http://informahealthcare.com/ddi ISSN: 0363-9045 (print), 1520-5762 (electronic) Drug Dev Ind Pharm, Early Online: 1–7 ! 2014 Informa Healthcare USA, Inc. DOI: 10.3109/03639045.2014.935393
RESEARCH ARTICLE
Ethylcellulose nanoparticles with bimodal size distribution as precursors for the production of very small nanoparticles Drug Dev Ind Pharm Downloaded from informahealthcare.com by Chinese University of Hong Kong on 02/10/15 For personal use only.
Philip Wachsmann1 and Alf Lamprecht1,2 1
Laboratory of Pharmaceutical Technology and Biopharmaceutics, University of Bonn, Bonn, Germany and 2Laboratory of Pharmaceutical Engineering, EA4267, University of Franche-Comte´, Besanc¸on, France Abstract
Keywords
A common technique for the preparation of polymeric nanoparticles (NPs) from preformed polymers is the emulsification solvent evaporation (ESE) method. However, the particle size of such carriers can typically not reduced below 100 nm. A bimodal distribution of particle size when applying ESE to the preparation of ethylcellulose (EC) NPs was intended to obtain very small particles in a size range below 50 nm. The proportion and size of the small particle fraction (SPF) depended on the surfactant as well as on the EC type and concentration. The preparation was conducted with different pharmaceutically relevant surfactants (polyoxyethylene (23) lauryl ether, sodium dodecyl sulfate, cetyltrimethylammonium bromide, polyvinyl alcohol and polysorbate 20) and all permitted obtaining very small NPs. After purification from excess surfactant by diafiltration and separation of the SPF by centrifugation, monodispersed particles with mean sizes between 20.6 ± 2.3 nm and 49.7 ± 4.8 nm could be isolated. The entrapment of a lipophilic model drug led to encapsulation rates between 34.0 ± 2.4% and 78.2 ± 12.6%, which were size and surfactant dependent. The preparation of polymeric NPs in a size below 50 nm by a simple centrifugation step holds promise for therapeutic applications where larger particles would be inefficient.
Emulsification solvent evaporation, indometacin, nanosphere, polymeric nanoparticle, surfactant
Introduction Nanoparticulate drug carrier systems are promising tools for diverse biopharmaceutical applications. The incorporation of active pharmaceutical ingredients into them and the choice of an adequate carrier system allow the modification of the biological and physicochemical behavior compared to the free drug. For example, the body distribution can be altered allowing more specific drug dosing, selective drug targeting and also intracellular drug delivery can be achieved1. Polymeric nanoparticles (NPs) are a common drug carrier type in this context and used for a wide range of the before-mentioned therapeutic applications. One crucial factor for the biological activity of nanocarriers is their size. On cellular level, the size determines the way of particle interaction, e.g. the adhesion to cell surface and the way of internalization. Possible ways of internalization are phagocytosis, clathrin-mediated endocytosis, caveolae-mediated endocytosis and macropinocytosis2. The mentioned pathways lead to different intracellular localizations of the internalized particles and show different size specificities. Phagocytosis and macropinocytosis exhibit minor size specificity, although a preference for larger particles is reported for the former3,4. Whereas clathrinmediated endocytosis and caveolae-mediated endocytosis have a size specificity for particles smaller than 200 nm and 500 nm, Address for correspondence: Alf Lamprecht, Laboratory of Pharmaceutical Technology and Biopharmaceutics, University of Bonn, Gerhard-Domagk-Str. 3, D-53121 Bonn, Germany. Tel: +49 228 73 52 43. Fax: +49 228 73 52 68. E-mail: [email protected]
History Received 28 February 2014 Revised 26 May 2014 Accepted 5 June 2014 Published online 7 July 2014
respectively5,6. Furthermore, it was shown that particles with diameters below 42 nm are internalized by a non-clathrin and noncaveolae-dependent pathway7. Thus, variation of particle size allows influencing particle–cell interactions. However, the standard techniques for the preparation of polymeric NPs from preformed polymers lead to sizes above approximately 100 nm. One of the main preparation methods for the incorporation of lipophilic drugs into the polymeric matrix of NPs is the oilin-water emulsification solvent evaporation (ESE) technique8. In this study, the lipophilic drug and the polymer are dissolved in an organic solvent, like methylene chloride or ethyl acetate. This organic solution is then emulsified in an external aqueous phase containing one or more surfactants with the intention to stabilize the emulsion interface. Removing the organic solvent by continuous stirring, increase of temperature or by reduced pressure eventually forms the drug-loaded NPs. ESE enables the preparation of NPs consisting of a wide variety of homo- and copolymers, such as biodegradable polymers like poly(lactic acid) (PLA) and poly(lactic-co-glycolic acid) (PLGA) or non-biodegradable polymers like ethylcellulose (EC) or ammonium methacrylate copolymers9. Similarly, the use of diverse surfactants for the preparation of such NPs has been reported in literature influencing the surface properties of the drug carriers10. The available range includes polymeric and non-polymeric as well as charged and non-charged surfactants. The choice leads to altered surface properties and biological activity of the nanospheres11, since a part of the surfactant remains on the particle surface even after extensive washing12.
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Chinese University of Hong Kong on 02/10/15 For personal use only.
2
P. Wachsmann & A. Lamprecht
The ESE technique allows a particle size reduction down to approximately 100 nm and diameters below that ‘‘threshold’’ are hardly found in the literature13–15. However, in several cases, a further size reduction appears desirable to increase targeting selectivity11 or to circumvent certain cellular mechanisms that may impede full delivery efficiency. In this study, ESE has been applied to the formulation of EC-NP with slight modifications. EC has served as model polymer due to its accessibility in different qualities. Its broad use for peroral and dermal applications is also well characterized. We found a bimodal size distribution of NPs when high surfactant concentrations were used and the small particle fraction (SPF) has a mean diameter of approximately 25–70 nm. The aim of this study was to analyze whether the production and isolation of small NPs (550 nm) was possible by efficient separation method and to what extent a variation of surfactant type or polymer properties and the respective concentrations would influence the NP size distribution and the SPF proportion.
Methods
Drug Dev Ind Pharm, Early Online: 1–7
was pumped (200 ml/min) through the filtration device, and the filtrate was replaced gradually with 1 l (Brij 35) or 500 ml (SDS, CTAB) distilled water. Thereafter, the suspension was concentrated to 10 ml and then diluted to 15.0 ml with distilled water. The monomodal distributed particles with diameters around 200 nm were purified by two centrifugation steps at 3840 g for 1 h (Brij 35) or 30 min (SDS, CTAB) and redispersed in distilled water to 10.0 ml. Isolation of SPF After purification from excess surfactant by diafiltration, the bimodal distributed particle suspension was centrifuged for 90 min with 8000g (1 g/100 ml Brij 35, 100 mg EC), 12 000 g (3 g/100 ml Brij 35, 300 mg EC) or 19 000 g (all other formulations) to separate the SPF from the large particle fraction (LPF). The lowest centrifugation force, which allowed the recovery of a monomodal distribution, was chosen. The supernatant was analyzed for its size distribution by photon correlation spectroscopy (PCS).
