. Kumar, A.K. Rai, Evaluation of anti-inflammatory and anti-arthritic activities of floating microspheres of herbal drugs, Int. Res. J. Pharm. 3 (2012) 186e193. [10] M. Sun, X. Su, B. Ding, X. He, X. Liu, A. Yu, H. Lou, G. Zhai, Advances in nanotechnology-based delivery systems for curcumin, Nanomed. (Lond.) 7 (2012) 1085e1100. [11] U. Hani, H.G. Shivakumar, Solubility enhancement and delivery systems of curcumin, a herbal medicine: a review, Curr. Drug Deliv. 11 (2014) 792e804. [12] S. Prasad, A.K. Tyagi, B.B. Aggarwal, Recent developments in delivery, bioavailability, absorption and metabolism of curcumin: the golden pigment from golden spice, Cancer Res. Treat. 46 (2014) 2e18. [13] H.-F. Ji, L. Shen, Can improving bioavailability improve the bioactivity of curcumin? Trends Pharmacol. Sci. 35 (2014) 265e266. [14] L. Mazzarino, C.L. Dora, I.C. Bellettini, E. Minatti, S.G. Cardoso, E. Lemos-Senna, Curcumin-loaded polymeric and lipid nanocapsules: preparation, characterization and chemical stability evaluation, Lat. Am. J. Pharm. 29 (2010) 933e940. [15] J. Li, I.W. Lee, G.H. Shin, X. Chen, H.J. Park, Curcumin-Eudragit® E PO solid dispersion: a simple and potent method to solve the problems of curcumin, Eur. J. Pharm. Biopharm. 94 (2015) 322e332. [16] T.M.G. Pecora, T. Musumeci, L. Musumeci, M. Fresta, R. Pignatello, Evaluation of Eudragit® Retard polymers for the microencapsulation of alpha-lipoic acid, Curr. Drug Deliv. (2016) (Epub ahead of print); (DOI:10.2174/ 1567201812666151016095342#sthash.53mTUQtv.dpuf). [17] R. Bodmeier, H. Chen, Preparation and characterization of microspheres containing the anti-inflammatory agents, indomethacin, ibuprofen, and ketoprofen, J. Contr. Release 10 (1989) 167e175. [18] Y. Kawashima, T. Iwamoto, T. Niwa, H. Takeuchi, T. Hino, Size control of ibuprofen microspheres with an acrylic polymer by changing the pH in an aqueous dispersion medium and its mechanism, Chem. Pharm. Bull. 41 (1993)

97

191e195. [19] R. Pignatello, M.A. Vandelli, P. Giunchedi, G. Puglisi, Properties of tolmetinloaded Eudragit RL100 and RS100 microparticles prepared by different techniques, STP Pharma Sci. 7 (1997) 148e157. [20] R. Pignatello, D. Amico, S. Chiechio, P. Giunchedi, C. Spadaro, G. Puglisi, Preparation and analgesic activity of Eudragit RS100 microparticles containing diflunisal, Drug Deliv. 8 (2001) 35e45. [21] R. Pignatello, C. Bucolo, P. Ferrara, A. Maltese, A. Puleo, G. Puglisi, Eudragit RS100 nanosuspensions for the ophthalmic controlled delivery of ibuprofen, Eur. J. Pharm. Sci. 16 (2002) 53e61. [22] R. Pignatello, C. Bucolo, G. Puglisi, Ocular tolerability of Eudragit RS100 and RL100 nanosuspensions as carriers for ophthalmic controlled drug delivery, J. Pharm. Sci. 91 (2002) 2636e2641. [23] R. Pignatello, C. Bucolo, G. Spedalieri, A. Maltese, G. Puglisi, Flurbiprofenloaded acrylate polymer nanosuspensions for ophthalmic application, Biomaterials 23 (2002) 3247e3255.
