Preparation of Biodegradable Polycaprolactone/Poly (ethylene glycol)/Polycaprolactone (PCEC) Nanoparticles

Abstract

Biodegradable polyetherester copolymer (PCL/PEG/PCL, PCEC) was synthesized by ring-opening polymerization of ε-caprolactone initiated by poly(ethylene glycol) (PEG). The PCEC nanoparticles were prepared by solvent diffusion method or w/o/w double emulsion method. The obtained particles' morphology was observed on scanning electron microscopy, and the particle size distribution was determined using Malvern laser particle sizer. Bovine serum albumin was used as the model water-soluble protein drug, which was successfully encapsulated in PCEC nanoparticles, the drug release behavior was studied in detail. The hydrolytic degradation behavior of the PCEC nanoparticles was also studied.

Keywords:

• Drug Delivery System
• Nanoparticle
• Polycaprolactone/Poly(Ethylene Glycol)/Polycaprolactone
• Protein Drug

Aliphatic polyesters, in particular FDA-approved bioresorbable poly(DL-lactide) (PLA) and poly(ε-caprolactone) (PCL), constitute a class of biomedical materials of growing interest in the field of temporary therapeutic applications in surgery, sustained drug delivery, and tissue engineering (Birnbaum, Kosmala, and Peppas 2000; Kwon et al. 2001; Molpeceres, Aberturas, and Guzman 2000; Chen et al. 2002; Zhang and Zhou 2005).

PCL is a kind of biocompatible and biodegradable polyester, which has attracted great attention in controlled drug delivery system due to its nontoxicity and great penetrability. However, its further application in drug delivery system has been restricted by its some shortcomings, such as slow degradation rate due to its relatively high crystallinity and great hydrophobicity. So, hydrophilic poly(ethylene glycol) (PEG) was incorporated into or blended with PCL to increase the hydrophilicity and the biodegradation rate. PEG is a nontoxic and nonimmunogenic water soluble polymer, and it has a great effect on preventing protein, adsorption and escaping capture of the reticuloendothelial system (Ryu et al. 2001; Zhang and Zhuo 2005). So, in this study, we prepared PCL/PEG/PCL (PCEC) triblock copolymers by ring-opening polymerization of ε-caprolactone initiated by PEG. Due to the hydrophilicity of PEG and hydrophobicity of PCL, the obtained PCEC copolymer is amphiphilic.

Nanoparticles are colloidal polymeric drug carriers, and their potential uptake as well as their stability in the gastrointestinal tract indicates that nanoparticles are expected to be promising carriers for delivery of therapeutic drugs (Birnbaum et al. 2000; Molpeceres et al. 2000; Kwon et al. 2001; Ryu et al. 2001; Chen et al. 2002; Ubrich et al. 2004; Zhang and Zhuo 2005). So, nanoparticles have been widely employed for drug delivery systems and many other biomedical applications recently. Since nanoparticle drug delivery systems can provide intravenous injection of drugs for site-specific drug delivery, this local drug delivery rather than systemic administration might be a more effective way to obtain higher tissue drug levels and decrease potential adverse systemic side effects (Kwon et al. 2001).

Proteins and peptides are new kinds of bioactive drugs developed more recently due to great development in biotechnology, and they have the advantage of curing many diseases over traditional drugs. But unfortunately, it is very difficult to keep bioactivity of such bioactive protein or peptide drugs. So, it is very important to avoid their denaturation of protein or peptide drugs before delivery to the target organs (Zhou, Deng, and Li 2001). In our study, bovine serum albumin (BSA) was employed as a model water soluble protein to be encapsulated in PCEC nanoparticles by solvent diffusion method or w/o/w double emulsion methods, and in vitro release behavior was also studied in our work.

EXPERIMENTAL

Materials

Sn(Oct)₂ and ε-caprolactone (CL) were obtained from Aldrich Company (USA). BSA was BR grade and purchased from Shanghai Biotechnology Company (P.R. China) Poly (ethylene glycol) (PEG, M_n = 4000) and poly(vinyl alcohol) (PVA, M_n = 8000–12000, 88% hydrolyzed) were purchased from Sigma (USA). All the other materials were obtained from Chengdu KeLong Chemicals (China). In this study all the materials except BSA were analytical grade and used as received without any purification.