Materials EC (EthocelÕ standard 4 premium, EthocelÕ standard 10 premium and EthocelÕ standard 45 premium) was a gift from Colorcon (Dartford, England). Polyoxyethylene (23) lauryl ether (Brij 35), cetyltrimethylammonium bromide (CTAB) and indomethacin were purchased from Sigma-Aldrich (Steinheim, Germany). Sodium dodecyl sulfate (SDS) was bought from Carl Roth (Karlsruhe, Germany). Polyvinyl alcohol (PVA; Mowiol 4-88) was a gift from Kuraray (Frankfurt, Germany). Polysorbate 20 was bought from Caelo (Hilden, Germany). All other chemicals were of analytical grade. Particle preparation and purification The NPs were prepared by an oil-in-water ESE method; 30, 100 or 300 mg EC was dissolved in 5 ml ethylacetate. To produce drug-loaded NPs, 10 mg indometacin was added. The organic phase was poured into 10 ml aqueous surfactant solution. Different concentrations of polysorbate 20, Brij 35, PVA, SDS or CTAB were used as surfactant. After building an emulsion by ultrasonication for 4 min at 50 W (Sonoplus, Bandelin Electronic), the organic solvent was removed under reduced pressure. Excess surfactant was removed either by dialysis (SpectraPor Biotech MWCO 100 kDa, Spectrum Labs) or by diafiltration as follows: for the separation and encapsulation experiments, the NP suspension was diluted to 100 ml and purified by diafiltration (VivaFlow 50, Sartorius Sedim GmbH). The suspension
Particle size and zeta potential The NPs were analyzed for their size distribution by PCS (ZetaPlus, Brookhaven Instruments Corporation) at a fixed angle of 90 at 25 causing either the mode for lognormal or multimodal distributions. All results are calculated as volume distributions. When the mode for multimodal distributions was used, the distribution was either monomodal or bimodal. The size distribution was characterized by the proportion of the two fractions and their mean diameter. The mean diameter was calculated as frequency-weighted mean of the class interval middles, which were obtained from the particle size analysis software. Table 1. Properties of isolated small particle fraction (SPF) and 200 nm particles prepared with 100 mg EthocelÕ 4 (mean ± SD; n ¼ 3).
Surfactant Brij 35 Brij 35 CTAB CTAB SDS SDS
Concentration [g/100 ml] 3 0.035 0.3 0.04 1 0.004
Diameter [nm] 25.4 246.8 30.9 221.4 34.5 219.0
(±2.3) (±18.8) (±0.2) (±29.9) (±2.1) (±22.9)
Polydispersity index 0.223 0.029 0.166 0.083 0.196 0.053
(±0.024) (±0.042) (±0.005) (±0.010) (±0.063) (±0.022)
Zeta potential [mV] 11.4 14.6 22.0 4.8 34.5 22.1
(±4.3) (±0.6) (±5.2) (±2.3) (±1.7) (±4.6)
Figure 1. SEM images of ethylcellulose nanospheres prepared with 3 g/100 ml Brij 35 and 100 mg EthocelÕ 4 before (A) and after (B) centrifugation.
DOI: 10.3109/03639045.2014.935393
Zeta potential was determined with the same device. Small particles were concentrated with Amicon Ultra filtration devices (Molecular weight cut off 50 kDa, Merck Millipore), because signal intensity of non-concentrated samples was too low to obtain reliable results. Scanning electron microscopy Particles were imaged by Field Emission Scanning Electron Microscopy (FESEM). The nanosphere suspension was placed on a cover glass and dried overnight in a desiccator. The samples were examined by FESEM using a Hitachi SU8030 at 0.1–0.5 kV (Hitachi, Tokyo, Japan).
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Chinese University of Hong Kong on 02/10/15 For personal use only.
Drop shape analysis Drop shape analysis was performed to compare the different EC types in terms of interface activity; 100 mg EthocelÕ 4, EthocelÕ 10 or EthocelÕ 45 was dissolved in 10 ml methylene chloride. These solutions or pure methylene chloride were measured against distilled water. The interfacial tension was calculated from the drop shape using an EasyDropdevice (Kru¨ss GmbH). The interfacial tension was measured every 30 s for a period of 4 min, and the values were averaged. Ethyl acetate had to be substituted by methylene chloride due to instability of the ethyl acetate droplet with water as external phase. Encapsulation rate and drug release Ten milliliters of bimodal-distributed NPs suspension was centrifuged to isolate the SPF as described above; 4 ml centrifuged suspension was freeze-dried and weighed. The particles were dissolved in 10.0 ml methanol, and indomethacin was quantified with a spectrophotometer at 320 nm. Besides, 200 nm particles were resuspended in 10 ml distilled water, and 1 ml was freeze-dried and analyzed as described above. Purified and isolated SPFs and monomodal-distributed 200 nm particles were tested for their indomethacin release by a modified basket method16. Therefore, 4 ml (Brij 35, SDS) or 3 ml (CTAB) SPF were placed in the dissolution apparatus (PTW S, Pharmatest), and 25 ml pre-warmed release medium (phosphate buffer pH 6.8, 37 C) was added. The 200 nm particles were diluted with distilled water to an appropriate concentration.
Ethylcellulose nanoparticles with bimodal size distribution
3
Four milliliters of the dilution were placed in the dissolution apparatus. At predefined time points, 1 ml of release medium was withdrawn and replaced by fresh buffer. The amount of released indomethacin was determined by HPLC according to the following method: acetonitrile:phosphoric acid (0.2%) 60:40; RP-18 column; flow rate: 1.5 ml/min; UV absorption ¼ 320 nm17. Statistical analysis All results are expressed as mean ± standard deviation (SD). The results were statistically analyzed by analysis of variance (ANOVA) followed by the Students–Newman–Keuls post-test. SigmaStat for Windows 2.0 was used to calculate the p values. Differences were considered to be significant at p50.05.
Results Particle characterization The ESE preparation process for EC-NP led to their bimodal particle size distribution with higher concentrations of surfactant. The SPF was isolated from the LPF by centrifugation of the particle dispersion as was proven by SEM imaging and exemplarily shown for Brij35-NP before and after the centrifugation step (Figure 1). The former bimodal distribution of particle size was monomodal after the centrifugation. Lower surfactant concentrations allowed the production of monomodal-distributed particles, which were optimized for sizes around 200 nm. Size and zeta potential of the particles depended on surfactant type and concentration (Table 1). The absolute value of zeta potential increases for the small particles prepared with ionic surfactants. Influence of preparation parameters Increased surfactant concentrations led to decreasing particle sizes and increasing polydispersity indices (data not shown). Bimodal size distribution was observed above a surfactantspecific concentration for each formulation type. Mean diameters in the SPF varied between 31 nm and 73 nm, while the LPF lay between 98 nm and 617 nm (Figure 2). Increasing surfactant concentrations resulted in an increasing SPF proportion, while mean particle size of the SPF was slightly
Figure 2. Ethylcellulose nanospheres were prepared with 100 mg EthocelÕ 4 and different surfactant types and concentrations. Data of the distribution are shown as diameter (A) of the small particle fraction (SPF) and the large particle fraction (LPF) as well as proportion of the SPF (B) (mean ± SD; n ¼ 3).
4
P. Wachsmann & A. Lamprecht
Drug Dev Ind Pharm, Early Online: 1–7
decreased. However, above an optimal concentration, a higher surfactant concentration led generally to a decreasing SPF proportion and an increasing particle size. This effect was not visible with CTAB due to its limited solubility in water. The highest proportion of the SPF was reached with the nonionic surfactants polysorbate 20, Brij 35 and PVA with approximately 95% (Table 2). However, PVA led to comparatively large SPF
Table 2. Size of large particle fraction (LPF) and small particle fraction (SPF) and proportion of SPF after optimizing the surfactant concentration for formation of SPF (mean ± SD; n ¼ 3).
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Chinese University of Hong Kong on 02/10/15 For personal use only.
