- Email: [email protected]
Ion milling coupled field emission scanning electron microscopy reveals current misunderstanding of morphology of polymeric nanoparticles
European Journal of Pharmaceutics and Biopharmaceutics 89 (2015) 56–61
Contents lists available at ScienceDirect
European Journal of Pharmaceutics and Biopharmaceutics journal homepage: www.elsevier.com/locate/ejpb
Research paper
Ion milling coupled field emission scanning electron microscopy reveals current misunderstanding of morphology of polymeric nanoparticles Donny Francis a, Samiha Mouftah a, Robert Steffen b, Arnaud Beduneau c, Yann Pellequer c, Alf Lamprecht a,c,⇑ a b c
Department of Pharmaceutics and Biopharmaceutics, Institute of Pharmacy, University of Bonn, Bonn, Germany Nanotechnology Equipment Department, Hitachi High-Technologies Europe GmbH, Krefeld, Germany Laboratory of Pharmaceutical Engineering (EA4267), University of Franche-Comté, Besançon, France
a r t i c l e
i n f o
Article history: Received 24 August 2014 Accepted in revised form 12 November 2014 Available online 18 November 2014 Keywords: Ion milling Nanoparticles Nanoparticle morphology Double emulsion technique Protein encapsulation Polyester Protein adsorption
a b s t r a c t Nanoparticles (NPs) are currently used as drug delivery systems for numerous therapeutic macromolecules, e.g. proteins or DNA. Based on the preparation by double emulsion solvent evaporation a sponge-like structure was postulated entrapping hydrophilic drugs inside an internal aqueous phase. However, a direct proof of this hypothesized structure is still missing today. NPs were prepared from different polymers using a double-emulsion method and characterized for their physicochemical properties. Combining ion milling with field emission scanning electron microscopy allowed to cross section single NP and to visualize their internal morphology. The imaging procedure permitted cross-sectioning of NPs and visualization of the internal structure as well as localizing drugs associated with NPs. It was observed that none of the model actives was encapsulated inside the polymeric matrix when particle diameters were below around 470 nm but predominantly adsorbed to the particle surface. Even at larger diameters only a minority of particles of a diameter below 1 lm contained an internal phase. The properties of such drug loaded NPs, i.e. drug release or the observations in cellular uptake or even drug targeting needs to be interpreted carefully since in most cases NP surface properties are potentially dominated by the ‘encapsulated’ drug characteristics. Ó 2014 Elsevier B.V. All rights reserved.
1. Introduction Nanotechnology as a drug delivery platform offers very promising applications in drug delivery due to the small size of the carrier systems. Such drug carriers have been found to be promising tools in therapeutic approaches such as selective or targeted drug delivery toward a specific tissue or organ, enhanced drug transport across biological barriers or intracellular drug delivery which is interesting in gene and cancer therapy [1]. Besides the ability to enhance the transport across biological barriers [2], a major advantage of their use is their protective
Abbreviations: NPs, nanoparticles; PVA, polyvinyl alcohol; PLGA, poly (D,L-lactide-co-glycolide); SEM, scanning electron microscopy; FESEM, field emission scanning electron microscope; PAMM, poly(ethyl acrylate-co-methyl methacrylate-co-trimethylammonioethyl methacrylate chloride); BSA, bovine serum albumin; API, active pharmaceutical ingredient; IM, ion milling. ⇑ Corresponding author. Institute of Pharmacy, Department of Pharmaceutics and Biopharmaceutics, Gerhard-Domagk-Str. 3, 53121 Bonn, Germany. Tel.: +49 228 735243; fax: +49 228 735268. E-mail address: [email protected] (A. Lamprecht). http://dx.doi.org/10.1016/j.ejpb.2014.11.008 0939-6411/Ó 2014 Elsevier B.V. All rights reserved.
influence against drug degradation. This is of particular interest in the case of oral peptide and protein delivery where the ‘hostile’ luminal environment easily degrades the active. Nonetheless, also physiological conditions in other tissues can lead to premature drug metabolism [3,4]. An enormous therapeutic potential emanates from the possibility of an intracellular delivery via such nanocarriers which was also reported to lead to significant therapeutic progress in cancer therapy and vaccinations [5–7]. All of these quite varied therapeutic applications were found to be primarily dependent on the particle size and the surface properties of nanocarriers, which are decisive on therapeutic outcome [8,9]. Within the last decade, a substantial number of diverse nanoobjects have been designed and tested as novel drug delivery strategies for a broad range of therapeutic applications [7]. Among them polymeric, lipid-based and even inorganic carriers have been prepared [10] applying a multitude of different preparation techniques. For the choice of the nanocarrier constituents, a preference is given to auxiliary materials that have proven their unhazardous
D. Francis et al. / European Journal of Pharmaceutics and Biopharmaceutics 89 (2015) 56–61
properties. Specifically, biocompatible and biodegradable polyesters with approval from the health authorities are very often used for therapeutic applications [11]. Standard processes allow for the encapsulation of lipophilic low-molecular weight compounds into a multitude of particle matrix materials (lipids and polymers), though their applicability to water-soluble compounds such as peptides, proteins, and genes is limited due to the immediate and uncontrollable leakage of the entrapped active during the preparation process [4]. Therefore, one of the most currently used preparation method for the entrapment of hydrophilic drugs in nanoparticles (NPs) is the double emulsion solvent evaporation method shown to be associated with significant therapeutic progresses [4,6,12–15]. The method employs the theory that a double emulsion is trapping the therapeutically active cargo inside a polymeric shell leading to a subsequent encapsulation of it. Earlier, drug loaded microspheres have been developed by a similar emulsion method, and such a sponge-like internal structure was observed [16] leading to the assumption that the drug was encapsulated inside the nanoparticle matrix. Accordingly, a variety of conclusions have been drawn such as NPs’ protecting effects against harsh physiological influences such as enzymatic degradation or drug denaturation due to low pHs [17]. Furthermore, this structure has been suggested to be useful for controlled release of proteins and other pharmacologically relevant compounds [4,18]. However, current analytical procedures for quantification of encapsulated peptides and proteins do not differ between entrapped drug and adsorbed drug on the NP surface. Moreover, it remained hardly considered that in many cases an immediate and uncontrolled release of the entrapped drug has occurred, although a degradation of the particle matrix in such short delay was very unlikely. A few imaging approaches using transmission electron microscopy [12,19–22] and scanning electron microscopy (SEM) [13,15,17,20,21,23–25] have not been able to reveal reliable information on the internal structural peculiarities of NPs. Though SEM is used to determine the shape and to confirm the particle size of the NPs, the interior morphology remains undiscovered. Besides, these approaches have been so far unsuccessful to yield information about the localization of the incorporated active. In order to provide comprehensive insights in the internal particle morphology and the localization of encapsulated protein drugs, we set up an analytical method based on an ion milling system for sample preparation followed by particle imaging with a field emission scanning electron microscope (IM-FESEM). A variety of NPs were produced and their physicochemical properties were compared with the morphological peculiarities after cross-sectioning with an argon ion beam followed by electron microscopic imaging.
2. Materials and methods 2.1. Materials The copolymer poly(D,L-lactide-co-glycolide) (PLGA) with a 50/ 50 M ratio (ResomerÒ RG 502H) was kindly provided by Boehringer Ingelheim (Ingelheim, Germany). Poly(ethyl acrylate-co-methyl methacrylate-co-trimethylammonioethyl methacrylate chloride) (PAMM, EudragitÒ RS) was a kind gift from Evonik (Darmstadt, Germany). Gold labeled bovine serum albumin (Gold BSA) with a nominal diameter of 6 nm was purchased from AurionÒ (Wageningen, Netherlands). Polyvinyl alcohol (PVA) (Mowiol 4-88) was purchased from Sigma–Aldrich (Steinheim, Germany) and bovine serum albumin (BSA) by Carl Roth (Karlsruhe, Germany). All other chemicals were supplied by Sigma–Aldrich and Fisher Scientific (Loughborough, UK) and were of analytical grade.
