Journal of Microencapsulation, 2009; 26(6): 556–561
Effect of microfluidization parameters on the physical properties of PEG-PLGA nanoparticles prepared using high pressure microfluidization Shabnam N. Sani1,2, Nandita G. Das3 & Sudip K. Das3 1
Department of Pharmaceutical Sciences, Idaho State University, Pocatello, ID, USA, 2Department of Pharmaceutical Sciences, Massachusetts College of Pharmacy and Health Sciences, Boston, MA, USA, 3Department of Pharmaceutical Sciences, Butler University, Indianapolis, IN, USA Abstract The objective of this work was to develop uniformly distributed poly(ethylene glycol) grafted poly(lactideco-glycolide) (PEG-PLGA) nanoparticles of mean size range 100–200 nm using ethyl acetate as the solvent. In the multiple emulsion solvent evaporation method a high pressure microfluidization process was adopted to produce the W/O/W multiple emulsion. Non-toxic ethyl acetate was used to solubilize PEG-PLGA. The mean size of nanoparticles obtained was less than 180 nm. The particle size and size distribution were dependent on the microfluidization conditions applied. Mean particle size steadily increased from 121 nm at three passes to 172 nm at 20 passes of the microfluidizer, indicating that over-processing may be detrimental to PEG-PLGA nanoparticles prepared using this technique. There was no significant alteration of the PEG-PLGA matrix, as evidenced from the differential scanning calorimetric studies. Key words: Nanoparticle, microfluidization, poly(ethylene glycol) grafted poly(lactide-co-glycolide) (PEG-PLGA), ethyl acetate
Introduction Early reports on the multiple emulsion (W/O/W) solvent evaporation method for the preparation of poly(DL-lactide) (PLA) and poly(lactide-co-glycolide) (PLGA) biodegradable nanoparticles by Bodmeier and McGinity (1987, 1988) and Ogawa et al. (1988) started appearing in the literature in the 1980s. This method was subsequently modified and applied towards the delivery of proteins and other small molecule drugs by a number of different research groups (Jeffery et al. 1993, Chattaraj et al. 1999). The major existing challenges of this method for production of nanoparticles are the parameters to control the particle size and the outcome of uniform size distribution for small particles, i.e. particles in the range of 100–200 nm or smaller. A recent i.e., review by Astete and Sabliov (2006) discussed various approaches taken by researchers to control particle size and obtain a narrow size
distribution. Recently, Gaumet et al. (2007). reviewed the size requirements identified by researchers for nanoparticles that are specific for passive targeting to various tissues. Since very small particles (520–30 nm) are cleared readily by renal excretion and bigger particles are taken up by the mononuclear phagocyte system (MPS), particle size control is an extremely important issue in the intravenous delivery of nanoparticulate drugs. Surface modification of the PLA or PLGA nanoparticles with hydrophilic polyethylene glycol (PEG), forming di-PEG-PLA or di-PEG-PLGA copolymers, has been shown to provide a longer circulation time (Stolnik et al. 1994). The presence of PEG on the surface of the nanoparticles prevents their adsorption of plasma proteins (i.e. opsonization) and thereby prevents them from being identified by the MPS and subsequently removed (Otsuka et al. 2003, Avgoustakis 2004). This study utilized a pegylated polymer with the perspective of future in vivo biodistribution studies, as well as to explore
Address for correspondence: Sudip K. Das, Butler University, Department of Pharmaceutical Sciences, 4600 Sunset Avenue, Indianapolis, IN 46208, USA. E-mail:
[email protected] (Received 5 Nov 2007; accepted 24 Sep 2008) ISSN 0265-2048 print/ISSN 1464-5246 online ß 2009 Informa UK Ltd DOI: 10.1080/02652040802500655
http://www.informahealthcare.com/mnc