Surfactant Brij 35 Polysorbate 20 PVA CTAB SDS
Concentration [g/100 ml] 3 10 10 0.3 1
Diameter SPF [nm] 40.3 31.2 62.4 35.7 38.1
(±5.9) (±4.7) (±6.0) (±0.1) (±5.1)
Diameter LPF [nm] 173.1 150.1 194.8 124.5 97.7
(±10.5) (±2.7) (±19.7) (±6.4) (±3.4)
Proportion SPF [%] 93.3 95.7 94.7 85.7 83.3
(±0.6) (±1.5) (±1.5) (±6.4) (±3.8)
sizes. The ionic surfactants SDS and CTAB showed a SPF proportion of approximately 85%. When total surface areas were calculated based on the mean size distribution of SPF and LPF, a correlation was observed between surfactant concentration and total surface area when excluding concentrations above the SPF optimum where solution viscosity became a dominant factor. Even the regression of the complete data set suggested a correlation of total surface area and surfactant amount seemingly being independent from the surfactant type (Figure 3). Besides surfactant type and concentration, also the amount and molecular weight of EC confirmed to have an influence on the size distribution of EC-NP (Figure 4). The minimum surfactant concentration for the formation of the bimodal size distribution was a dependent variable of the EC amount. Furthermore, the proportion of the SPF in the optimum slightly differed depending on the EC concentration (30 mg EC: 95.3 ± 0.6%; 100 mg EC: 95.7 ± 1.5%; 300 mg EC: 86.7 ± 2.9%), whereas a statistically significant difference was obtained for 300 mg EC compared to the other two EC amounts. However, the size of the SPF was surprisingly unaffected (30 mg EC: 37.4 ± 3.5 nm; 100 mg EC: 31.2 ± 4.7 nm; 300 mg EC: 36.6 ± 3.3 nm). Similarly, EC of different viscosities influenced SPF proportion and minimum surfactant concentration for bimodal distribution significantly (Figure 5). The highest proportion of the SPF is achieved with a concentration of 81 mM polysorbate 20 for all EC types simultaneously. The mean NP diameter at this surfactant concentration varied, however, significantly with the molecular weight of EC (EthocelÕ 4: 31.2 ± 4.7 nm; EthocelÕ 10: 42.5 ± 6.4 nm; and EthocelÕ 45: 52.2 ± 1.4 nm). In order to determine the additional influence of surface-active properties of EC, drop shape analyses were performed with an EC-methylene chloride/water system. The interface tension lowering effect was less pronounced for EthocelÕ 45, while the other two types resulted in similar observations (Figure 6). Encapsulation rate and indomethacin release
Figure 3. Non-linear regression of surface area and surfactant amount. Surface area was calculated from the proportion of SPF and the mean sizes of SPF and LPF. y ¼ 16.8 ln(x) 23.1; R2 ¼ 0.683.
The concentration of surfactant was adjusted to obtain particles of comparable sizes around 200 nm after encapsulation of indomethacin. Therefore, the concentration of CTAB and SDS for blank particles and indomethacin-loaded particles differs. The encapsulation rates varied between 34.0% and 78.2% (Table 3). SDS and Brij 35 showed a comparable encapsulation
Figure 4. Ethylcellulose nanospheres were prepared with different amounts of EthocelÕ 4. The diameter (A) of the small particle fraction (SPF) and the large particle fraction (LPF) as well as proportion of the SPF (B) of ethylcellulose nanospheres are plotted against the surfactant concentration (mean ± SD; n ¼ 3).
Ethylcellulose nanoparticles with bimodal size distribution
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Chinese University of Hong Kong on 02/10/15 For personal use only.
DOI: 10.3109/03639045.2014.935393
5
Figure 5. Ethylcellulose nanospheres were prepared with different ethyl cellulose types (EthocelÕ 4, 10 and 45). The diameter (A) of the small particle fraction (SPF) and the large particle fraction (LPF) as well as proportion of the SPF (B) are plotted against the surfactant concentration (mean ± SD; n ¼ 3).
profiles with t50% values of 445 ± 82 min for CTAB, 489 ± 154 min for SDS and 503 ± 102 min for Brij 35. However, release kinetics of SPF were influenced by the surfactant type: SDS and Brij 35 released 50% of the encapsulated drug within 553 ± 88 min and 406 ± 184 min, respectively, while CTAB showed a significantly slower indomethacin release (1262 ± 53 min).
Discussion
Figure 6. Interfacial tension at the methylene chloride/water interface. The interfacial tension was analyzed by drop shape analysis using different ethylcellulose types (EthocelÕ 4, 10 and 45) (mean ± SD; n ¼ 3). Table 3. Hydrodynamic diameter, polydispersity index and encapsulation rate of indomethacin-loaded EC-NPs prepared with EthocelÕ 4.
Surfactant Brij 35 Brij 35 CTAB CTAB SDS SDS
Concentration [g/100ml] 3 0.035 0.3 0.045 1 0.003
Diameter [nm] 27.9 200.5 33.9 208.6 30.5 198.6
(±2.3) (±19.3) (±1.4) (±8.2) (±2.1) (±7.5)
Polydispersity Encapsulation index rate [%] 0.144 0.059 0.197 0.109 0.149 0.070
(±0.016) (±0.009) (±0.019) (±0.025) (±0.008) (±0.011)
34.0 80.0 78.2 96.7 41.1 56.1
(±2.4) (±1.6) (±12.6) (±5.0) (±2.7) (±5.5)
For each surfactant, the SPF was isolated, and monomodal-distributed 200 nm particles were prepared separately (mean ± SD; n ¼ 3).
rate, while EC-NP prepared with CTAB had nearly twice higher drug content. Indomethacin release of isolated SPF and monomodal-distributed 200 nm particles is shown in Figure 7. Two-hundred-nanometer particles exhibited comparable release
In the ESE method, the formulation of the nanoemulsion and removal of the organic solvent under reduced pressure can essentially influence the size distribution of NPs. Consequently, the formation of the SPF depends highly on energy input and amount of surfactant that is available for the surface stabilization18 by which the re-coalescence rate of newly formed small droplets can be reduced19. Thereafter, coalescence can also take place during evaporation of the organic phase. EC-NP prepared by ESE with ethyl acetate as solvent are formed due to coalescence of several unstable emulsion droplets, whereas PLA droplets show lower tendency to coalesce20. On the other hand, polystyrene NPs prepared by the same technique but with methylene chloride, chloroform or toluene as solvent show a very low coalescence rate during the evaporation21. So the extent of coalescence during evaporation depends on the solvent and the polymer properties. The size distribution of particles seems to depend mainly on an appropriate surfactant concentration, which hinders coalescence of emulsion droplets that may take place during the emulsion formation or the evaporation process or during both phases. The good correlation between surfactant concentration and surface area of the NPs throughout all surfactant types indicated that the molar concentration of the surfactant is far more determinant than the surfactant type. Besides the surfactant concentration, the polymer type and concentration also influenced the size distribution. An increasing EC concentration led to two different effects. The first effect was the co-stabilizing effect at the oil–water interface that was needed for the formation of a bimodal distribution and which EC contributed due to its surface-active properties22. A second effect was observed when the surfactant amount was able to stabilize the interface. Then, higher EC concentrations led to lower SPF
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Chinese University of Hong Kong on 02/10/15 For personal use only.