57
2.2. Methods 2.2.1. Preparation and characterization of NPs BSA was employed as a model protein due to its ease of handling and strong interface stabilizing effect. BSA loaded NPs were prepared by using a water–oil–water (w/o/w) double emulsion technique [12,17]. Briefly, 100 ll of protein solution (BSA at 100 mg/ml in PBS) was homogenized with PLGA containing dichloromethane (100 mg in 3 ml) using a Sonopuls HD 2200 sonicator (Bandelin electronic, Germany) at 50 W for 1 min. The water in oil emulsion (w/o) was then added to 10 ml outer water phase containing various amounts of PVA. After ultrasonification for 1 min at 50 W the organic phase of the w/o/w-double emulsion was removed under reduced pressure. For the visualization of BSA location the inner phase was alternatively composed from 50 ll of BSA and 50 ll of gold-labeled BSA which were mixed prior to homogenization with the organic phase (Figs. 2B and 5A + B). Alternatively, heparin loaded nanoparticles were prepared as previously described [26] and modified as in the following: 0.1 ml heparin solution (1000 IU/ml) was emulsified in polyaminomethylmethacrylate (PAMM, EudragitÒ RS)/ethyl acetate solution (250 mg in 10 ml) at 50 W for 1 min. The resulted primary emulsion was then homogenized with 10 ml 0.1% w/v PVA using an ultrasound nozzle for 1 min at 50 W. Ethyl acetate was evaporated under reduced pressure. 2.2.2. Loading rate of protein loaded particles NPs were centrifuged for 20 min at 15,000g and the BSA loading was determined through the measurement of the free protein within the supernatant using a BCA Assay [27] (Roti-Quant universal assay, Carl Roth, Germany). Each sample was assayed in triplicates. 2.2.3. Particle size analysis Particle size and polydispersity (PDI) were determined by photon-correlation spectroscopy (PCS) using a ZetaPlus particle sizer (Brookhaven Instruments Corporation, UK) at a fixed angle of 90° at 25 °C and cumulants analysis. Each sample was measured in triplicates. 2.2.4. Release profile of protein loaded particles 1 ml of nanoparticle suspension was mixed with 9 ml of PBS and incubated in a shaking water bath at 37 °C. Samples were withdrawn at various times (0.5 h, 1 h, 2 h, 4 h and 24 h) for the analyses of drug release. The protein content was determined as described above. Each sample was measured in triplicates. 2.2.5. Morphology of protein loaded particles Particle suspensions were dried overnight on a bare silicon chip as sample preparation for argon ion milling cross sectioning, followed by low-voltage high-resolution field emission electron microscopy (FESEM). The silicon chip carrying the particle film was cleaved and mounted inside the ion milling system IM4000cryo (Hitachi, Japan) with the particle-coated side facing away from the ion source; that way the silicon chip acted as mechanical carrier with good heat conductivity to ensure minimum heat load by the impacting ion beam to the particles. Prior to start of the milling the sample stage was cooled down with liquid nitrogen to 100 °C. An argon ion beam (4 keV ion energy, 140 lA ion current) was applied to gently cut away a slice of about 40 lm from the cleaved chip edge, thus providing cross sections of particles present below the chip in this cross sectional cut plain. Since the glass transition temperature of PLGA is slightly above the physiological temperature of 37 °C [28], a low temperature is crucial to avoid particle collapses during the milling process. The sample was recovered from the ion milling system and
58
D. Francis et al. / European Journal of Pharmaceutics and Biopharmaceutics 89 (2015) 56–61
investigated for the internal particle morphology with a SU8030 FESEM (Hitachi, Japan) at 0.5–1 kV and low probe currents below 20 pA, using the in-column Super ExB secondary electron/lowangle backscattered electron detector. 14 batches of nanoparticles were cross-sectioned and in total 269 nanoparticles were analyzed. The determination of the particle diameters was performed with ImageJ2 (http://imagej.net). 3. Results and discussion The design of NPs has followed current preparation procedures for such protein loaded carriers. The protein was dissolved in an inner aqueous phase which is first emulsified in the water-immiscible organic solvent containing the particle matrix polymer which is again dispersed in an external aqueous phase. High-energy shear stress is necessary to provide emulsification at nanometric scale, which is usually exerted using ultrasound or high-pressure homogenization. The quantity of surfactants stabilizing the large external interface is of major importance and has a direct effect on the final particle size. Preparing particles by applying the standard polymers PLGA and PAMM with a double emulsion setup led to particles with mean diameters ranging from 80 nm to >5 lm. BSA was selected as a highly water soluble and strongly surface active model protein in order to ensure a significant contribution of the protein itself onto the stability of the inner water/oil interface [17]. Increased protein loading for smaller particles was observed with increasing surfactant amounts in the external phase (Fig. 1). A high BSA loading rate of 14.5 ± 0.6 lmol for PLGA particles was achieved using 0.5% w/v PVA as a surfactant, resulting in an eventual mean particle size of 320 nm (Table 1). Interestingly, the loading rate decreased for bigger particles, i.e. 5.6 ± 0.9 lmol loading of BSA for PLGA particles at a size of 693 nm. In spite of these contradictory observations, the hypothetic morphology of the submicron and nano-particles prepared by the double emulsion technique has not been elucidated by an appropriate imaging technique yet. As encapsulation rates were mainly determined by an indirect method (quantifying the non-encapsulated amount) usually the simple absence of the drug from the external phase was concluded confirmatory for the postulated particle structure [14,17]. Besides, the recovery of the encapsulated drug after particle degradation does not allow for distinguishing between protein amounts located on the surface and actual entrapment inside the particle. Surprisingly high drug encapsulation rates with polymeric particles in the size below 300 nm and even below 100 nm were reported [11,19,22]. When establishing IM-FESEM as an imaging technique for drug loaded nanocarriers, significant differences from the hypothesized particle structures were found. Cavities in the polymer matrix, representing the inner
PVA (g/100 ml)
Particle size (nm)
Loading rate (%)
0.01 0.03 0.1 0.5 1 3
1290.4 ± 124.9 1080.8 ± 28.8 648.0 ± 80.2 331.9 ± 35.6 291.3 ± 13.6 268.7 ± 3.8
37.6 ± 5.8 51.4 ± 1.2 53.0 ± 3.6 92.8 ± 5.5 93.6 ± 1.7 95.4 ± 0.1
water phase, could be visualized with IM-FESEM. PLGA or PAMM, used for the double emulsion step, led to surprisingly few particles with an inner aqueous phase (Figs. 2 and 3). Only 2.2% of all crosssections of PLGA particles and 7.6% of PAMM particles contained an inner water phase. This is far from explaining high encapsulation rates of up to 80%. Furthermore, image analyses revealed that the smallest PAMM particle still containing an inner aqueous phase was of around 470 nm while an equivalent PLGA particle had a size of at least 600 nm (Fig. 4). This difference observed between the polymer types is essentially related to the surface active properties of PAMM which provides higher interface stability in general. According to the Young–Laplace equation and the steep increase of the resulting pressure at the droplet interface for smaller droplets such instability is coherent with existing observation and explains the incapacity of the stabilization of the double emulsion at such small inner phase diameters. From precedent studies it is known that oil–water and water–oil interfaces differ in curvature pressure while effects are further enhanced by the limited
PLGA particles R2=0.963
15
protein binding [µmol]
Table 1 Protein loading and particle size of PLGA nanoparticles prepared with doubleemulsion method at different PVA concentrations.
10
5
0
50
100
150
200
total particle surface [cm2] Fig. 1. Protein loading of particles correlates with the total available particle surface area.
Fig. 2. (A + B) Cross sections obtained by IM-FESEM imaging show particles at different magnifications prepared by double emulsion method using PLGA.