Synthesis of Triblock PCEC Copolymers

Triblock PCEC copolymer was synthesized from ε-caprolactone by ring-opening method initiated by PEG, (Diagram 1) (Ryu et al. 2001). The typical PCEC100000 copolymer was prepared as follows: ε-CL (24 g), PEG4000 (1 g), and Sn(Oct)₂ (0.05 g) were added into reaction vessel under nitrogen atmosphere. The mixture was first kept at 110°C for 1 hr under nitrogen atmosphere. Later, the temperature was gradually elevated to 160°C over 30 min. Then, the mixture was rapidly heated to 180°C under vacuum for another 7 hr. At the end, the resultant melt was poured onto a steel plate under nitrogen atmosphere, thus PCEC100000 copolymer was obtained.

DIAG. 1.  Synthesis scheme of the PCEC copolymer.

Purification of PCEC Copolymers

The resultant PCEC copolymers were first dissolved in AR grade trichloromethane (TCM) and reprecipitated from the filtrate using AR grade n-hexane. Then, this mixture was filtered and vacuum dried to constant weight. The purified PCEC copolymers were kept in air-tight bags in desiccators before use.

Fourier Transform Infrared Spectroscopy (FTIR) and ¹H Nuclear Magnetic Resonance

FTIR (KBr) spectrum of the copolymers was taken in NICOLET 200SXV560 infrared spectrophotometer (Nicolet, USA). ¹H-NMR spectrum (in CDCl₃) was recorded on Varian 400 spectrometer (Varian, USA) at 400 MHz using tetramethylsilane as internal reference standard.

Preparation of PCEC Nanoparticles

In this work, PCEC100000 copolymer was used to prepare PCEC polyetherester nanoparticles. The solvent displacement (diffusion) method and solvent evaporation method were both used. PCEC100000 was first dissolved in 5 ml of organic solvents (DMSO, DMSO/CH₂Cl₂, or CH₂Cl₂). The resultant polymer solution was then added into 100 ml of aqueous polyvinyl alcohol (PVA) solution with continuous stirring at predetermined rate. Then, PCEC organic solution was immediately dispersed into drops by extensive stirring with a four-blade stirrer at a predetermined rate. For the diffusion or evaporation of organic solvent, the drops became smaller and smaller, and PCEC nanoparticles formed after 2 hr. The resultant PCEC nanoparticles suspension was ultracentrifuged for 20 min at 30000 rpm. The obtained nanoparticles were washed with distilled water and then ultracentrifuged three times. Then, the suspension was obtained by addition of distilled water before test. The preparing parameters of different PCEC nanoparticles are shown in Table 1.

TABLE 1  Preparing parameters for PCEC nanoparticles

Sample	Vouter aqueous (ml)	CPVA (%)	CPCEC (%)	Vinner aqueous (ml)	VSolvent (ml)	Theoretical drug loading (%)
N1	100	4	2.5	1	DMSO: 5	10
N2	100	4	2.5	0.75	DMSO: 5	10
N3	100	4	2.5	0.5	DMSO: 5	10
R2	100	4	2.5	1	DMSO: 2.5+CH2Cl2: 2.5	10
R3	100	4	2.5	1	CH2Cl2: 5	10
Z2	100	4	2.5	1	DMSO: 5	5
Z3	100	4	2.5	1	DMSO: 5	15

V = volume, C = concentration.

Determination of Particle Size Distribution

PCEC nanoparticles were first dispersed in distilled water using ultrasound dispersion, and the particle size distribution was determined on Malvern Instrument (NanoZS90, Malvern Instruments, UK).