6
P. Wachsmann & A. Lamprecht
Drug Dev Ind Pharm, Early Online: 1–7
Figure 7. Release profile of monomodal-distributed 200 nm particles (A) and of isolated SPF (B). All particles were prepared with EthocelÕ 4 as polymer. Two-hundred nanometer particles were prepared with 0.035 g/100 ml Brij 35, 0.045 g/100 ml CTAB or 0.003 g/100 ml SDS. SPF was isolated from formulations prepared with 3 g/100 ml Brij 35, 0.3 g/100 ml CTAB or 1 g/100 ml SDS. The cumulated indomethacin release is plotted against the time (mean ± SD; n ¼ 3).
proportions with larger NP sizes, because the higher viscosity of the inner phase. Surprisingly, all variations of the preparation procedure led to a bimodal size distribution. A possible explanation is the inability of the surfactant-EC combination to fully stabilize the large water/ethyl acetate interface that is generated during the preparation. Thus, a part of the small droplets coalesces and forms the LPF, whereas the other part remains as the SPF. This is in accordance with the above-mentioned observation, that during the preparation of nanospheres using ESE, the formation of EC-NP results from coalescence of several unstable emulsion droplets20. Because we used higher concentrations of surfactant, an increased stabilization of small droplets was possible and therewith the production of small NPs. However, a part of droplets still coalesces and builds the LPF. The SPF that was formed under optimized conditions exhibited mean diameters down to approximately 25 nm, which is in the size range of other nano-sized drug carriers, like lipid nanocapsules (20–100 nm)23 or polymeric micelles (15–150 nm)24. Although the appearance of LPF cannot be removed completely, its ratio can be reduced down to a level of around vol-5% of the entire formulation; and moreover, the remaining amount of a few percent LPF can be separated by several techniques. Filtration was already described to isolate the SPF of bimodal distributed PLGA particles25. However, the major drawback of filtration in laboratory scale is caking, which leads to low particle recovery, whereas centrifugation seems to be more appropriate as separation procedure26. The lipophilic model drug indomethacin was successfully encapsulated into EC-NP with three different surfactants. Even after purification by diafiltration, an acceptable encapsulation rate was achieved. The appropriateness of this method for the purification of indometacin nanocapsules was already shown27. Indomethacin-loaded LPF or SPF prepared with SDS and Brij 35 exhibited fast drug release, which was on a comparable level. The SPF of CTAB particles showed a nearly twice higher drug content than the particles prepared with the anionic and the nonionic surfactant. During the preparation of 30 nm particles, a more than 6-fold higher amount of CTAB is used than for 200 nm particles, which leads to an increased zeta potential of the small particles. The molar ratio of CTAB to indomethacin increases
from 0.4 to 2.9. Subsequently, electrostatic interaction of the cationic surfactant and the anionic drug can take place leading to a higher drug load and a considerable delay of drug release.
Conclusion EC-NPs allow the preparation of drug-loaded NPs in the size range of 30 nm via the isolation of a SPF from a bimodal particle population. The optimal settings led to a process where only minimal amounts of LPF needed to be removed. The size and proportion of the SPF depended on the surfactant concentration but much less on its type as well as on the EC type and concentration.
Declaration of interest Alf Lamprecht is thankful for the financial support from the Institut Universitaire de France. This work was partially supported by a French Government grant managed by the French National Research Agency under the program ‘‘Investissementsd’Avenir’’ with reference ANR-11-LABX-0021.
References 1. Barratt G. Therapeutic applications of colloidal drug carriers. Pharm Sci Technol Today 2000;3:163–71. 2. Hillaireau H, Couvreur P. Nanocarriers’ entry into the cell: relevance to drug delivery. Cell Mol Life Sci 2009;66:2873–96. 3. Scha¨fer V, Briesen von H, Andreesen R, et al. Phagocytosis of nanoparticles by human immunodeficiency virus (HlV)-infected macrophages: a possibility for antiviral drug targeting. Pharm Res 1992;9:541–6. 4. Tabata Y, Ikada Y. Effect of the size and surface charge of polymer microspheres on their phagocytosis by macrophage. Biomaterials 1988;9:356–62. 5. Rejman J, Oberle V, Zuhorn I, Hoekstra D. Size-dependent internalization of particles via the pathways of clathrin-and caveolae-mediated endocytosis. Biochem J 2004;377:159–69. 6. Wang Z, Tiruppathi C, Minshall RD, Malik AB. Size and dynamics of caveolae studied using nanoparticles in living endothelial cells. ACS Nano 2009;3:4110–16. 7. Lai SK, Hida K, Man ST, et al. Privileged delivery of polymer nanoparticles to the perinuclear region of live cells via a non-clathrin, non-degradative pathway. Biomaterials 2007;28: 2876–84.
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Chinese University of Hong Kong on 02/10/15 For personal use only.
DOI: 10.3109/03639045.2014.935393
8. Pinto Reis C, Neufeld RJ, Ribeiro AJ, Veiga F. Nanoencapsulation I. Methods for preparation of drug-loaded polymeric nanoparticles. Nanomedicine 2006;2:8–21. 9. Letchford K, Burt H. A review of the formation and classification of amphiphilic block copolymer nanoparticulate structures: micelles, nanospheres, nanocapsules and polymersomes. Eur J Pharm Biopharm 2007;65:259–69. 10. Wachsmann P, Moulari B, Be´duneau A, et al. Surfactant-dependence of nanoparticle treatment in murine experimental colitis. J Control Release 2013;172:62–8. 11. Abdel-Mottaleb MMA, Moulari B, Be´duneau A, et al. Surfacecharge-dependent nanoparticles accumulation in inflamed skin. J Pharm Sci 2012;101:4231–9. 12. Boury F, Marchais H, Benoit JP, Proust JE. Surface characterization of poly(alpha-hydroxy acid) microspheres prepared by a solvent evaporation/extraction process. Biomaterials 1997;18:125–36. 13. Gaucher G, Asahina K, Wang J, Leroux J-C. Effect of poly(N-vinylpyrrolidone)-block-poly(D,L-lactide) as coating agent on the opsonization, phagocytosis, and pharmacokinetics of biodegradable nanoparticles. Biomacromolecules 2009;10:408–16. 14. Go´mez-Gaete C, Tsapis N, Besnard M, et al. Encapsulation of dexamethasone into biodegradable polymeric nanoparticles. Int J Pharm 2007;331:153–9. 15. Vauthier C, Bouchemal K. Methods for the preparation and manufacture of polymeric nanoparticles. Pharm Res 2009;26: 1025–58. 16. Abdel-Mottaleb MMA, Lamprecht A. Standardized in vitro drug release test for colloidal drug carriers using modified USP dissolution apparatus I. Drug Dev Ind Pharm 2011;37:178–84.
Ethylcellulose nanoparticles with bimodal size distribution
7
17. Limnell T, Heikkila¨ T, Santos HA, et al. Physicochemical stability of high indomethacin payload ordered mesoporous silica MCM-41 and SBA-15 microparticles. Int J Pharm 2011;416:242–51. 18. Anton N, Benoit J-P, Saulnier P. Design and production of nanoparticles formulated from nano-emulsion templates – a review. J Control Release 2008;128:185–99. 19. Jafari SM, Assadpoor E, He Y, Bhandari B. Re-coalescence of emulsion droplets during high-energy emulsification. Food Hydrocolloids 2008;22:1191–202. 20. Desgouilles S, Vauthier C, Bazile D, et al. The design of nanoparticles obtained by solvent evaporation: a comprehensive study. Langmuir 2003;19:9504–10. 21. Staff RH, Schaeffel D, Turshatov A, et al. Particle formation in the emulsion-solvent evaporation process. Small 2013;9:3514–22. 22. Melzer E, Kreuter J, Daniels R. Ethylcellulose: a new type of emulsion stabilizer. Eur J Pharm Biopharm 2003;56:23–7. 23. Huynh NT, Passirani C, Saulnier P, Benoit JP. Lipid nanocapsules: a new platform for nanomedicine. Int J Pharm 2009;379:201–9. 24. Jones M, Leroux J. Polymeric micelles – a new generation of colloidal drug carriers. Eur J Pharm Biopharm 1999;48:101–11. 25. Prabha S, Zhou W-Z, Panyam J, Labhasetwar V. Size-dependency of nanoparticle-mediated gene transfection: studies with fractionated nanoparticles. Int J Pharm 2002;244:105–15. 26. Gaumet M, Gurny R, Delie F. Fluorescent biodegradable PLGA particles with narrow size distributions: preparation by means of selective centrifugation. Int J Pharm 2007;342:222–30. 27. Limayem I, Charcosset C, Fessi H. Purification of nanoparticle suspensions by a concentration/diafiltration process. Sep Purif Technol 2004;38:1–9.