D. Francis et al. / European Journal of Pharmaceutics and Biopharmaceutics 89 (2015) 56–61
59
presence of BSA on the surface of the particles (Fig. 5). As a consequence for the absence of an inner phase trapping the BSA, the entire drug amount associated with the particles is physically adsorbed on the surface of the polymeric particles. It should also be stated that the gold-BSA possess a highly increased mass compared to the unlabeled protein. This however, does not influence the structural arrangements of the interfaces of the nanoparticles as can be seen by the comparison of Figs. 2, 3 and 5. These findings are subsequently in line with earlier reports where smaller particles led to large drug ‘entrapment’ rates. Due to the larger total particle surface offered by smaller particles compared to that provided by bigger particles, more albumin is adsorbed onto such smaller particles, hence higher loading rates are found (Fig. 1). The release studies show that the protein is dissociated completely from the particles within 1 h (Fig. 6). Taking into account that polymer matrix degradation occurs within weeks [32], a protein release from inside of the particles is very improbable within this short period. This confirms strongly the observation of IMFESEM. Even the release from particle prepared with the lowest surfactant concentration to stabilize the oil–water interface during preparation allowing a more intimate contact between protein and particle surface releases the drug within 24 h, by far too short for a significant polymer degradation. We believe that this study gives a significant indication that the current structural understanding of drug loaded NPs prepared by double emulsion methods is not appropriate for most of the cases.
Fig. 3. (A + B) Cross sections obtained by IM-FESEM imaging show particles at different magnifications prepared by double emulsion method using PAMM.
particle diameter [µm]
PLGA0.1 PLGA0.5 PAMM
1
0.1
w/o int. phase
int. phase
Fig. 4. More than 300 particles were analyzed for their diameter and divided into two groups, particles containing an inner aqueous phase or not. Results show that no particle with a diameter below 400 nm is containing an inner aqueous phase droplet. PLGA0.1 represents PLGA-NP prepared with 0.1% PVA and PLGA0.5 represents PLGA-NP prepared with 0.5% PVA.
surfactant availability to stabilize the internal interface [29]. First indications for this instability of the inner water–oil phase were reported much earlier for microspheres, where in case of insufficient homogenization step of the inner aqueous phase most of the protein was attached to the particle surface [30,31]. Similar suggestions for smaller scale particles have been confirmed with the technique described here. Based on this observation, the question of drug location is surely of eminent interest. Subsequent studies, in which BSA was substituted with gold-BSA clearly elucidated the exclusive
Fig. 5. (A) Cross section obtained by ion milling and SEM imaging of gold-labeled BSA loaded PLGA particles prepared by double emulsion method. The gold label of the protein is clearly visible by the bright spots. (B) Gold-BSA can only be found within particles larger than 1 lm (left arrow) but the large majority of particles are empty (middle and right arrow) and in smaller particles (<400 nm) the protein is exclusively located on the particle surface.
60
D. Francis et al. / European Journal of Pharmaceutics and Biopharmaceutics 89 (2015) 56–61
protein release [%]
125 100 75 50
PVA 0.1% PVA 0.1% post PVA 0.5% PVA 0.5% post PVA 1.0% PVA 1.0% post
25 0 0
4
8
12
16
20
24
time [h] Fig. 6. Protein release kinetic obtained with different particle formulations having a nominal size of 299–693 nm underlines the assumption that the whole process is triggered by a desorption step and not by a diffusion control or polymer matrix degradation. Adding albumin to the inner aqueous phase (PVA X%) or at the end of the preparation process (PVA X%post) leads to similar kinetics, except in the case for particles in a size range of above 600 nm, where part of the albumin is encapsulated inside. Nanoparticle formulations were prepared with different PVA concentrations (0.1–1%).
PLGA needs to be seen as one of the standard polymers of the double emulsion technique and compared to alternative polyesters such as poly(lactic acid) or polycaprolactone it is much less hydrophobic. In order to ensure a maximum interface stabilizing effect by the polymer PAMM was used for NP formulation as an alternative. The effect of the polymer choice is clearly visible however still encapsulation of hydrophilic drugs in particles smaller than 400 nm can be seemingly excluded for the analyzed preparations obtained under standard preparation conditions of the doubleemulsion method. Concerning the interface stabilizing effect of the selected protein BSA is surely of primary choice. Therapeutic peptides and proteins usually possess much less interface stabilizing properties which will deteriorate the inner water/oil stability compared to the system presented here combining BSA with PLGA. Preliminary data from the ‘encapsulation’ of insulin and lysozyme indicate the absence of the inner aqueous phase in these cases (data not shown). Potentially, further proteins need to be tested at the long term, but the properties of the employed excipients in our study suggest that we deal here with a general phenomenon. Nanoparticles were cross-sectioned randomly which cannot ensure that images were taken at the sphere diameter only. Subsequently, diameters mentioned from image analyses can only be approximate data. However, this does not turn the observation invalid. It could lead only to the underestimation of the particle diameter. In consequence, this means that the minimum diameter of nanoparticles containing an inner phase could be even larger than it is estimated in this paper. Nonetheless, the IM-FESEM technique allowed for the first time to unambiguously confirm the absence of the inner aqueous phase. This discovery does not contest the observation on the therapeutic outcome of previous studies. It shows, however, the necessity to revisit mechanistic conclusions based on the initially hypothesized particle structure. The results of this study demonstrate that the protein load is mainly adsorbed to the particle surface of the nanocarrier. Consistently, the protein release is dominated by desorption kinetics which become essential for the drug availability in biological media, and not the diffusion of drugs through the polymeric matrix as currently mentioned in the literature [14]. As protein association with the particle surface plays the dominant role, the characterization of particle surface properties and the resulting interactions with the active compound needs
to be analyzed in much more detail than it is currently done in most of the studies. It is currently unclear to what extent the different forces influence the protein–particle interaction and to what magnitude each factor can modify this physisorptive link, as visibly the adsorptive binding is strong enough to ensure a drug transport into cells and to exercise an intracellular effect [6]. Another aspect also arises from the fact that surface properties of NPs are essentially dominated by the properties of the absorbed proteins. It needs to be studied in detail how the protein properties influence the biodistribution of the drug carrier, and how other factor such as toxicity are dominated or modified by the simple presence of the protein on the particle surface. First steps in this direction indicate the importance of this aspect [33]. Moreover, those can evolve according to complex adsorption and desorption pattern in their biological environment. A final issue is the currently complex preparation of such polymeric particles based on the double emulsion method, which limits the possibilities to transfer this method to an industrial scale due to its complexity and the therefrom resulting regulatory aspects requested by the health authorities worldwide. As drug load is essentially located on the surface of the particles manufacturing could be simplified by omitting complex steps within the current particle preparation methods, accelerating the development toward market entry of such nanocarrier products.
4. Conclusion The imaging technique based on IM-FESEM was able to visualize the internal morphology of drug loaded polymeric NPs prepared by a double emulsion method. It revealed that the current understanding of the internal structure of such particles needs to be revisited and suggests a further in-depth analysis of the conclusions drawn on the mechanisms of action and interaction of such drug carriers with the biological environment after administration.