Effect of microfluidization parameters on the physical properties of PEG-PLGA whether microfluidization (high pressure homogenization) is suitable for the production of nanoparticles using this type of polymer that conform to the defined acceptance criteria. Previous studies on development of submicron emulsion systems reported the efficacy of a microfluidization process in achieving a higher degree of emulsion stability (Pinnamaneni et al. 2003). In this study, in order to achieve small but stable droplet sizes of W/O/W emulsions, microfluidization was used instead of processes more commonly used, such as high-speed stirring, homogenization alone or sonication. From the literature review, it appears that the most commonly used modifications of the multiple emulsion (W/O/W) solvent evaporation technique utilize dichloromethane to solubilize the polymer. While some researchers (Yoncheva et al. 2003, Lee et al. 2003, Dong and Feng 2007) studied the properties of PLA/PLGA nanoparticles that were prepared using a high pressure homogenizer, these studies were done using dichloromethane as the solvent for the polymer. Dichloromethane is a Class 2 solvent that poses problems in use in pharmaceutical preparations due to its potential toxicity (Impurities: Guideline for residual solvents 2007); therefore, for this research project ethyl acetate was 4 C), chosen (boiling point 77 C and flash point a relatively non-toxic Class 3 solvent that would be more justified for pharmaceutical applications. Another common Class 3 solvent, acetone, was not chosen because of its low boiling point (56.5 C) and low flash point ( 20 C) which produces highly porous particles (Ruan and Feng 2003) that eventually adversely facilitate the drug release. Moreover, processing with acetone must be done very carefully because of its high flammability. Although a group of investigators have reported the use of ethyl acetate in dissolving PLGA as the oil phase (Blanco and Alonso 1997, Soppimath and Aminabhavi 2002, Weissenboeck et al. 2004) in the primary W/O emulsion for the preparation of nanoparticles, their application did not involve high pressure homogenization as part of the preparation process. Therefore, the objective of this work was to develop a novel method for preparation of PEGPLGA nanoparticles in the size range less than 200 nm using a high pressure microfluidization process, with narrow size distribution and using ethyl acetate as the solvent for the polymer in the oil phase.
Experimental Materials Polyethylene glycol 5000/75 : 25 Poly(DL-lactide-co-glycolide, LactelTM) was a gift from Birmingham Polymers (AL). Polyvinyl alcohol (PVA), Mw 30,000–70,000 was obtained
557
from Sigma-Aldrich (St. Louis, MO). HPLC grade ethyl acetate was obtained from Fisher Scientific (PA). Double distilled water, filtered through 0.2 mm filter was used for all studies.
Method Preparation of nanoparticles. A W/O primary emulsion was formed by dissolving 200 mg of the PEG-PLGA polymer in 15 ml ethyl acetate and adding 1 ml of water, followed by emulsification by vortexing for 3 min. The primary emulsion was added to 250 ml PVA 2.0% w/v aqueous solution (external phase) and mixed for 30 s using an Ultra-Turrax stirrer (IKA, USA) to prepare a W/O/W emulsion. The resultant coarse multiple emulsion was processed through a MicrofluidizerTM 110L (Microfluidics Corp., Newton, MA) operated at room temperature and 18,000 psi air pressure in the interaction chamber. The fluid output from the microfluidizer was collected in its entirety and reprocessed in discrete cycles as opposed to continuous cycling of the emulsion through the microfluidizer. A complete cycle of microfluidization of the emulsion volume being processed is being referred to as a ‘pass’. Fractions of the processed emulsion were collected at 1, 3, 5, 8, 10, 15 and 20 passes to compare the particle size and characteristics of the resultant nanoparticles. The stainless steel capillary coil of the microfluidizer was immersed in running cold water to dissipate the heat produced during the microfluidization process to avoid uncontrolled thermal variables influencing the outcomes of the current study. For slow solvent evaporation while keeping the emulsion droplets fluidized, the output emulsion fractions were homogenized at 6400 rpm for 4 h using a Silverson SL2T high-speed homogenizer (Silverson machines, East Longmeadow, MA) in a vacuum hood. Any potential large polymer masses or aggregates were removed using