RESEARCH ARTICLE
Ethylcellulose nanoparticles with bimodal size distribution as precursors for the production of very small nanoparticles Drug Dev Ind Pharm Downloaded from informahealthcare.com by Chinese University of Hong Kong on 02/10/15 For personal use only.
Philip Wachsmann1 and Alf Lamprecht1,2 1
Laboratory of Pharmaceutical Technology and Biopharmaceutics, University of Bonn, Bonn, Germany and 2Laboratory of Pharmaceutical Engineering, EA4267, University of Franche-Comte´, Besanc¸on, France Abstract
Keywords
A common technique for the preparation of polymeric nanoparticles (NPs) from preformed polymers is the emulsification solvent evaporation (ESE) method. However, the particle size of such carriers can typically not reduced below 100 nm. A bimodal distribution of particle size when applying ESE to the preparation of ethylcellulose (EC) NPs was intended to obtain very small particles in a size range below 50 nm. The proportion and size of the small particle fraction (SPF) depended on the surfactant as well as on the EC type and concentration. The preparation was conducted with different pharmaceutically relevant surfactants (polyoxyethylene (23) lauryl ether, sodium dodecyl sulfate, cetyltrimethylammonium bromide, polyvinyl alcohol and polysorbate 20) and all permitted obtaining very small NPs. After purification from excess surfactant by diafiltration and separation of the SPF by centrifugation, monodispersed particles with mean sizes between 20.6 ± 2.3 nm and 49.7 ± 4.8 nm could be isolated. The entrapment of a lipophilic model drug led to encapsulation rates between 34.0 ± 2.4% and 78.2 ± 12.6%, which were size and surfactant dependent. The preparation of polymeric NPs in a size below 50 nm by a simple centrifugation step holds promise for therapeutic applications where larger particles would be inefficient.
Emulsification solvent evaporation, indometacin, nanosphere, polymeric nanoparticle, surfactant
Introduction Nanoparticulate drug carrier systems are promising tools for diverse biopharmaceutical applications. The incorporation of active pharmaceutical ingredients into them and the choice of an adequate carrier system allow the modification of the biological and physicochemical behavior compared to the free drug. For example, the body distribution can be altered allowing more specific drug dosing, selective drug targeting and also intracellular drug delivery can be achieved1. Polymeric nanoparticles (NPs) are a common drug carrier type in this context and used for a wide range of the before-mentioned therapeutic applications. One crucial factor for the biological activity of nanocarriers is their size. On cellular level, the size determines the way of particle interaction, e.g. the adhesion to cell surface and the way of internalization. Possible ways of internalization are phagocytosis, clathrin-mediated endocytosis, caveolae-mediated endocytosis and macropinocytosis2. The mentioned pathways lead to different intracellular localizations of the internalized particles and show different size specificities. Phagocytosis and macropinocytosis exhibit minor size specificity, although a preference for larger particles is reported for the former3,4. Whereas clathrinmediated endocytosis and caveolae-mediated endocytosis have a size specificity for particles smaller than 200 nm and 500 nm, Address for correspondence: Alf Lamprecht, Laboratory of Pharmaceutical Technology and Biopharmaceutics, University of Bonn, Gerhard-Domagk-Str. 3, D-53121 Bonn, Germany. Tel: +49 228 73 52 43. Fax: +49 228 73 52 68. E-mail: [email protected]
History Received 28 February 2014 Revised 26 May 2014 Accepted 5 June 2014 Published online 7 July 2014
respectively5,6. Furthermore, it was shown that particles with diameters below 42 nm are internalized by a non-clathrin and noncaveolae-dependent pathway7. Thus, variation of particle size allows influencing particle–cell interactions. However, the standard techniques for the preparation of polymeric NPs from preformed polymers lead to sizes above approximately 100 nm. One of the main preparation methods for the incorporation of lipophilic drugs into the polymeric matrix of NPs is the oilin-water emulsification solvent evaporation (ESE) technique8. In this study, the lipophilic drug and the polymer are dissolved in an organic solvent, like methylene chloride or ethyl acetate. This organic solution is then emulsified in an external aqueous phase containing one or more surfactants with the intention to stabilize the emulsion interface. Removing the organic solvent by continuous stirring, increase of temperature or by reduced pressure eventually forms the drug-loaded NPs. ESE enables the preparation of NPs consisting of a wide variety of homo- and copolymers, such as biodegradable polymers like poly(lactic acid) (PLA) and poly(lactic-co-glycolic acid) (PLGA) or non-biodegradable polymers like ethylcellulose (EC) or ammonium methacrylate copolymers9. Similarly, the use of diverse surfactants for the preparation of such NPs has been reported in literature influencing the surface properties of the drug carriers10. The available range includes polymeric and non-polymeric as well as charged and non-charged surfactants. The choice leads to altered surface properties and biological activity of the nanospheres11, since a part of the surfactant remains on the particle surface even after extensive washing12.
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Chinese University of Hong Kong on 02/10/15 For personal use only.
2
P. Wachsmann & A. Lamprecht
The ESE technique allows a particle size reduction down to approximately 100 nm and diameters below that ‘‘threshold’’ are hardly found in the literature13–15. However, in several cases, a further size reduction appears desirable to increase targeting selectivity11 or to circumvent certain cellular mechanisms that may impede full delivery efficiency. In this study, ESE has been applied to the formulation of EC-NP with slight modifications. EC has served as model polymer due to its accessibility in different qualities. Its broad use for peroral and dermal applications is also well characterized. We found a bimodal size distribution of NPs when high surfactant concentrations were used and the small particle fraction (SPF) has a mean diameter of approximately 25–70 nm. The aim of this study was to analyze whether the production and isolation of small NPs (550 nm) was possible by efficient separation method and to what extent a variation of surfactant type or polymer properties and the respective concentrations would influence the NP size distribution and the SPF proportion.
Methods
Drug Dev Ind Pharm, Early Online: 1–7
was pumped (200 ml/min) through the filtration device, and the filtrate was replaced gradually with 1 l (Brij 35) or 500 ml (SDS, CTAB) distilled water. Thereafter, the suspension was concentrated to 10 ml and then diluted to 15.0 ml with distilled water. The monomodal distributed particles with diameters around 200 nm were purified by two centrifugation steps at 3840 g for 1 h (Brij 35) or 30 min (SDS, CTAB) and redispersed in distilled water to 10.0 ml. Isolation of SPF After purification from excess surfactant by diafiltration, the bimodal distributed particle suspension was centrifuged for 90 min with 8000g (1 g/100 ml Brij 35, 100 mg EC), 12 000 g (3 g/100 ml Brij 35, 300 mg EC) or 19 000 g (all other formulations) to separate the SPF from the large particle fraction (LPF). The lowest centrifugation force, which allowed the recovery of a monomodal distribution, was chosen. The supernatant was analyzed for its size distribution by photon correlation spectroscopy (PCS).