Conflict of interest There is no conflict of interest statement. Acknowledgments Alf Lamprecht is thankful for funding in the framework of his membership at 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 ‘‘Investissements d’Avenir’’ with reference ANR-11-LABX-0021. Furthermore, we thank Dr. Martin Folger and Dr. Ragna Hoffmann from Boehringer Ingelheim Vetmedica GmbH for their support in this project. References [1] A. Lamprecht, IBD: selective nanoparticle adhesion can enhance colitis therapy, Nat. Rev. Gastroenterol. Hepatol. 7 (2010) 311–312. [2] J. Kreuter, Mechanism of polymeric nanoparticle-based drug transport across the blood–brain barrier (BBB), J. Microencapsul. 30 (2013) 49–54. [3] C. Damge, C. Reis, P. Maincent, Nanoparticle strategies for the oral delivery of insulin, Exp. Opin. Drug Deliv. 5 (2008) 45–68. [4] S.D. Putneyand, P.A. Burke, Improving protein therapeutics with sustainedrelease formulations, Nat. Biotechnol. 16 (1998) 153–157. [5] R.A. Petrosand, J.M. DeSimone, Strategies in the design of nanoparticles for therapeutic applications, Nat. Rev. Drug Discov. 9 (2010) 615–627. [6] S.P. Kasturi, I. Skountzou, R.A. Albrecht, D. Koutsonanos, T. Hua, H.I. Nakaya, R. Ravindran, S. Stewart, M. Alam, M. Kwissa, F. Villinger, N. Murthy, J. Steel, J. Jacob, R.J. Hogan, A. Garcia-Sastre, R. Compans, B. Pulendran, Programming the magnitude and persistence of antibody responses with innate immunity, Nature 470 (2011) 543–547.
D. Francis et al. / European Journal of Pharmaceutics and Biopharmaceutics 89 (2015) 56–61 [7] Z. Cheng, A. Al Zaki, J.Z. Hui, V.R. Muzykantov, A. Tsourkas, Multifunctional nanoparticles: cost versus benefit of adding targeting and imaging capabilities, Science 338 (2012) 903–910. [8] S. Hasani, Y. Pellequer, A. Lamprecht, Selective adhesion of nanoparticles to inflamed tissue in gastric ulcers, Pharm. Res. 26 (2009) 1149–1154. [9] S. McClean, E. Prosser, E. Meehan, D. O’Malley, N. Clarke, Z. Ramtoola, D. Brayden, Binding and uptake of biodegradable poly-DL-lactide micro- and nanoparticles in intestinal epithelia, Eur. J. Pharm. Sci. 6 (1998) 153–163. [10] T.A.P.F. Doll, S. Raman, R. Dey, P. Burkhard, Nanoscale assemblies and their biomedical applications, J. R. Soc. Interface 10 (2013) 80. [11] U. Bilati, E. Allemann, E. Doelker, Poly(D,L-lactide-co-glycolide) protein-loaded nanoparticles prepared by the double emulsion method – processing and formulation issues for enhanced entrapment efficiency, J. Microencapsul. 22 (2005) 205–214. [12] M.D. Blancoand, M.J. Alonso, Development and characterization of proteinloaded poly(lactide-co-glycolide) nanospheres, Eur. J. Pharm. Biopharm. 43 (1997) 287–294. [13] M.F. Zambaux, F. Bonneaux, R. Gref, P. Maincent, E. Dellacherie, M.J. Alonso, P. Labrude, C. Vigneron, Influence of experimental parameters on the characteristics of poly(lactic acid) nanoparticles prepared by a double emulsion method, J. Control. Release 50 (1998) 31–40. [14] A. Lamprecht, N. Ubrich, M. Hombreiro Perez, C. Lehr, M. Hoffman, P. Maincent, Biodegradable monodispersed nanoparticles prepared by pressure homogenization–emulsification, Int. J. Pharm. 184 (1999) 97–105. [15] B.A. Pfeifer, J.A. Burdick, R. Langer, Formulation and surface modification of poly(ester-anhydride) micro- and nanospheres, Biomaterials 26 (2005) 117– 124. [16] A. Lamprecht, H. Rodero Torres, U. Schäfer, C.-M. Lehr, Biodegradable microparticles as a two-drug controlled release formulation: a potential treatment of inflammatory bowel disease, J. Control. Release 69 (2000) 445– 454. [17] A. Lamprecht, N. Ubrich, M. Hombreiro Perez, C. Lehr, M. Hoffman, P. Maincent, Influences of process parameters on nanoparticle preparation performed by a double emulsion pressure homogenization technique, Int. J. Pharm. 196 (2000) 177–182. [18] R.C. Mundargi, V.R. Babu, V. Rangaswamy, P. Patel, T.M. Aminabhavi, Nano/ micro technologies for delivering macromolecular therapeutics using poly(D,Llactide-co-glycolide) and its derivatives, J. Control. Release 125 (2008) 193– 209. [19] E. Cohen-Sela, M. Chorny, N. Koroukhov, H.D. Danenberg, G. Golomb, A new double emulsion solvent diffusion technique for encapsulating hydrophilic molecules in PLGA nanoparticles, J. Control. Release 133 (2009) 90–95. [20] S. Khoee, A. Sattari, F. Atyabi, Physico-chemical properties investigation of cisplatin loaded polybutyladipate (PBA) nanoparticles prepared by w/o/w, Mater. Sci. Eng. C 32 (2012) 1078–1086.
61
[21] S. Khoeeand, M. Yaghoobian, An investigation into the role of surfactants in controlling particle size of polymeric nanocapsules containing penicillin-G in double emulsion, Eur. J. Med. Chem. 44 (2009) 2392–2399. [22] Q. He, J. Liu, X. Sun, Z.R. Zhang, Preparation and characteristics of DNAnanoparticles targeting to hepatocarcinoma cells, World J. Gastroenterol. 10 (2004) 660–663. [23] C. Giovino, I. Ayensu, J. Tetteh, J.S. Boateng, Development and characterisation of chitosan films impregnated with insulin loaded PEG-b-PLA nanoparticles (NPs): a potential approach for buccal delivery of macromolecules, Int. J. Pharm. 428 (2012) 143–151. [24] I. Gutierro, R.M. Hernandez, M. Igartua, A.R. Gascon, J.L. Pedraz, Size dependent immune response after subcutaneous, oral and intranasal administration of BSA loaded nanospheres, Vaccine 21 (2002) 67–77. [25] B. Sahana, K. Santra, S. Basu, B. Mukherjee, Development of biodegradable polymer based tamoxifen citrate loaded nanoparticles and effect of some manufacturing process parameters on them: a physicochemical and in-vitro evaluation, Int. J. Nanomed. 5 (2010) 621–630. [26] V. Hoffart, A. Lamprecht, P. Maincent, T. Lecompte, C. Vigneron, N. Ubrich, Oral bioavailability of a low molecular weight heparin using a polymeric delivery system, J. Control. Release 113 (2006) 38–42. [27] P.K. Smith, R.I. Krohn, G.T. Hermanson, A.K. Mallia, F.H. Gartner, M.D. Provenzano, E.K. Fujimoto, N.M. Goeke, B.J. Olson, D.C. Klenk, Measurement of protein using bicinchoninic acid, Anal. Biochem. 150 (1985) 76–85. [28] R.A. Jain, The manufacturing techniques of various drug loaded biodegradable poly(lactide-co-glycolide) (PLGA) devices, Biomaterials 21 (2000) 2475–2490. [29] A.J. Castellanos, S.J. Toro-Mendoza, M. Garcia-Sucre, Correction to the interfacial tension by curvature radius: differences between droplets and bubbles, J. Phys. Chem. B 113 (2009) 5891–5896. [30] C. Wischke, H.H. Borchert, Influence of the primary emulsification procedure on the characteristics of small protein-loaded PLGA microparticles for antigen delivery, J. Microencapsul. 23 (2006) 435–448. [31] S. Schuster, R. Bernewitz, G. Guthausen, J. Zapp, A.M. Greiner, K. Köhler, H.P. Schuchmann, Analysis of W1/O/W2 double emulsions with CLSM: statistical image processing for droplet size distribution, Chem. Eng. Sci. 81 (2012) 84– 90. [32] K. Mäder, B. Bittner, Y. Li, W. Wohlauf, T. Kissel, Monitoring microviscosity and microacidity of the albumin microenvironment inside degrading microparticles from poly(lactide-co-glycolide) (PLG) or ABA-triblock polymers containing hydrophobic poly(lactide-co-glycolide) A blocks and hydrophilic poly(ethyleneoxide) B blocks, Pharm. Res. 15 (1998) 787–793. [33] M. Arruebo, M. Valladares, A. Gonzalez-Fernandez, Antibody-conjugated nanoparticles for biomedical applications, J. Nanomater. (2009) 1–24 (Article ID 439389).