a metal screen filter (pore size 500 mm) and the filtrate was centrifuged (Beckman L8-80 with 70.1 Ti rotor) at 260,000 g for 1 h. The supernatant was discarded and the nanoparticle plug was washed twice with water. The nanoparticles were re-suspended with 1 mL water for particle sizing. A 0.5 ml aliquot of this nanoparticle suspension was lyophilized (done without any cryoprotectant) and reserved for differential scanning calorimetry studies. Results in Table 1 and Figure 1 represent mean values SEM for three different batches of formulations, prepared on different days. Unless mentioned otherwise, all experiments were done in triplicate. In vitro degradation of PEG-PLGA was determined by quantitation of lactate as per the method described by Avgoustakis et al. (2002) This test is based on an enzymatic assay of quantitative oxidation of lactate formed by the degradation of the polymer. During nanoparticle
558
S. N. Sani et al. Table 1. Effect of microfluidization on the polydispersity index of the nanoparticles. Number of microfluidization passes 1 3 5 8 10 15 20
lyophilized nanoparticles. The solid samples (3–5 mg) were placed in non-hermetically sealed aluminum pans and an empty pan and lid assembly was used as reference. The reference and the sample pans were allowed to equilibrate for 10 min at 0 C and then heated at the rate of 10 min 1 from 0 C to 400 C in a nitrogen atmosphere.
Polydispersity index SEM 1.69 0.27 0.59 0.19 1.49 0.17 1.34 0.23 1.32 0.25 1.34 0.14 1.32 0.26
Results and discussion
Hydrodynamic diameter (nm)
900 800 500 450 400 350 300 250 200 150 100 50 0 0
2
4 6 8 10 12 14 16 18 Number of microfluidization passes
20
Figure 1. Effect of the number of microfluidization passes on the particle size of PEG-PLGA nanoparticles.
preparation, the supernatant fluid was collected immediately after centrifugation and analysed for lactate content. Lactate generated by the microfluidization method was compared to the more commonly used nanoparticle preparation method of using high speed stirrers, using pure polymer soaked in water as the control. Particle characterization. The purified nanoparticles were re-suspended in water and diluted as necessary in disposable cuvettes. The particle size was determined using the dynamic light scattering technique at a fixed 90 angle using a Beckman Coulter N4 Plus particle size analyser (Beckman Coulter, Miami, FL). Particle size and polydispersity indices are reported as mean particle diameter (nm) standard error of mean (SEM, nm). Particle surface morphology was studied using scanning electron micrography (Philips XL30ESEM) after drying diluted nanoparticle suspensions on stubs and coating with gold. Differential scanning calorimetry. Differential scanning calorimetry (Modulated Differential Scanning Calorimeter TA 2920, TA instruments, DE) was carried out on the
Several methods have been proposed for the preparation of PLGA (Astete and Sabliov 2006) and PEG-PLGA (Avgoustakis 2004) nanoparticles using multiple emulsion solvent evaporation methods. Nevertheless, to the authors’ knowledge, no study has been reported on the preparation of PEG-PLGA nanoparticles using ethyl acetate as the solvent for the polymer and using microfluidization for droplet size reduction. It is to be noted that using a surfactant PoloxamerTM 188, Blanco and Alonso reported a particle size between 320–521 nm using ethyl acetate as the solvent. Other reported use of ethyl acetate have resulted in particle sizes over 1 micron (Soppimath and Aminabhavi 2002, Weissenboeck et al. 2004). Also, in a recent study by Dong and Feng using high pressure homogenization, the smallest particle size reported was 245 nm. In these studies, the nanoparticles obtained from the non-microfluidized batches demonstrated a wide size range from 220– 800 nm with high polydispersity indices. The polydispersity index values of greater than 1 indicated multiple size populations and justified the high-pressure homogenization approach for reducing the particle size of the nanoparticles and potential for producing narrow size distributions. As described below, the microfluidized batches yielded mean particle size ranges