, R. Pignatello, Eudragit [24] C. Bucolo, A. Maltese, F. Maugeri, G. Puglisi, B. Busa RL100 nanoparticle system for the ophthalmic delivery of cloricromene, J. Pharm. Pharmacol. 56 (2004) 841e846. [25] A. Trapani, V. Laquintana, N. Denora, A. Lopedota, A. Cutrignelli, M. Franco, G. Trapani, G. Liso, Eudragit RS100 microparticles containing 2hydroxypropyl-beta-cyclodextrin and glutathione: physicochemical characterization, drug release and transport studies, Eur. J. Pharm. Sci. 30 (2007) 64e74. [26] S. Pandav, J.B. Naik, Preparation and in vitro evaluation of ethylcellulose and polymethacrylate resins loaded microparticles containing hydrophilic drug, J. Pharm. (2014), http://dx.doi.org/10.1155/2014/904036. Article ID 904036, 5 pages. [27] U.K. Sharma, A. Verma, S.K. Prajapati, H. Pandey, A.C. Pandey, In vitro, in vivo and pharmacokinetic assessment of amikacin sulphate laden polymeric nanoparticles meant for controlled ocular drug delivery, Appl. Nanosci. 5 (2015) 143e155, http://dx.doi.org/10.1007/s13204-014-0300-y. [28] H.H. Tønnesen, J. Karlsen, G.B. Van Henegouwen, Studies on curcumin and curcuminoids. VIII. Photochemical stability of curcumin, Z. Leb. Unters Forsch 183 (1986) 116e122. ~ amares, J.V. Garcia-Ramos, S. Sanchez-Cortes, Degradation of curcu[29] M.V. Can min dye in aqueous solution and on ag nanoparticles studied by ultravioletevisible absorption and surface-enhanced Raman spectroscopy, Appl. Spectrosc. 60 (2006) 1386e1391. [30] E. Stippler, S. Kopp, J.B. Dressman, Comparison of US Pharmacopeia Simulated Intestinal Fluid TS (without pancreatin) and Phosphate Standard Buffer pH 6.8, TS of the International Pharmacopoeia with respect to their use in in vitro dissolution testing. [31] Evonik data sheets (http://eudragit.evonik.com/product/eudragit/en/ products-services/eudragit-products/sustained-release-formulations/Pages/ default.aspx). Last access on November, 2015. [32] M. Kiliçarslan, T. Baykara, Effects of the permeability characteristics of different polymethacrylates on the pharmaceutical characteristics of verapamil hyhdrochloride-loaded microspheres, J. Microencapsul. 21 (2004) 175e189. [33] M.M.L. Del Socorro, F.G. Teves, M.R.S.B. Madamba, DNA-binding activity and partial characterization by Fourier Transform Infrared Spectroscopy (FT-IR) of Curcuma longa L. SC-CO2 extracts, Int. Res. J. Biol. Sci. 2 (2013) 40e44. [34] D. Guzman-Villanueva, I.M. El-Sherbiny, D. Herrera-Ruiz, H.D.C. Smyth, Design and in vitro evaluation of a new nano-microparticulate system for enhanced aqueous-phase solubility of curcumin, BioMed. Res. Int. (2013), http:// dx.doi.org/10.1155/2013/724763. Article ID 724763, 9 pages. [35] C.C.C. Teixeira, L.M. Mendonça, M.M. Bergamaschi, R.H.C. Queiroz, G.E.P. Souza, L.M.G. Antunes, L.A.P. Freitas, Microparticles containing curcumin solid dispersion: stability, bioavailability and anti-inflammatory activity, AAPS PharmSciTech (2015) [Epub ahead of print] (DOI: 10.1208/s12249-0150337-6). [36] I.F. Nata, K.J. Chen, C.K. Lee, Facile microencapsulation of curcumin in acetylated starch microparticles, J. Microencapsul. 31 (2014) 344e349. [37] S. Chandran, K.S. Sanjay, A. Asghar, Microspheres with pH modulated release: design and characterization of formulation variables for colonic delivery, J. Microencapsul. 26 (2009) 420e431. [38] F. Grimpen, P. Pavli, Advances in the management of inflammatory bowel disease, Intern. Med. J. 40 (2010) 258e264. at, pH[39] A. Beloqui, R. Coco, P.B. Memvanga, B. Ucakar, A. des Rieux, V. Pre sensitive nanoparticles for colonic delivery of curcumin in inflammatory bowel disease, Int. J. Pharm. 473 (2014) 203e212. [40] L. Vecchi Brumatti, A. Marcuzzi, P.M. Tricarico, V. Zanin, M. Girardelli, A.M. Bianco, Curcumin and inflammatory bowel disease: potential and limits of innovative treatments, Molecules 19 (2014) 21127e21153.