Materials EC (EthocelÕ standard 4 premium, EthocelÕ standard 10 premium and EthocelÕ standard 45 premium) was a gift from Colorcon (Dartford, England). Polyoxyethylene (23) lauryl ether (Brij 35), cetyltrimethylammonium bromide (CTAB) and indomethacin were purchased from Sigma-Aldrich (Steinheim, Germany). Sodium dodecyl sulfate (SDS) was bought from Carl Roth (Karlsruhe, Germany). Polyvinyl alcohol (PVA; Mowiol 4-88) was a gift from Kuraray (Frankfurt, Germany). Polysorbate 20 was bought from Caelo (Hilden, Germany). All other chemicals were of analytical grade. Particle preparation and purification The NPs were prepared by an oil-in-water ESE method; 30, 100 or 300 mg EC was dissolved in 5 ml ethylacetate. To produce drug-loaded NPs, 10 mg indometacin was added. The organic phase was poured into 10 ml aqueous surfactant solution. Different concentrations of polysorbate 20, Brij 35, PVA, SDS or CTAB were used as surfactant. After building an emulsion by ultrasonication for 4 min at 50 W (Sonoplus, Bandelin Electronic), the organic solvent was removed under reduced pressure. Excess surfactant was removed either by dialysis (SpectraPor Biotech MWCO 100 kDa, Spectrum Labs) or by diafiltration as follows: for the separation and encapsulation experiments, the NP suspension was diluted to 100 ml and purified by diafiltration (VivaFlow 50, Sartorius Sedim GmbH). The suspension
Particle size and zeta potential The NPs were analyzed for their size distribution by PCS (ZetaPlus, Brookhaven Instruments Corporation) at a fixed angle of 90 at 25 causing either the mode for lognormal or multimodal distributions. All results are calculated as volume distributions. When the mode for multimodal distributions was used, the distribution was either monomodal or bimodal. The size distribution was characterized by the proportion of the two fractions and their mean diameter. The mean diameter was calculated as frequency-weighted mean of the class interval middles, which were obtained from the particle size analysis software. Table 1. Properties of isolated small particle fraction (SPF) and 200 nm particles prepared with 100 mg EthocelÕ 4 (mean ± SD; n ¼ 3).
Surfactant Brij 35 Brij 35 CTAB CTAB SDS SDS
Concentration [g/100 ml] 3 0.035 0.3 0.04 1 0.004
Diameter [nm] 25.4 246.8 30.9 221.4 34.5 219.0
(±2.3) (±18.8) (±0.2) (±29.9) (±2.1) (±22.9)
Polydispersity index 0.223 0.029 0.166 0.083 0.196 0.053
(±0.024) (±0.042) (±0.005) (±0.010) (±0.063) (±0.022)
Zeta potential [mV] 11.4 14.6 22.0 4.8 34.5 22.1
(±4.3) (±0.6) (±5.2) (±2.3) (±1.7) (±4.6)
Figure 1. SEM images of ethylcellulose nanospheres prepared with 3 g/100 ml Brij 35 and 100 mg EthocelÕ 4 before (A) and after (B) centrifugation.
DOI: 10.3109/03639045.2014.935393
Zeta potential was determined with the same device. Small particles were concentrated with Amicon Ultra filtration devices (Molecular weight cut off 50 kDa, Merck Millipore), because signal intensity of non-concentrated samples was too low to obtain reliable results. Scanning electron microscopy Particles were imaged by Field Emission Scanning Electron Microscopy (FESEM). The nanosphere suspension was placed on a cover glass and dried overnight in a desiccator. The samples were examined by FESEM using a Hitachi SU8030 at 0.1–0.5 kV (Hitachi, Tokyo, Japan).
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Chinese University of Hong Kong on 02/10/15 For personal use only.
Drop shape analysis Drop shape analysis was performed to compare the different EC types in terms of interface activity; 100 mg EthocelÕ 4, EthocelÕ 10 or EthocelÕ 45 was dissolved in 10 ml methylene chloride. These solutions or pure methylene chloride were measured against distilled water. The interfacial tension was calculated from the drop shape using an EasyDropdevice (Kru¨ss GmbH). The interfacial tension was measured every 30 s for a period of 4 min, and the values were averaged. Ethyl acetate had to be substituted by methylene chloride due to instability of the ethyl acetate droplet with water as external phase. Encapsulation rate and drug release Ten milliliters of bimodal-distributed NPs suspension was centrifuged to isolate the SPF as described above; 4 ml centrifuged suspension was freeze-dried and weighed. The particles were dissolved in 10.0 ml methanol, and indomethacin was quantified with a spectrophotometer at 320 nm. Besides, 200 nm particles were resuspended in 10 ml distilled water, and 1 ml was freeze-dried and analyzed as described above. Purified and isolated SPFs and monomodal-distributed 200 nm particles were tested for their indomethacin release by a modified basket method16. Therefore, 4 ml (Brij 35, SDS) or 3 ml (CTAB) SPF were placed in the dissolution apparatus (PTW S, Pharmatest), and 25 ml pre-warmed release medium (phosphate buffer pH 6.8, 37 C) was added. The 200 nm particles were diluted with distilled water to an appropriate concentration.
Ethylcellulose nanoparticles with bimodal size distribution
3
Four milliliters of the dilution were placed in the dissolution apparatus. At predefined time points, 1 ml of release medium was withdrawn and replaced by fresh buffer. The amount of released indomethacin was determined by HPLC according to the following method: acetonitrile:phosphoric acid (0.2%) 60:40; RP-18 column; flow rate: 1.5 ml/min; UV absorption ¼ 320 nm17. Statistical analysis All results are expressed as mean ± standard deviation (SD). The results were statistically analyzed by analysis of variance (ANOVA) followed by the Students–Newman–Keuls post-test. SigmaStat for Windows 2.0 was used to calculate the p values. Differences were considered to be significant at p50.05.
Results Particle characterization The ESE preparation process for EC-NP led to their bimodal particle size distribution with higher concentrations of surfactant. The SPF was isolated from the LPF by centrifugation of the particle dispersion as was proven by SEM imaging and exemplarily shown for Brij35-NP before and after the centrifugation step (Figure 1). The former bimodal distribution of particle size was monomodal after the centrifugation. Lower surfactant concentrations allowed the production of monomodal-distributed particles, which were optimized for sizes around 200 nm. Size and zeta potential of the particles depended on surfactant type and concentration (Table 1). The absolute value of zeta potential increases for the small particles prepared with ionic surfactants. Influence of preparation parameters Increased surfactant concentrations led to decreasing particle sizes and increasing polydispersity indices (data not shown). Bimodal size distribution was observed above a surfactantspecific concentration for each formulation type. Mean diameters in the SPF varied between 31 nm and 73 nm, while the LPF lay between 98 nm and 617 nm (Figure 2). Increasing surfactant concentrations resulted in an increasing SPF proportion, while mean particle size of the SPF was slightly
Figure 2. Ethylcellulose nanospheres were prepared with 100 mg EthocelÕ 4 and different surfactant types and concentrations. Data of the distribution are shown as diameter (A) of the small particle fraction (SPF) and the large particle fraction (LPF) as well as proportion of the SPF (B) (mean ± SD; n ¼ 3).
4
P. Wachsmann & A. Lamprecht
Drug Dev Ind Pharm, Early Online: 1–7
decreased. However, above an optimal concentration, a higher surfactant concentration led generally to a decreasing SPF proportion and an increasing particle size. This effect was not visible with CTAB due to its limited solubility in water. The highest proportion of the SPF was reached with the nonionic surfactants polysorbate 20, Brij 35 and PVA with approximately 95% (Table 2). However, PVA led to comparatively large SPF
Table 2. Size of large particle fraction (LPF) and small particle fraction (SPF) and proportion of SPF after optimizing the surfactant concentration for formation of SPF (mean ± SD; n ¼ 3).
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Chinese University of Hong Kong on 02/10/15 For personal use only.