Drug Loading Determination

BSA content of the BSA/PCEC nanoparticles was determined using alkaline digestion method, which was shown as follows: ∼20 mg of BSA-loaded PCEC nanoparticles was suspended in 2 ml of 0.01 M sodium hydroxide, and the suspension was stirred at room temperature for 24 hr, until complete hydrolysis of the PCEC copolymers. The resultant transparent hydrolysate was then spectrophotometrically measured at 278 nm (UV spectrophotometer, Analytik Jena Specord 200, Germany), and the BSA concentration in hydrolysate was calculated according to the standard curve of BSA. Each sample was assayed in triplicate. In this study, drug loading (DL) and entrapment efficiency (EE) were calculated according to Equations 1 and 2 (Jia et al. 2007);

In Vitro Degradation Study

The PCEC100000 nanoparticles were incubated in 0.01 mol/L alkaline NaOH solution (pH = 12) at 37°C. The samples were removed from the bottles at predetermined time. Then the surface morphology of PCEC nanoparticle was observed on SEM (Amray, USA) after sputtered with gold.

In Vitro Drug Release Studies

First, 100 mg of PCEC nanoparticles were immersed in 25 ml of PBS solution (0.01M, pH 7.4) in Eppendorf test tube shaken at 60 rpm at 37°C. Second, predetermined intervals, the Eppendorf test tube was ultracentrifuged at 30000 rpm for 20 min, and 5.0 ml PBS solution was removed from the supernatant and replaced by fresh release medium. BSA concentration was determined by UV method as already mentioned. The accumulative release weight of BSA was calculated according to equation 3 (Jia et al. 2007): where Q is accumulative release weight, C_n is measured concentration of the drug, V_t is volume of the medium, and V_s is volume of solution removed from supernatant. The results shown in this work are the mean of three parallel experiments.

Polyacrylamide Gel Electrophoresis (PAGE)

The structural integrity of BSA extracted from nanoparticles or released in vitro was detected by SDS-PAGE, compared with native BSA and reference markers. Protein samples were diluted with Tris-buffer (pH 6.8) with 2% SDS. Electrophoresis of samples was performed at a constant voltage of 200V in a Tris/glycine/SDS buffer using a BioRad Mini-Protein II electrophoresis system. After migration, the gel was stained with coomassie bright blue in methanol-acetic-water (2.5:1:6.4) to reveal protein, then destained and dried.

Statistical Data Analysis

Multiple experiment data were obtained and presented as mean values. Statistical comparisons were done at a significant level of P < 0.05 using Student's t-test.

RESULTS AND DISCUSSION

Polymer Characterization

The biodegradable PCEC triblock copolymer was successfully prepared by ring opening copolymerization of ε-caprolactone initiated by PEG. FTIR and ¹H-NMR were used to characterize the chemical structure of PCEC copolymer. The FTIR spectrum is shown in Figure 1. The absorption band at 1733 cm⁻¹ was attributed to the C═ O stretching vibrations of ester carbonyl group. The absorption bands at 1142 cm⁻¹ and 1243 cm⁻¹ were attributed to the characteristic C─ O─ C stretching vibrations of the repeated OCH₂CH₂ units of PEG and the ─ COO/bond bands stretching vibrations, respectively. It is obvious that the PCEC copolymer exhibits characteristic peaks of both PEG and PCL.

FIG. 1.  FTIR spectra of the PCEC copolymer.

To furtherly confirm the formation of PCEC copolymer, ¹H-NMR spectrum was also recorded and is shown in Figure 2. The characteristic absorption peaks were also indicated in this figure. Peaks at 1.40, 1.65, 2.32, and 4.06 ppm were assigned to methylene protons of ─ (CH₂)₃─, ─ OCCH₂─, and ─ CH₂OOC─ in PCL chains, respectively. The sharp peak at 3.65 ppm is attributed to the methylene protons of PEG segments. The very weak peaks at 4.23 ppm and 3.82 ppm were attributed, respectively, to the methylene protons of ─ O─ CH₂─ CH₂─ in PEG end unit. All the FTIR and ¹H-NMR results indicated that the PCEC copolymer formed successfully.

FIG. 2.  ¹H-NMR spectrum of the PCEC copolymer.