Contents lists available at ScienceDirect
European Journal of Pharmaceutics and Biopharmaceutics journal homepage: www.elsevier.com/locate/ejpb
Research paper
Ion milling coupled field emission scanning electron microscopy reveals current misunderstanding of morphology of polymeric nanoparticles Donny Francis a, Samiha Mouftah a, Robert Steffen b, Arnaud Beduneau c, Yann Pellequer c, Alf Lamprecht a,c,⇑ a b c
Department of Pharmaceutics and Biopharmaceutics, Institute of Pharmacy, University of Bonn, Bonn, Germany Nanotechnology Equipment Department, Hitachi High-Technologies Europe GmbH, Krefeld, Germany Laboratory of Pharmaceutical Engineering (EA4267), University of Franche-Comté, Besançon, France
a r t i c l e
i n f o
Article history: Received 24 August 2014 Accepted in revised form 12 November 2014 Available online 18 November 2014 Keywords: Ion milling Nanoparticles Nanoparticle morphology Double emulsion technique Protein encapsulation Polyester Protein adsorption
a b s t r a c t Nanoparticles (NPs) are currently used as drug delivery systems for numerous therapeutic macromolecules, e.g. proteins or DNA. Based on the preparation by double emulsion solvent evaporation a sponge-like structure was postulated entrapping hydrophilic drugs inside an internal aqueous phase. However, a direct proof of this hypothesized structure is still missing today. NPs were prepared from different polymers using a double-emulsion method and characterized for their physicochemical properties. Combining ion milling with field emission scanning electron microscopy allowed to cross section single NP and to visualize their internal morphology. The imaging procedure permitted cross-sectioning of NPs and visualization of the internal structure as well as localizing drugs associated with NPs. It was observed that none of the model actives was encapsulated inside the polymeric matrix when particle diameters were below around 470 nm but predominantly adsorbed to the particle surface. Even at larger diameters only a minority of particles of a diameter below 1 lm contained an internal phase. The properties of such drug loaded NPs, i.e. drug release or the observations in cellular uptake or even drug targeting needs to be interpreted carefully since in most cases NP surface properties are potentially dominated by the ‘encapsulated’ drug characteristics. Ó 2014 Elsevier B.V. All rights reserved.
1. Introduction Nanotechnology as a drug delivery platform offers very promising applications in drug delivery due to the small size of the carrier systems. Such drug carriers have been found to be promising tools in therapeutic approaches such as selective or targeted drug delivery toward a specific tissue or organ, enhanced drug transport across biological barriers or intracellular drug delivery which is interesting in gene and cancer therapy [1]. Besides the ability to enhance the transport across biological barriers [2], a major advantage of their use is their protective
Abbreviations: NPs, nanoparticles; PVA, polyvinyl alcohol; PLGA, poly (D,L-lactide-co-glycolide); SEM, scanning electron microscopy; FESEM, field emission scanning electron microscope; PAMM, poly(ethyl acrylate-co-methyl methacrylate-co-trimethylammonioethyl methacrylate chloride); BSA, bovine serum albumin; API, active pharmaceutical ingredient; IM, ion milling. ⇑ Corresponding author. Institute of Pharmacy, Department of Pharmaceutics and Biopharmaceutics, Gerhard-Domagk-Str. 3, 53121 Bonn, Germany. Tel.: +49 228 735243; fax: +49 228 735268. E-mail address: [email protected] (A. Lamprecht). http://dx.doi.org/10.1016/j.ejpb.2014.11.008 0939-6411/Ó 2014 Elsevier B.V. All rights reserved.
influence against drug degradation. This is of particular interest in the case of oral peptide and protein delivery where the ‘hostile’ luminal environment easily degrades the active. Nonetheless, also physiological conditions in other tissues can lead to premature drug metabolism [3,4]. An enormous therapeutic potential emanates from the possibility of an intracellular delivery via such nanocarriers which was also reported to lead to significant therapeutic progress in cancer therapy and vaccinations [5–7]. All of these quite varied therapeutic applications were found to be primarily dependent on the particle size and the surface properties of nanocarriers, which are decisive on therapeutic outcome [8,9]. Within the last decade, a substantial number of diverse nanoobjects have been designed and tested as novel drug delivery strategies for a broad range of therapeutic applications [7]. Among them polymeric, lipid-based and even inorganic carriers have been prepared [10] applying a multitude of different preparation techniques. For the choice of the nanocarrier constituents, a preference is given to auxiliary materials that have proven their unhazardous
D. Francis et al. / European Journal of Pharmaceutics and Biopharmaceutics 89 (2015) 56–61
properties. Specifically, biocompatible and biodegradable polyesters with approval from the health authorities are very often used for therapeutic applications [11]. Standard processes allow for the encapsulation of lipophilic low-molecular weight compounds into a multitude of particle matrix materials (lipids and polymers), though their applicability to water-soluble compounds such as peptides, proteins, and genes is limited due to the immediate and uncontrollable leakage of the entrapped active during the preparation process [4]. Therefore, one of the most currently used preparation method for the entrapment of hydrophilic drugs in nanoparticles (NPs) is the double emulsion solvent evaporation method shown to be associated with significant therapeutic progresses [4,6,12–15]. The method employs the theory that a double emulsion is trapping the therapeutically active cargo inside a polymeric shell leading to a subsequent encapsulation of it. Earlier, drug loaded microspheres have been developed by a similar emulsion method, and such a sponge-like internal structure was observed [16] leading to the assumption that the drug was encapsulated inside the nanoparticle matrix. Accordingly, a variety of conclusions have been drawn such as NPs’ protecting effects against harsh physiological influences such as enzymatic degradation or drug denaturation due to low pHs [17]. Furthermore, this structure has been suggested to be useful for controlled release of proteins and other pharmacologically relevant compounds [4,18]. However, current analytical procedures for quantification of encapsulated peptides and proteins do not differ between entrapped drug and adsorbed drug on the NP surface. Moreover, it remained hardly considered that in many cases an immediate and uncontrolled release of the entrapped drug has occurred, although a degradation of the particle matrix in such short delay was very unlikely. A few imaging approaches using transmission electron microscopy [12,19–22] and scanning electron microscopy (SEM) [13,15,17,20,21,23–25] have not been able to reveal reliable information on the internal structural peculiarities of NPs. Though SEM is used to determine the shape and to confirm the particle size of the NPs, the interior morphology remains undiscovered. Besides, these approaches have been so far unsuccessful to yield information about the localization of the incorporated active. In order to provide comprehensive insights in the internal particle morphology and the localization of encapsulated protein drugs, we set up an analytical method based on an ion milling system for sample preparation followed by particle imaging with a field emission scanning electron microscope (IM-FESEM). A variety of NPs were produced and their physicochemical properties were compared with the morphological peculiarities after cross-sectioning with an argon ion beam followed by electron microscopic imaging.
2. Materials and methods 2.1. Materials The copolymer poly(D,L-lactide-co-glycolide) (PLGA) with a 50/ 50 M ratio (ResomerÒ RG 502H) was kindly provided by Boehringer Ingelheim (Ingelheim, Germany). Poly(ethyl acrylate-co-methyl methacrylate-co-trimethylammonioethyl methacrylate chloride) (PAMM, EudragitÒ RS) was a kind gift from Evonik (Darmstadt, Germany). Gold labeled bovine serum albumin (Gold BSA) with a nominal diameter of 6 nm was purchased from AurionÒ (Wageningen, Netherlands). Polyvinyl alcohol (PVA) (Mowiol 4-88) was purchased from Sigma–Aldrich (Steinheim, Germany) and bovine serum albumin (BSA) by Carl Roth (Karlsruhe, Germany). All other chemicals were supplied by Sigma–Aldrich and Fisher Scientific (Loughborough, UK) and were of analytical grade.