Surfactant Brij 35 Polysorbate 20 PVA CTAB SDS
Concentration [g/100 ml] 3 10 10 0.3 1
Diameter SPF [nm] 40.3 31.2 62.4 35.7 38.1
(±5.9) (±4.7) (±6.0) (±0.1) (±5.1)
Diameter LPF [nm] 173.1 150.1 194.8 124.5 97.7
(±10.5) (±2.7) (±19.7) (±6.4) (±3.4)
Proportion SPF [%] 93.3 95.7 94.7 85.7 83.3
(±0.6) (±1.5) (±1.5) (±6.4) (±3.8)
sizes. The ionic surfactants SDS and CTAB showed a SPF proportion of approximately 85%. When total surface areas were calculated based on the mean size distribution of SPF and LPF, a correlation was observed between surfactant concentration and total surface area when excluding concentrations above the SPF optimum where solution viscosity became a dominant factor. Even the regression of the complete data set suggested a correlation of total surface area and surfactant amount seemingly being independent from the surfactant type (Figure 3). Besides surfactant type and concentration, also the amount and molecular weight of EC confirmed to have an influence on the size distribution of EC-NP (Figure 4). The minimum surfactant concentration for the formation of the bimodal size distribution was a dependent variable of the EC amount. Furthermore, the proportion of the SPF in the optimum slightly differed depending on the EC concentration (30 mg EC: 95.3 ± 0.6%; 100 mg EC: 95.7 ± 1.5%; 300 mg EC: 86.7 ± 2.9%), whereas a statistically significant difference was obtained for 300 mg EC compared to the other two EC amounts. However, the size of the SPF was surprisingly unaffected (30 mg EC: 37.4 ± 3.5 nm; 100 mg EC: 31.2 ± 4.7 nm; 300 mg EC: 36.6 ± 3.3 nm). Similarly, EC of different viscosities influenced SPF proportion and minimum surfactant concentration for bimodal distribution significantly (Figure 5). The highest proportion of the SPF is achieved with a concentration of 81 mM polysorbate 20 for all EC types simultaneously. The mean NP diameter at this surfactant concentration varied, however, significantly with the molecular weight of EC (EthocelÕ 4: 31.2 ± 4.7 nm; EthocelÕ 10: 42.5 ± 6.4 nm; and EthocelÕ 45: 52.2 ± 1.4 nm). In order to determine the additional influence of surface-active properties of EC, drop shape analyses were performed with an EC-methylene chloride/water system. The interface tension lowering effect was less pronounced for EthocelÕ 45, while the other two types resulted in similar observations (Figure 6). Encapsulation rate and indomethacin release
Figure 3. Non-linear regression of surface area and surfactant amount. Surface area was calculated from the proportion of SPF and the mean sizes of SPF and LPF. y ¼ 16.8 ln(x) 23.1; R2 ¼ 0.683.
The concentration of surfactant was adjusted to obtain particles of comparable sizes around 200 nm after encapsulation of indomethacin. Therefore, the concentration of CTAB and SDS for blank particles and indomethacin-loaded particles differs. The encapsulation rates varied between 34.0% and 78.2% (Table 3). SDS and Brij 35 showed a comparable encapsulation
Figure 4. Ethylcellulose nanospheres were prepared with different amounts of EthocelÕ 4. The diameter (A) of the small particle fraction (SPF) and the large particle fraction (LPF) as well as proportion of the SPF (B) of ethylcellulose nanospheres are plotted against the surfactant concentration (mean ± SD; n ¼ 3).
Ethylcellulose nanoparticles with bimodal size distribution
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Chinese University of Hong Kong on 02/10/15 For personal use only.
DOI: 10.3109/03639045.2014.935393
5
Figure 5. Ethylcellulose nanospheres were prepared with different ethyl cellulose types (EthocelÕ 4, 10 and 45). The diameter (A) of the small particle fraction (SPF) and the large particle fraction (LPF) as well as proportion of the SPF (B) are plotted against the surfactant concentration (mean ± SD; n ¼ 3).
profiles with t50% values of 445 ± 82 min for CTAB, 489 ± 154 min for SDS and 503 ± 102 min for Brij 35. However, release kinetics of SPF were influenced by the surfactant type: SDS and Brij 35 released 50% of the encapsulated drug within 553 ± 88 min and 406 ± 184 min, respectively, while CTAB showed a significantly slower indomethacin release (1262 ± 53 min).
Discussion
Figure 6. Interfacial tension at the methylene chloride/water interface. The interfacial tension was analyzed by drop shape analysis using different ethylcellulose types (EthocelÕ 4, 10 and 45) (mean ± SD; n ¼ 3). Table 3. Hydrodynamic diameter, polydispersity index and encapsulation rate of indomethacin-loaded EC-NPs prepared with EthocelÕ 4.
Surfactant Brij 35 Brij 35 CTAB CTAB SDS SDS
Concentration [g/100ml] 3 0.035 0.3 0.045 1 0.003
Diameter [nm] 27.9 200.5 33.9 208.6 30.5 198.6
(±2.3) (±19.3) (±1.4) (±8.2) (±2.1) (±7.5)
Polydispersity Encapsulation index rate [%] 0.144 0.059 0.197 0.109 0.149 0.070
(±0.016) (±0.009) (±0.019) (±0.025) (±0.008) (±0.011)
34.0 80.0 78.2 96.7 41.1 56.1
(±2.4) (±1.6) (±12.6) (±5.0) (±2.7) (±5.5)
For each surfactant, the SPF was isolated, and monomodal-distributed 200 nm particles were prepared separately (mean ± SD; n ¼ 3).
rate, while EC-NP prepared with CTAB had nearly twice higher drug content. Indomethacin release of isolated SPF and monomodal-distributed 200 nm particles is shown in Figure 7. Two-hundred-nanometer particles exhibited comparable release
In the ESE method, the formulation of the nanoemulsion and removal of the organic solvent under reduced pressure can essentially influence the size distribution of NPs. Consequently, the formation of the SPF depends highly on energy input and amount of surfactant that is available for the surface stabilization18 by which the re-coalescence rate of newly formed small droplets can be reduced19. Thereafter, coalescence can also take place during evaporation of the organic phase. EC-NP prepared by ESE with ethyl acetate as solvent are formed due to coalescence of several unstable emulsion droplets, whereas PLA droplets show lower tendency to coalesce20. On the other hand, polystyrene NPs prepared by the same technique but with methylene chloride, chloroform or toluene as solvent show a very low coalescence rate during the evaporation21. So the extent of coalescence during evaporation depends on the solvent and the polymer properties. The size distribution of particles seems to depend mainly on an appropriate surfactant concentration, which hinders coalescence of emulsion droplets that may take place during the emulsion formation or the evaporation process or during both phases. The good correlation between surfactant concentration and surface area of the NPs throughout all surfactant types indicated that the molar concentration of the surfactant is far more determinant than the surfactant type. Besides the surfactant concentration, the polymer type and concentration also influenced the size distribution. An increasing EC concentration led to two different effects. The first effect was the co-stabilizing effect at the oil–water interface that was needed for the formation of a bimodal distribution and which EC contributed due to its surface-active properties22. A second effect was observed when the surfactant amount was able to stabilize the interface. Then, higher EC concentrations led to lower SPF
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Chinese University of Hong Kong on 02/10/15 For personal use only.