Preparation of PCEC Nanoparticles

In this study, several kinds of PCEC nanoparticles were prepared successfully, and the detailed results are shown in Table 2. The particle size was greatly affected by inner aqueous phase, organic solvent, and drug loading, which is discussed in detail as follows.

TABLE 2  Basic data of PCEC nanoparticles prepared in mean values (n = 3)

Number	D (nm)	PDI	DL (%)	EE (%)	Burst effect (%) (Q in 0.5 hr)
N1	598.18	0.105	4.95	49.5	61.7
N2	512.37	0.063	4.02	40.2	40.8
N3	431.58	0.043	2.18	21.8	46.4
R2	1044.97	0.111	1.75	17.5	62.4
R3	5532.01 (67%), 2712.84 (33%)	0.383	1.67	16.7	53.5
Z2	468.87	0.041	2.15	42.99	45.1
Z3	452.13	0.047	5.28	35.21	87.8

D = the average diameter; PDI = polydispersity index; DL = drug loading; EE = encapsulation efficiency; Q = accumulative release weight.

Effect of Volume of Inner Aqueous Phase (w₁)

Table 2 shows the basic data of nanoparticles. Samples N1, N2, and N3 have the same parameter except for the volume of inner aqueous phase. The average diameter of N1, N2, and N3 are 598.18, 512.37, and 431.58 nm, respectively. With the increase in volume of inner aqueous phase, the particle size of PCEC nanoparticles increased accordingly. This increase in particle size might be due to a coarse and less stable W₁/O emulsion, which favored the coalescence of suspended droplets. At the same time, the larger the volume of inner aqueous phase, the higher the entrapment efficiency, because the larger nanoparticles have higher drug-loading potential. And the PDI (polydispersity index) of PCEC nanoparticles are all less than or nearly 0.1, so these kinds of nanoparticles can be considered a monodispersed system (Lamprecht, Ubrich, and Perez 1999).

Effect of Organic Phase

DMSO is water soluble, and it could be diffused into water very quickly. CH₂Cl₂ cannot be dissolved in water, but it could evaporate from water. When DMSO is substituted or partly substituted by CH₂Cl₂, the particle size of the obtained PCEC nanoparticles will change, greatly.

The partly substitution of DMSO by CH₂Cl₂ significantly makes the size of the nanoparticle larger, and even the size of R₃ (CH₂Cl₂ only) is more than 5000 nm. This phenomenon might be due to the first W₁/O emulsion becoming coarse and less stable, which favored the coalescence of suspended droplets, as already mentioned in the section discussing the effect of the volume of inner aqueous phase. When the DMSO was partly substituted by CH₂Cl₂, the entrapment efficiency of the nanoparticles decreased accordingly, which might be due to slower solidification speed evoked by participation of CH₂Cl₂, and partial drug diffusion into the outer aqueous phase. Thus, the more the CH₂Cl₂ in the solvent, the lower the drug entrapment efficiency. All the results are shown in Table 2.

Effect of Nominal Drug Loading

In this study, the nominal drug loading is from 5% to 15%. In this region, the higher the nominal drug loading, the higher the actual drug loading. But the entrapment efficiency of Z₂ and Z₃ are 42.99% and 35.21%, respectively, which decreased accordingly compared with N₁ whose entrapment efficiency is 49.5%.

In Vitro Degradation Study

Figure 3 shows the surface morphology of PCEC nanoparticles during degradation, According to Figure 3, we found that the original PCEC particles had a spherical shape and the particle size ranged from 400 to 5000 nm with change in organic phase. The particles prepared by DMSO had a much smaller size and great uniformity, but the particles were easy to conglutinate with each other; on the contrary, when the organic phase was substituted or partly substituted by CH₂Cl₂, the size of the particles became larger, and the uniformity decreased, but the particles became less conglutinated. All of these results are consistent with the basic data discussed above. When degraded for 3 days, the particles became syncretic and the surface of the particles started flaking off, so the boundary of the particles was blurry and the surface obviously became coarse. When degraded for 5 days, the particles degraded into pieces, and many holes appeard on the fragment after degradation.