between 100–200 nm. One-way analysis of variance of the data indicated that the mean nanoparticle size and polydispersity index depend on the number of microfluidization passes. As evidenced in Figure 1, significant reduction of particle size of the coarse emulsion was obtained with microfluidization, which reduced particle size of the resultant nanoparticles up to three passes. More importantly, the polydispersity index was lowered significantly at three passes (polydispersity index 0.59 0.19), indicating that the size distribution of the particles was narrow and uniform. The polydispersity indices for the various passes are presented in Table 1; a polydispersity index value of 1 or less is indicative of a unimodal size population which is an important characteristic for quality control, particularly batch-tobatch variation. While the droplet size reduction from pass 1 (126 14 nm) to pass 3 (121 4 nm) was not found to be significantly different, the improvement in polydispersity index and marked reduction in standard error of mean at pass 3 indicate positive gains from
Effect of microfluidization parameters on the physical properties of PEG-PLGA microfluidization up to this stage. It is likely that the shear stress produced by the microfluidizer between 1–3 processing passes is favourable for size reduction of the emulsion droplets and resultant size reduction of the nanoparticles. Interestingly, there was no further reduction in particle size or improvement in polydispersity index characteristics with further processing after pass 3 (passes 5–20). In fact, a reversing trend was observed and the biggest particle size amongst microfluidized batches was obtained at 20 passes (172 14 nm, polydispersity index 1.32 0.26). With the exception of pass 15 data, statistical differences were not observed between the particle size and polydispersity index values from passes 5–20. The increase in particle size and the worsening of the polydispersity indices could be indicative of coalescence of the internal phase droplets during processing. However, ageing of the emulsions for 1 week did not produce any statistically significant change in particle size of the various batches (data not presented), indicating that the coalescence of the particles was likely self-limiting and dependent on the amount of shear stress and not progressive in nature. Hence, the observed behaviours were potentially correlated to thermodynamic variables and not kinetic changes. As evidenced from the degradation study of PEG-PLGA, minor degradation of the copolymer occurs during nanoparticle preparation. It was observed that 1.1% w/w lactate was produced during the manufacturing process. As expected, the pure polymer control soaked in water in unstirred condition showed even less degradation, at 0.06% w/w lactate. It was observed that the lactate produced by conventional high speed stirring process was comparable to the microfluidization process. Scanning electron microscopy indicated that between 3–5 microfluidization passes, the resultant nanoparticles harvested from the dispersion were spherical in shape with occasional pores visible on the surface (Figure 2). Since diffusion of the organic phase from the emulsion droplet during nanoparticle formation is believed to lead to the formation of the pores in the matrix (Sah 1997), controlling the rate of solvent evaporation should effectively control the degree of pores in the porous matrix. As this study adopted a slow evaporation process with the intention of minimizing the number of pores, the observations from Figure 2 are consistent with the finding that slow evaporation of the organic solvent could minimize the number of pores in the nanoparticle matrix. As far as particle size and morphology were concerned, the scanning electron micrographs offered important clues to the observed reversing pattern in particle size and polydispersity index with increase in shear stress placed on the particles. For batches microfluidized beyond pass 5, coalesced masses of nanoparticles consisting of a wide variety
559
Figure 2. Scanning electron micrographs of PEG-PLGA nanoparticles at three microfluidization passes.