6
P. Wachsmann & A. Lamprecht
Drug Dev Ind Pharm, Early Online: 1–7
Figure 7. Release profile of monomodal-distributed 200 nm particles (A) and of isolated SPF (B). All particles were prepared with EthocelÕ 4 as polymer. Two-hundred nanometer particles were prepared with 0.035 g/100 ml Brij 35, 0.045 g/100 ml CTAB or 0.003 g/100 ml SDS. SPF was isolated from formulations prepared with 3 g/100 ml Brij 35, 0.3 g/100 ml CTAB or 1 g/100 ml SDS. The cumulated indomethacin release is plotted against the time (mean ± SD; n ¼ 3).
proportions with larger NP sizes, because the higher viscosity of the inner phase. Surprisingly, all variations of the preparation procedure led to a bimodal size distribution. A possible explanation is the inability of the surfactant-EC combination to fully stabilize the large water/ethyl acetate interface that is generated during the preparation. Thus, a part of the small droplets coalesces and forms the LPF, whereas the other part remains as the SPF. This is in accordance with the above-mentioned observation, that during the preparation of nanospheres using ESE, the formation of EC-NP results from coalescence of several unstable emulsion droplets20. Because we used higher concentrations of surfactant, an increased stabilization of small droplets was possible and therewith the production of small NPs. However, a part of droplets still coalesces and builds the LPF. The SPF that was formed under optimized conditions exhibited mean diameters down to approximately 25 nm, which is in the size range of other nano-sized drug carriers, like lipid nanocapsules (20–100 nm)23 or polymeric micelles (15–150 nm)24. Although the appearance of LPF cannot be removed completely, its ratio can be reduced down to a level of around vol-5% of the entire formulation; and moreover, the remaining amount of a few percent LPF can be separated by several techniques. Filtration was already described to isolate the SPF of bimodal distributed PLGA particles25. However, the major drawback of filtration in laboratory scale is caking, which leads to low particle recovery, whereas centrifugation seems to be more appropriate as separation procedure26. The lipophilic model drug indomethacin was successfully encapsulated into EC-NP with three different surfactants. Even after purification by diafiltration, an acceptable encapsulation rate was achieved. The appropriateness of this method for the purification of indometacin nanocapsules was already shown27. Indomethacin-loaded LPF or SPF prepared with SDS and Brij 35 exhibited fast drug release, which was on a comparable level. The SPF of CTAB particles showed a nearly twice higher drug content than the particles prepared with the anionic and the nonionic surfactant. During the preparation of 30 nm particles, a more than 6-fold higher amount of CTAB is used than for 200 nm particles, which leads to an increased zeta potential of the small particles. The molar ratio of CTAB to indomethacin increases
from 0.4 to 2.9. Subsequently, electrostatic interaction of the cationic surfactant and the anionic drug can take place leading to a higher drug load and a considerable delay of drug release.
Conclusion EC-NPs allow the preparation of drug-loaded NPs in the size range of 30 nm via the isolation of a SPF from a bimodal particle population. The optimal settings led to a process where only minimal amounts of LPF needed to be removed. The size and proportion of the SPF depended on the surfactant concentration but much less on its type as well as on the EC type and concentration.
Declaration of interest Alf Lamprecht is thankful for the financial support from the Institut Universitaire de France. This work was partially supported by a French Government grant managed by the French National Research Agency under the program ‘‘Investissementsd’Avenir’’ with reference ANR-11-LABX-0021.
References 1. Barratt G. Therapeutic applications of colloidal drug carriers. Pharm Sci Technol Today 2000;3:163–71. 2. Hillaireau H, Couvreur P. Nanocarriers’ entry into the cell: relevance to drug delivery. Cell Mol Life Sci 2009;66:2873–96. 3. Scha¨fer V, Briesen von H, Andreesen R, et al. Phagocytosis of nanoparticles by human immunodeficiency virus (HlV)-infected macrophages: a possibility for antiviral drug targeting. Pharm Res 1992;9:541–6. 4. Tabata Y, Ikada Y. Effect of the size and surface charge of polymer microspheres on their phagocytosis by macrophage. Biomaterials 1988;9:356–62. 5. Rejman J, Oberle V, Zuhorn I, Hoekstra D. Size-dependent internalization of particles via the pathways of clathrin-and caveolae-mediated endocytosis. Biochem J 2004;377:159–69. 6. Wang Z, Tiruppathi C, Minshall RD, Malik AB. Size and dynamics of caveolae studied using nanoparticles in living endothelial cells. ACS Nano 2009;3:4110–16. 7. Lai SK, Hida K, Man ST, et al. Privileged delivery of polymer nanoparticles to the perinuclear region of live cells via a non-clathrin, non-degradative pathway. Biomaterials 2007;28: 2876–84.
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Chinese University of Hong Kong on 02/10/15 For personal use only.
DOI: 10.3109/03639045.2014.935393
8. Pinto Reis C, Neufeld RJ, Ribeiro AJ, Veiga F. Nanoencapsulation I. Methods for preparation of drug-loaded polymeric nanoparticles. Nanomedicine 2006;2:8–21. 9. Letchford K, Burt H. A review of the formation and classification of amphiphilic block copolymer nanoparticulate structures: micelles, nanospheres, nanocapsules and polymersomes. Eur J Pharm Biopharm 2007;65:259–69. 10. Wachsmann P, Moulari B, Be´duneau A, et al. Surfactant-dependence of nanoparticle treatment in murine experimental colitis. J Control Release 2013;172:62–8. 11. Abdel-Mottaleb MMA, Moulari B, Be´duneau A, et al. Surfacecharge-dependent nanoparticles accumulation in inflamed skin. J Pharm Sci 2012;101:4231–9. 12. Boury F, Marchais H, Benoit JP, Proust JE. Surface characterization of poly(alpha-hydroxy acid) microspheres prepared by a solvent evaporation/extraction process. Biomaterials 1997;18:125–36. 13. Gaucher G, Asahina K, Wang J, Leroux J-C. Effect of poly(N-vinylpyrrolidone)-block-poly(D,L-lactide) as coating agent on the opsonization, phagocytosis, and pharmacokinetics of biodegradable nanoparticles. Biomacromolecules 2009;10:408–16. 14. Go´mez-Gaete C, Tsapis N, Besnard M, et al. Encapsulation of dexamethasone into biodegradable polymeric nanoparticles. Int J Pharm 2007;331:153–9. 15. Vauthier C, Bouchemal K. Methods for the preparation and manufacture of polymeric nanoparticles. Pharm Res 2009;26: 1025–58. 16. Abdel-Mottaleb MMA, Lamprecht A. Standardized in vitro drug release test for colloidal drug carriers using modified USP dissolution apparatus I. Drug Dev Ind Pharm 2011;37:178–84.
Ethylcellulose nanoparticles with bimodal size distribution
7
17. Limnell T, Heikkila¨ T, Santos HA, et al. Physicochemical stability of high indomethacin payload ordered mesoporous silica MCM-41 and SBA-15 microparticles. Int J Pharm 2011;416:242–51. 18. Anton N, Benoit J-P, Saulnier P. Design and production of nanoparticles formulated from nano-emulsion templates – a review. J Control Release 2008;128:185–99. 19. Jafari SM, Assadpoor E, He Y, Bhandari B. Re-coalescence of emulsion droplets during high-energy emulsification. Food Hydrocolloids 2008;22:1191–202. 20. Desgouilles S, Vauthier C, Bazile D, et al. The design of nanoparticles obtained by solvent evaporation: a comprehensive study. Langmuir 2003;19:9504–10. 21. Staff RH, Schaeffel D, Turshatov A, et al. Particle formation in the emulsion-solvent evaporation process. Small 2013;9:3514–22. 22. Melzer E, Kreuter J, Daniels R. Ethylcellulose: a new type of emulsion stabilizer. Eur J Pharm Biopharm 2003;56:23–7. 23. Huynh NT, Passirani C, Saulnier P, Benoit JP. Lipid nanocapsules: a new platform for nanomedicine. Int J Pharm 2009;379:201–9. 24. Jones M, Leroux J. Polymeric micelles – a new generation of colloidal drug carriers. Eur J Pharm Biopharm 1999;48:101–11. 25. Prabha S, Zhou W-Z, Panyam J, Labhasetwar V. Size-dependency of nanoparticle-mediated gene transfection: studies with fractionated nanoparticles. Int J Pharm 2002;244:105–15. 26. Gaumet M, Gurny R, Delie F. Fluorescent biodegradable PLGA particles with narrow size distributions: preparation by means of selective centrifugation. Int J Pharm 2007;342:222–30. 27. Limayem I, Charcosset C, Fessi H. Purification of nanoparticle suspensions by a concentration/diafiltration process. Sep Purif Technol 2004;38:1–9.