57
2.2. Methods 2.2.1. Preparation and characterization of NPs BSA was employed as a model protein due to its ease of handling and strong interface stabilizing effect. BSA loaded NPs were prepared by using a water–oil–water (w/o/w) double emulsion technique [12,17]. Briefly, 100 ll of protein solution (BSA at 100 mg/ml in PBS) was homogenized with PLGA containing dichloromethane (100 mg in 3 ml) using a Sonopuls HD 2200 sonicator (Bandelin electronic, Germany) at 50 W for 1 min. The water in oil emulsion (w/o) was then added to 10 ml outer water phase containing various amounts of PVA. After ultrasonification for 1 min at 50 W the organic phase of the w/o/w-double emulsion was removed under reduced pressure. For the visualization of BSA location the inner phase was alternatively composed from 50 ll of BSA and 50 ll of gold-labeled BSA which were mixed prior to homogenization with the organic phase (Figs. 2B and 5A + B). Alternatively, heparin loaded nanoparticles were prepared as previously described [26] and modified as in the following: 0.1 ml heparin solution (1000 IU/ml) was emulsified in polyaminomethylmethacrylate (PAMM, EudragitÒ RS)/ethyl acetate solution (250 mg in 10 ml) at 50 W for 1 min. The resulted primary emulsion was then homogenized with 10 ml 0.1% w/v PVA using an ultrasound nozzle for 1 min at 50 W. Ethyl acetate was evaporated under reduced pressure. 2.2.2. Loading rate of protein loaded particles NPs were centrifuged for 20 min at 15,000g and the BSA loading was determined through the measurement of the free protein within the supernatant using a BCA Assay [27] (Roti-Quant universal assay, Carl Roth, Germany). Each sample was assayed in triplicates. 2.2.3. Particle size analysis Particle size and polydispersity (PDI) were determined by photon-correlation spectroscopy (PCS) using a ZetaPlus particle sizer (Brookhaven Instruments Corporation, UK) at a fixed angle of 90° at 25 °C and cumulants analysis. Each sample was measured in triplicates. 2.2.4. Release profile of protein loaded particles 1 ml of nanoparticle suspension was mixed with 9 ml of PBS and incubated in a shaking water bath at 37 °C. Samples were withdrawn at various times (0.5 h, 1 h, 2 h, 4 h and 24 h) for the analyses of drug release. The protein content was determined as described above. Each sample was measured in triplicates. 2.2.5. Morphology of protein loaded particles Particle suspensions were dried overnight on a bare silicon chip as sample preparation for argon ion milling cross sectioning, followed by low-voltage high-resolution field emission electron microscopy (FESEM). The silicon chip carrying the particle film was cleaved and mounted inside the ion milling system IM4000cryo (Hitachi, Japan) with the particle-coated side facing away from the ion source; that way the silicon chip acted as mechanical carrier with good heat conductivity to ensure minimum heat load by the impacting ion beam to the particles. Prior to start of the milling the sample stage was cooled down with liquid nitrogen to 100 °C. An argon ion beam (4 keV ion energy, 140 lA ion current) was applied to gently cut away a slice of about 40 lm from the cleaved chip edge, thus providing cross sections of particles present below the chip in this cross sectional cut plain. Since the glass transition temperature of PLGA is slightly above the physiological temperature of 37 °C [28], a low temperature is crucial to avoid particle collapses during the milling process. The sample was recovered from the ion milling system and
58
D. Francis et al. / European Journal of Pharmaceutics and Biopharmaceutics 89 (2015) 56–61
investigated for the internal particle morphology with a SU8030 FESEM (Hitachi, Japan) at 0.5–1 kV and low probe currents below 20 pA, using the in-column Super ExB secondary electron/lowangle backscattered electron detector. 14 batches of nanoparticles were cross-sectioned and in total 269 nanoparticles were analyzed. The determination of the particle diameters was performed with ImageJ2 (http://imagej.net). 3. Results and discussion The design of NPs has followed current preparation procedures for such protein loaded carriers. The protein was dissolved in an inner aqueous phase which is first emulsified in the water-immiscible organic solvent containing the particle matrix polymer which is again dispersed in an external aqueous phase. High-energy shear stress is necessary to provide emulsification at nanometric scale, which is usually exerted using ultrasound or high-pressure homogenization. The quantity of surfactants stabilizing the large external interface is of major importance and has a direct effect on the final particle size. Preparing particles by applying the standard polymers PLGA and PAMM with a double emulsion setup led to particles with mean diameters ranging from 80 nm to >5 lm. BSA was selected as a highly water soluble and strongly surface active model protein in order to ensure a significant contribution of the protein itself onto the stability of the inner water/oil interface [17]. Increased protein loading for smaller particles was observed with increasing surfactant amounts in the external phase (Fig. 1). A high BSA loading rate of 14.5 ± 0.6 lmol for PLGA particles was achieved using 0.5% w/v PVA as a surfactant, resulting in an eventual mean particle size of 320 nm (Table 1). Interestingly, the loading rate decreased for bigger particles, i.e. 5.6 ± 0.9 lmol loading of BSA for PLGA particles at a size of 693 nm. In spite of these contradictory observations, the hypothetic morphology of the submicron and nano-particles prepared by the double emulsion technique has not been elucidated by an appropriate imaging technique yet. As encapsulation rates were mainly determined by an indirect method (quantifying the non-encapsulated amount) usually the simple absence of the drug from the external phase was concluded confirmatory for the postulated particle structure [14,17]. Besides, the recovery of the encapsulated drug after particle degradation does not allow for distinguishing between protein amounts located on the surface and actual entrapment inside the particle. Surprisingly high drug encapsulation rates with polymeric particles in the size below 300 nm and even below 100 nm were reported [11,19,22]. When establishing IM-FESEM as an imaging technique for drug loaded nanocarriers, significant differences from the hypothesized particle structures were found. Cavities in the polymer matrix, representing the inner
PVA (g/100 ml)
Particle size (nm)
Loading rate (%)
0.01 0.03 0.1 0.5 1 3
1290.4 ± 124.9 1080.8 ± 28.8 648.0 ± 80.2 331.9 ± 35.6 291.3 ± 13.6 268.7 ± 3.8
37.6 ± 5.8 51.4 ± 1.2 53.0 ± 3.6 92.8 ± 5.5 93.6 ± 1.7 95.4 ± 0.1
water phase, could be visualized with IM-FESEM. PLGA or PAMM, used for the double emulsion step, led to surprisingly few particles with an inner aqueous phase (Figs. 2 and 3). Only 2.2% of all crosssections of PLGA particles and 7.6% of PAMM particles contained an inner water phase. This is far from explaining high encapsulation rates of up to 80%. Furthermore, image analyses revealed that the smallest PAMM particle still containing an inner aqueous phase was of around 470 nm while an equivalent PLGA particle had a size of at least 600 nm (Fig. 4). This difference observed between the polymer types is essentially related to the surface active properties of PAMM which provides higher interface stability in general. According to the Young–Laplace equation and the steep increase of the resulting pressure at the droplet interface for smaller droplets such instability is coherent with existing observation and explains the incapacity of the stabilization of the double emulsion at such small inner phase diameters. From precedent studies it is known that oil–water and water–oil interfaces differ in curvature pressure while effects are further enhanced by the limited
PLGA particles R2=0.963
15
protein binding [µmol]
Table 1 Protein loading and particle size of PLGA nanoparticles prepared with doubleemulsion method at different PVA concentrations.
10
5
0
50
100
150
200
total particle surface [cm2] Fig. 1. Protein loading of particles correlates with the total available particle surface area.
Fig. 2. (A + B) Cross sections obtained by IM-FESEM imaging show particles at different magnifications prepared by double emulsion method using PLGA.