Figure 3. Scanning electron micrographs of PEG-PLGA nanoparticles at 20 microfluidization passes.
of particle sizes from small (5100 nm) to large (4500 nm) were observed, which explained their increasing mean diameter as well as the larger polydispersity indices indicative of multiple size populations. With maximum shear stress, at 20 passes, in addition to nanoparticle aggregates elongated strands of polymer were observed, which appear to be shredded nanoparticle matrices (Figure 3). This provides conclusive evidence that over-processing and excessive shear stress placed upon nanoparticles is likely to be detrimental upon the size distribution and morphology of the polymer particles. Although microfluidization is greatly beneficial for producing droplet size reduction and narrow size distributions, as indicated by these studies, a minimal number of passes may produce the best outcomes in polymer particle formation. Differential scanning calorimetry (DSC) was conducted to investigate for potential physical interactions between the nanoparticle components, as well as possible changes occurring with microfluidization due to the shear stress placed upon the polymer. Figure 4 shows the DSC thermograms for PEG-PLGA nanoparticles, PEG-PLGA and PVA
560
S. N. Sani et al. 0.5
unique features of the microfluidization technique for preparation of PEG-PLGA nanoparticles were: (1) mean size 120 nm, (2) low polydispersity, indicating unimodal distribution of nanoparticle population at three passes and (3) use of a non-toxic solvent, ethyl acetate, for dissolving PEG-PLGA, thus requiring no additional processing to ensure removal of residual toxic solvent.
Heat Flow (W/g)
0.0
−0.5 −1.0 −1.5 −2.0 −100
Acknowledgement
3 passes PEG_PLGA nanoparticles PEG_PLGA polymer poly vinyl alcohol
Exo Up
0
100
200
Temperature (°C)
300
400
Universal V3.0G TA Instruments
Figure 4. Differential scanning calorimetry study of nanoparticles and its individual components.
at pass 3. In the presence of PVA, the onset temperature for melting as well as the melting peak of PEG-PLGA was shifted slightly downwards, likely due to the plasticization of the polymer in the presence of PVA. As plasticization of the polymer could favour even distribution of dispersed chemicals in the polymer matrix, this interaction was not deemed as detrimental. One did not observe any differences among the DSC thermograms for nanoparticles as a factor of microfluidization, indicating that shear stress did not induce any specific physical changes or interactions in the polymer as observable by differential scanning calorimetry.
Conclusion Controlled microfluidization of the W/O/W emulsion formed during the multiple emulsion solvent evaporation process was successful in producing uniform size distribution of PEG-PLGA nanoparticles in the size range of 120 nm. Nanoparticle morphology was uniform, as evidenced from scanning electron microscopy. Differential scanning calorimetry did not show any significant change in the polymer characteristics with respect to shear stress induced by microfluidization. Using this novel technique, this study successfully produced nanoparticles of PEG-PLGA using non-toxic ethyl acetate as the solvent for potential future intravenous administration of this surface modified nanoparticle carrier. It is also to be noted that other than PVA in the external phase these nanoparticles were produced without using any other cell membrane destabilizing surfactants that might have a detrimental effect on red blood cells and related tissues upon IV administration of the nanoparticles. Therefore,
The authors acknowledge Ms Tammy Townbridge of the Idaho National Laboratories (INL), Idaho Falls, ID, for obtaining the SEM photographs. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