FIG. 3.  SEM morphology of various nanoparticles prepared by different organic solvents (with different mixing ratio of DMSO and CH₂Cl₂). (A) Organic phase is DMSO, degradation Time T = 0d (A1), T = 3 d (A2), and T = 5 d (A3). (B) Organic phase is DMSO + CH₂Cl₂, degradation Time T = 0d (B1), T = 3d (B2), and T = 5d (B3). (C) Organic phase is CH₂Cl₂ only, degradation Time T = 0d (C1), T = 3d(C2), and T = 5d (C3).

In Vitro Drug Release Studies

Figure 4 illustrates the in vitro release profiles of BSA loaded nanoparticles in 120 hr, by representing the percentage of BSA release with respect to the total encapsulated BSA. For all kinds of nanoparticles, there were two stages for BSA release. Stage I: a first initial burst release, the BSA, adsorbed on the surface of nanoparticles, was immediately released. Stage II: after initial burst, the drug release displayed a plateau release profile, resulting from diffusion of BSA from polymer matrix after erosion of PCEC polymers.

FIG. 4.  Release profiles of BSA nanoparticles in phosphate buffer at 37°C and pH 7.4 in 120 hr. (a) Effect of inner aqueous phase on BSA release profiles of PCEC nanoparticles: release profiles of N1, N2, and N3. (b) Effect of organic solvent on BSA release profiles of PCEC nanoparticles: release profiles of N1, R2, and R3. (c) Effect of drug loading on BSA release profiles of PCEC nanoparticles: release profiles of N1, Z2, and Z3.

Figure 5 shows the release profiles in the first 4 hr. Burst effect is defined here as the percentage released in the first 0.5 hr, compared to the total released BSA, because all samples of nanoparticles showed obvious BSA release in the first 0.5 hr. All the samples had burst release effect because of the surface absorption of the nanoparticles. The burst effect of Z₂, N₁, and Z₃ were 45.1, 61.7, and 87.8%, respectively, which ordinarily increased with increase in nominal drug loading. Obviously, higher nominal drug loading leads to the tremendous burst effect. As a matter of fact, an increase in nominal drug loading to very high levels is not always desirable, because the amount of polymer could be insufficient to completely entrap the drug, and a sustained release is expected (Bilati, Allemann, and Doelker 2005).

FIG. 5.  Release profiles of BSA nanoparticles in phosphate buffer at 37°C and pH 7.4 in 4 hr. (a) Effect of inner aqueous phase on BSA release profiles of PCEC nanoparticles: release profiles of N1, N2, and N3. (b) Effect of organic solvent on BSA release profiles of PCEC nanoparticles: Release profiles of N1, R2 and R3. (c) Release profiles of N1, Z2, and Z3.

PAGE Behavior of BSA in Vitro Release

Figure 6 shows the SDS-PAGE results of original BSA solution (band A) and BSA released from N1 (band B), and R2 (band C). It we see that the samples—BSA before encapsulation, and BSA released from N1 and R2—showed no additional peak of higher or lower molecular BSA. This suggests that no chemical polymerization, noncovalent aggregation, or molecular hydrolysis occurred during the preparation and release processes. From the above discussion, we conclude that biodegradable (PCEC) nanoparticles have a great potential application in protein delivery system.

FIG. 6.  Typical SDS-PAGE results of BSA IN vitro release. Band A = Original BSA; band B = BSA released from N1; band C = BSA released from R2.

CONCLUSION

The PCEC triblock copolymer was synthesized by ring-opening of ε-caprolactone with PEG and characterized by ¹H-NMR and FT-IR. The PCEC nanoparticles were prepared by solvent diffusion method or w/o/w double emulsion method. The particles were characterized by size, SEM. Also, the effect of the experiment parameter on the particles, and the degradation, and drug release in vitro were all studied in detail. The model drug BSA can be efficiently entrapped in PCEC nanoparticles, and it can be released extendedly for 5 days without denaturation.