D. Francis et al. / European Journal of Pharmaceutics and Biopharmaceutics 89 (2015) 56–61
59
presence of BSA on the surface of the particles (Fig. 5). As a consequence for the absence of an inner phase trapping the BSA, the entire drug amount associated with the particles is physically adsorbed on the surface of the polymeric particles. It should also be stated that the gold-BSA possess a highly increased mass compared to the unlabeled protein. This however, does not influence the structural arrangements of the interfaces of the nanoparticles as can be seen by the comparison of Figs. 2, 3 and 5. These findings are subsequently in line with earlier reports where smaller particles led to large drug ‘entrapment’ rates. Due to the larger total particle surface offered by smaller particles compared to that provided by bigger particles, more albumin is adsorbed onto such smaller particles, hence higher loading rates are found (Fig. 1). The release studies show that the protein is dissociated completely from the particles within 1 h (Fig. 6). Taking into account that polymer matrix degradation occurs within weeks [32], a protein release from inside of the particles is very improbable within this short period. This confirms strongly the observation of IMFESEM. Even the release from particle prepared with the lowest surfactant concentration to stabilize the oil–water interface during preparation allowing a more intimate contact between protein and particle surface releases the drug within 24 h, by far too short for a significant polymer degradation. We believe that this study gives a significant indication that the current structural understanding of drug loaded NPs prepared by double emulsion methods is not appropriate for most of the cases.
Fig. 3. (A + B) Cross sections obtained by IM-FESEM imaging show particles at different magnifications prepared by double emulsion method using PAMM.
particle diameter [µm]
PLGA0.1 PLGA0.5 PAMM
1
0.1
w/o int. phase
int. phase
Fig. 4. More than 300 particles were analyzed for their diameter and divided into two groups, particles containing an inner aqueous phase or not. Results show that no particle with a diameter below 400 nm is containing an inner aqueous phase droplet. PLGA0.1 represents PLGA-NP prepared with 0.1% PVA and PLGA0.5 represents PLGA-NP prepared with 0.5% PVA.
surfactant availability to stabilize the internal interface [29]. First indications for this instability of the inner water–oil phase were reported much earlier for microspheres, where in case of insufficient homogenization step of the inner aqueous phase most of the protein was attached to the particle surface [30,31]. Similar suggestions for smaller scale particles have been confirmed with the technique described here. Based on this observation, the question of drug location is surely of eminent interest. Subsequent studies, in which BSA was substituted with gold-BSA clearly elucidated the exclusive
Fig. 5. (A) Cross section obtained by ion milling and SEM imaging of gold-labeled BSA loaded PLGA particles prepared by double emulsion method. The gold label of the protein is clearly visible by the bright spots. (B) Gold-BSA can only be found within particles larger than 1 lm (left arrow) but the large majority of particles are empty (middle and right arrow) and in smaller particles (<400 nm) the protein is exclusively located on the particle surface.
60
D. Francis et al. / European Journal of Pharmaceutics and Biopharmaceutics 89 (2015) 56–61
protein release [%]
125 100 75 50
PVA 0.1% PVA 0.1% post PVA 0.5% PVA 0.5% post PVA 1.0% PVA 1.0% post
25 0 0
4
8
12
16
20
24
time [h] Fig. 6. Protein release kinetic obtained with different particle formulations having a nominal size of 299–693 nm underlines the assumption that the whole process is triggered by a desorption step and not by a diffusion control or polymer matrix degradation. Adding albumin to the inner aqueous phase (PVA X%) or at the end of the preparation process (PVA X%post) leads to similar kinetics, except in the case for particles in a size range of above 600 nm, where part of the albumin is encapsulated inside. Nanoparticle formulations were prepared with different PVA concentrations (0.1–1%).
PLGA needs to be seen as one of the standard polymers of the double emulsion technique and compared to alternative polyesters such as poly(lactic acid) or polycaprolactone it is much less hydrophobic. In order to ensure a maximum interface stabilizing effect by the polymer PAMM was used for NP formulation as an alternative. The effect of the polymer choice is clearly visible however still encapsulation of hydrophilic drugs in particles smaller than 400 nm can be seemingly excluded for the analyzed preparations obtained under standard preparation conditions of the doubleemulsion method. Concerning the interface stabilizing effect of the selected protein BSA is surely of primary choice. Therapeutic peptides and proteins usually possess much less interface stabilizing properties which will deteriorate the inner water/oil stability compared to the system presented here combining BSA with PLGA. Preliminary data from the ‘encapsulation’ of insulin and lysozyme indicate the absence of the inner aqueous phase in these cases (data not shown). Potentially, further proteins need to be tested at the long term, but the properties of the employed excipients in our study suggest that we deal here with a general phenomenon. Nanoparticles were cross-sectioned randomly which cannot ensure that images were taken at the sphere diameter only. Subsequently, diameters mentioned from image analyses can only be approximate data. However, this does not turn the observation invalid. It could lead only to the underestimation of the particle diameter. In consequence, this means that the minimum diameter of nanoparticles containing an inner phase could be even larger than it is estimated in this paper. Nonetheless, the IM-FESEM technique allowed for the first time to unambiguously confirm the absence of the inner aqueous phase. This discovery does not contest the observation on the therapeutic outcome of previous studies. It shows, however, the necessity to revisit mechanistic conclusions based on the initially hypothesized particle structure. The results of this study demonstrate that the protein load is mainly adsorbed to the particle surface of the nanocarrier. Consistently, the protein release is dominated by desorption kinetics which become essential for the drug availability in biological media, and not the diffusion of drugs through the polymeric matrix as currently mentioned in the literature [14]. As protein association with the particle surface plays the dominant role, the characterization of particle surface properties and the resulting interactions with the active compound needs
to be analyzed in much more detail than it is currently done in most of the studies. It is currently unclear to what extent the different forces influence the protein–particle interaction and to what magnitude each factor can modify this physisorptive link, as visibly the adsorptive binding is strong enough to ensure a drug transport into cells and to exercise an intracellular effect [6]. Another aspect also arises from the fact that surface properties of NPs are essentially dominated by the properties of the absorbed proteins. It needs to be studied in detail how the protein properties influence the biodistribution of the drug carrier, and how other factor such as toxicity are dominated or modified by the simple presence of the protein on the particle surface. First steps in this direction indicate the importance of this aspect [33]. Moreover, those can evolve according to complex adsorption and desorption pattern in their biological environment. A final issue is the currently complex preparation of such polymeric particles based on the double emulsion method, which limits the possibilities to transfer this method to an industrial scale due to its complexity and the therefrom resulting regulatory aspects requested by the health authorities worldwide. As drug load is essentially located on the surface of the particles manufacturing could be simplified by omitting complex steps within the current particle preparation methods, accelerating the development toward market entry of such nanocarrier products.
4. Conclusion The imaging technique based on IM-FESEM was able to visualize the internal morphology of drug loaded polymeric NPs prepared by a double emulsion method. It revealed that the current understanding of the internal structure of such particles needs to be revisited and suggests a further in-depth analysis of the conclusions drawn on the mechanisms of action and interaction of such drug carriers with the biological environment after administration.
Conflict of interest There is no conflict of interest statement. Acknowledgments Alf Lamprecht is thankful for funding in the framework of his membership at 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 ‘‘Investissements d’Avenir’’ with reference ANR-11-LABX-0021. Furthermore, we thank Dr. Martin Folger and Dr. Ragna Hoffmann from Boehringer Ingelheim Vetmedica GmbH for their support in this project. References [1] A. Lamprecht, IBD: selective nanoparticle adhesion can enhance colitis therapy, Nat. Rev. Gastroenterol. Hepatol. 7 (2010) 311–312. [2] J. Kreuter, Mechanism of polymeric nanoparticle-based drug transport across the blood–brain barrier (BBB), J. Microencapsul. 30 (2013) 49–54. [3] C. Damge, C. Reis, P. Maincent, Nanoparticle strategies for the oral delivery of insulin, Exp. Opin. Drug Deliv. 5 (2008) 45–68. [4] S.D. Putneyand, P.A. Burke, Improving protein therapeutics with sustainedrelease formulations, Nat. Biotechnol. 16 (1998) 153–157. [5] R.A. Petrosand, J.M. DeSimone, Strategies in the design of nanoparticles for therapeutic applications, Nat. Rev. Drug Discov. 9 (2010) 615–627. [6] S.P. Kasturi, I. Skountzou, R.A. Albrecht, D. Koutsonanos, T. Hua, H.I. Nakaya, R. Ravindran, S. Stewart, M. Alam, M. Kwissa, F. Villinger, N. Murthy, J. Steel, J. Jacob, R.J. Hogan, A. Garcia-Sastre, R. Compans, B. Pulendran, Programming the magnitude and persistence of antibody responses with innate immunity, Nature 470 (2011) 543–547.