References Astete CE, Sabliov CM. 2006. Synthesis and characterization of PLGA nanoparticles. J Biomater Sci Polym Ed 17:247–289. Avgoustakis K. 2004. Pegylated poly(lactide) and poly(lactide-co-glycolide) nanoparticles: Preparation, properties and possible applications in drug delivery. Curr Drug Deliv 1:321–333. Avgoustakis K, Beletsi A, Panagi Z, Klepetsanis P, Karydas AG, Ithakissios DS. 2002. PLGA-mPEG nanoparticles of cisplatin: In vitro nanoparticle degradation, in vitro drug release and in vivo drug residence in blood properties. J Contr Rel 79:123–135. Blanco MD, Alonso MJ. 1997. Development and characterization of protein-loaded poly(lactide-co-glycolide) nanospheres. Eur J Pharm Biopharm 43:287–294. Bodmeier R, McGinity JW. 1988. Solvent selection in the preparation of poly(lactide) microspheres prepared by the solvent evaporation method. Int J Pharm 43:179–186. Bodmeier R, McGinity JW. 1987. The preparation and evaluation of drugcontaining poly(Dl-lactide) microspheres formed by the solvent evaporation method. Pharm Res 4:465–471. Chattaraj SC, Rathinavelu A, Das SK. 1999. Biodegradable microparticles of influenza viral vaccine: Comparison of the effects of routes of administration on the in vivo immune response in mice. J Contr Rel 58:223–232. Dong Y, Feng SS. 2007. Poly(D, L-lactide-co-glycolide) (PLGA) nanoparticles prepared by high pressure homogenization for paclitaxel chemotherapy. Int J Pharm 342:208–214. Gaumet M, Vargas A, Gurny R, Delie F. 2007. Nanoparticles for drug delivery: The need for precision in reporting particle size parameters. Eur J Pharm Biopharm, doi:10.1016/j.ejpb.2007.08.001. Impurities: Guideline for residual solvents, Q3C (R3), ICH Harmonized Tripartite Guideline. Available online at: http://www.ich.org/LOB/ media/MEDIA423.pdf, accessed 15 October 2007. Jeffery H, Davis SS, O’Hagan DT. 1993. The preparation and characterization of poly(lactide-co-glycolide) microparticles. II. The entrapment of a model protein using a (water-in-oil)-in-water emulsion solvent evaporation technique. Pharm Res 10:362–368. Lee WK, Park JY, Yang EH, Suh H, Kim SH, Chung DS, et al. 2003. Investigation of the factors influencing the release rates of cyclosporin A-loaded micro- and nanoparticles prepared by high-pressure homogenizer. J Contr Rel 84:115–123. Ogawa Y, Yamamoto M, Okada H, Yashiki T, Shimamoto T. 1988. A new technique to efficiently entrap leuprolide acetate into microcapsules of
Effect of microfluidization parameters on the physical properties of PEG-PLGA polylactic acid or copoly(lactic/glycolic) acid. Chem Pharm Bull (Tokyo) 36:1095–1103. Otsuka H, Nagasaki Y, Kataoka K. 2003. PEGylated nanoparticles for biological and pharmaceutical applications. Adv Drug Deliv Rev 55:403–419. Pinnamaneni S, Das NG, Das SK. 2003. Comparison of oil-in-water emulsions manufactured by microfluidization and homogenization. Pharmazie 58:554–558. Ruan G, Feng SS. 2003. Preparation and characterization of poly(lactic acid)-poly(ethylene glycol)-poly(lactic acid) (PLA-PEG-PLA) microspheres for controlled release of paclitaxel. Biomaterials 24:5037–5044. Sah H. 1997. Microencapsulation techniques using ethyl acetate as a dispersed solvent: Effects of its extraction rate on the characteristics of PLGA microspheres. J Contr Rel 47:233–245. Soppimath KS, Aminabhavi TM. 2002. Ethyl acetate as a dispersing solvent in the production of poly(Dl-lactide-co-glycolide) microspheres: Effect
561
of process parameters and polymer type. J Microencapsulation 19:281–292. Soppimath KS, Aminabhavi TM. 2002. Ethyl acetate as a dispersing solvent in the production of poly(Dl-lactide-co-glycolide) microspheres: Effect of process parameters and polymer type. J Microencapsulation 19:281–292. Stolnik S, Dunn SE, Garnett MC, Davies MC, Coombes AG, Taylor DC, et al. 1994. Surface modification of poly(lactide-co-glycolide) nanospheres by biodegradable poly(lactide)-poly(ethylene glycol) copolymers. Pharm Res 11:1800–1808. Weissenboeck A, Bogner E, Wirth M, Gabor F. 2004. Binding and uptake of wheat germ agglutinin-grafted PLGA-nanospheres by caco-2 monolayers. Pharm Res 21:1917–1923. Yoncheva K, Vandervoort J, Ludwig A. 2003. Influence of process parameters of high-pressure emulsification method on the properties of pilocarpine-loaded nanoparticles. J Microencapsulation 20:449–458.