D. Francis et al. / European Journal of Pharmaceutics and Biopharmaceutics 89 (2015) 56–61 [7] Z. Cheng, A. Al Zaki, J.Z. Hui, V.R. Muzykantov, A. Tsourkas, Multifunctional nanoparticles: cost versus benefit of adding targeting and imaging capabilities, Science 338 (2012) 903–910. [8] S. Hasani, Y. Pellequer, A. Lamprecht, Selective adhesion of nanoparticles to inflamed tissue in gastric ulcers, Pharm. Res. 26 (2009) 1149–1154. [9] S. McClean, E. Prosser, E. Meehan, D. O’Malley, N. Clarke, Z. Ramtoola, D. Brayden, Binding and uptake of biodegradable poly-DL-lactide micro- and nanoparticles in intestinal epithelia, Eur. J. Pharm. Sci. 6 (1998) 153–163. [10] T.A.P.F. Doll, S. Raman, R. Dey, P. Burkhard, Nanoscale assemblies and their biomedical applications, J. R. Soc. Interface 10 (2013) 80. [11] U. Bilati, E. Allemann, E. Doelker, Poly(D,L-lactide-co-glycolide) protein-loaded nanoparticles prepared by the double emulsion method – processing and formulation issues for enhanced entrapment efficiency, J. Microencapsul. 22 (2005) 205–214. [12] M.D. Blancoand, M.J. Alonso, Development and characterization of proteinloaded poly(lactide-co-glycolide) nanospheres, Eur. J. Pharm. Biopharm. 43 (1997) 287–294. [13] M.F. Zambaux, F. Bonneaux, R. Gref, P. Maincent, E. Dellacherie, M.J. Alonso, P. Labrude, C. Vigneron, Influence of experimental parameters on the characteristics of poly(lactic acid) nanoparticles prepared by a double emulsion method, J. Control. Release 50 (1998) 31–40. [14] A. Lamprecht, N. Ubrich, M. Hombreiro Perez, C. Lehr, M. Hoffman, P. Maincent, Biodegradable monodispersed nanoparticles prepared by pressure homogenization–emulsification, Int. J. Pharm. 184 (1999) 97–105. [15] B.A. Pfeifer, J.A. Burdick, R. Langer, Formulation and surface modification of poly(ester-anhydride) micro- and nanospheres, Biomaterials 26 (2005) 117– 124. [16] A. Lamprecht, H. Rodero Torres, U. Schäfer, C.-M. Lehr, Biodegradable microparticles as a two-drug controlled release formulation: a potential treatment of inflammatory bowel disease, J. Control. Release 69 (2000) 445– 454. [17] A. Lamprecht, N. Ubrich, M. Hombreiro Perez, C. Lehr, M. Hoffman, P. Maincent, Influences of process parameters on nanoparticle preparation performed by a double emulsion pressure homogenization technique, Int. J. Pharm. 196 (2000) 177–182. [18] R.C. Mundargi, V.R. Babu, V. Rangaswamy, P. Patel, T.M. Aminabhavi, Nano/ micro technologies for delivering macromolecular therapeutics using poly(D,Llactide-co-glycolide) and its derivatives, J. Control. Release 125 (2008) 193– 209. [19] E. Cohen-Sela, M. Chorny, N. Koroukhov, H.D. Danenberg, G. Golomb, A new double emulsion solvent diffusion technique for encapsulating hydrophilic molecules in PLGA nanoparticles, J. Control. Release 133 (2009) 90–95. [20] S. Khoee, A. Sattari, F. Atyabi, Physico-chemical properties investigation of cisplatin loaded polybutyladipate (PBA) nanoparticles prepared by w/o/w, Mater. Sci. Eng. C 32 (2012) 1078–1086.
61
[21] S. Khoeeand, M. Yaghoobian, An investigation into the role of surfactants in controlling particle size of polymeric nanocapsules containing penicillin-G in double emulsion, Eur. J. Med. Chem. 44 (2009) 2392–2399. [22] Q. He, J. Liu, X. Sun, Z.R. Zhang, Preparation and characteristics of DNAnanoparticles targeting to hepatocarcinoma cells, World J. Gastroenterol. 10 (2004) 660–663. [23] C. Giovino, I. Ayensu, J. Tetteh, J.S. Boateng, Development and characterisation of chitosan films impregnated with insulin loaded PEG-b-PLA nanoparticles (NPs): a potential approach for buccal delivery of macromolecules, Int. J. Pharm. 428 (2012) 143–151. [24] I. Gutierro, R.M. Hernandez, M. Igartua, A.R. Gascon, J.L. Pedraz, Size dependent immune response after subcutaneous, oral and intranasal administration of BSA loaded nanospheres, Vaccine 21 (2002) 67–77. [25] B. Sahana, K. Santra, S. Basu, B. Mukherjee, Development of biodegradable polymer based tamoxifen citrate loaded nanoparticles and effect of some manufacturing process parameters on them: a physicochemical and in-vitro evaluation, Int. J. Nanomed. 5 (2010) 621–630. [26] V. Hoffart, A. Lamprecht, P. Maincent, T. Lecompte, C. Vigneron, N. Ubrich, Oral bioavailability of a low molecular weight heparin using a polymeric delivery system, J. Control. Release 113 (2006) 38–42. [27] P.K. Smith, R.I. Krohn, G.T. Hermanson, A.K. Mallia, F.H. Gartner, M.D. Provenzano, E.K. Fujimoto, N.M. Goeke, B.J. Olson, D.C. Klenk, Measurement of protein using bicinchoninic acid, Anal. Biochem. 150 (1985) 76–85. [28] R.A. Jain, The manufacturing techniques of various drug loaded biodegradable poly(lactide-co-glycolide) (PLGA) devices, Biomaterials 21 (2000) 2475–2490. [29] A.J. Castellanos, S.J. Toro-Mendoza, M. Garcia-Sucre, Correction to the interfacial tension by curvature radius: differences between droplets and bubbles, J. Phys. Chem. B 113 (2009) 5891–5896. [30] C. Wischke, H.H. Borchert, Influence of the primary emulsification procedure on the characteristics of small protein-loaded PLGA microparticles for antigen delivery, J. Microencapsul. 23 (2006) 435–448. [31] S. Schuster, R. Bernewitz, G. Guthausen, J. Zapp, A.M. Greiner, K. Köhler, H.P. Schuchmann, Analysis of W1/O/W2 double emulsions with CLSM: statistical image processing for droplet size distribution, Chem. Eng. Sci. 81 (2012) 84– 90. [32] K. Mäder, B. Bittner, Y. Li, W. Wohlauf, T. Kissel, Monitoring microviscosity and microacidity of the albumin microenvironment inside degrading microparticles from poly(lactide-co-glycolide) (PLG) or ABA-triblock polymers containing hydrophobic poly(lactide-co-glycolide) A blocks and hydrophilic poly(ethyleneoxide) B blocks, Pharm. Res. 15 (1998) 787–793. [33] M. Arruebo, M. Valladares, A. Gonzalez-Fernandez, Antibody-conjugated nanoparticles for biomedical applications, J. Nanomater. (2009) 1–24 (Article ID 439389).