New amphiphilic PEG-b-P(ester–ether) micelles as potential drug nanocarriers

J Nanopart Res (2012) 14:1168 DOI 10.1007/s11051-012-1168-y

RESEARCH PAPER

New amphiphilic PEG-b-P(ester–ether) micelles as potential drug nanocarriers Roubeena Jeetah • Archana Bhaw-Luximon Dhanjay Jhurry

•

Received: 29 March 2012 / Accepted: 28 August 2012 / Published online: 8 September 2012 Ó Springer Science+Business Media B.V. 2012

Abstract A range of diblock and triblock copolymers of dioxanone and methyl dioxanone (MeDX) were synthesized by ring-opening polymerization of dioxanone and MeDX initiated by hydroxyl-terminated PEG (MPEG) or di-amino-terminated PEG (JeffamineÒ) as macroinitiator in the presence of Sn(Oct)2. The copolymers exhibit amphiphilic behavior in water forming core–shell micelles in the size range 120–300 nm as measured by DLS. DSC measurements exhibit only one melting transition for all copolymers and confirm that increasing MeDX content of the copolymers lead to decreasing crystalline character and hydrophobic–hydrophilic chain entanglement. Anti-inflammatory drug ketoprofen was successfully loaded into the hydrophobic core of the micelles. Various key parameters such as micelle size, drug entrapment efficiency and drug release, which are dependent on crystalline structure and biodegradability characteristics of the hydrophobic core, could effectively be controlled by varying the dioxanone/ MeDX ratio of the (ester–ether) copolymer.

R. Jeetah  A. Bhaw-Luximon  D. Jhurry (&) ANDI Centre of Excellence for Biomedical and Biomaterials Research, MSIRI Building, University of Mauritius, Re´duit, Mauritius e-mail: [email protected]; [email protected]

Keywords Amphiphilic block copolymers  Nanoparticles  Dioxanone  Methyl dioxanone  Drug delivery  Ketoprofen

Introduction In the past 20 years, polymeric nanoparticles have attracted a lot of interest as drug carriers as they can improve therapeutic efficacy, prolong biological activity, and decrease drug dosage frequency (Nishiyama and Kataoka 2006). PEG-containing polymers have been mostly used for their ability to impart stealth characteristic to polymeric nanoparticles (Owens and Peppas 2006). Polymeric micelle-like nanoparticles have been widely investigated with hydrophilic PEG and a biodegradable hydrophobic polyester such as polylactide/glycolide or polycaprolactone or polypeptide (Kumar et al. 2001; Liu et al. 2006; Yokoyama et al. 1992). Such block copolymer systems are particularly interesting for the encapsulation of highly hydrophobic BCS class II and IV drugs such as paclitaxel, lactoferrin, nimodipine, doxorubicin, camptothecin, among others (Danhier et al. 2009; Hu et al. 2003, 2009; Park et al. 2009; Derakhshandeh et al. 2010). The use of biodegradable poly(ester– ether)s in drug delivery applications has been more limited. Poly(dioxanone), first commercialized by Ethicon in 1981, has been mainly used as biodegradable suture material (PDS) or has found applications in medical devices for valvular repair (Takeishi et al.

123

Page 2 of 13

1992; Yang et al. 2002; Kalangos 2005). In addition to its biocompatibility and bioabsorbability, poly(dioxanone) shows greater softness and flexibility as well as a marked hydrolysis as compared to polylactides/ glycolides due to the presence of the ether bond and the additional methylene group. As regards to drug delivery systems, Li et al. (2003) used microparticles of poly(dioxanone) and substituted polydioxanones (5-benzyloxymethyl1,4-dioxan-2-one and 5-hydroxymethyl-1,4-dioxan-2one) to encapsulate bovine serum albumin (BSA), a model protein drug. Bhattarai et al. (2007) used a system based on the combination of the commercial cationic lipid lipofectin with a novel amphiphilic triblock copolymer, poly(dioxanone-co-L-lactide)-bpoly(ethylene glycol) to form micelles for the development of an aerosol system for topical gene delivery to the lungs of mice. We recently reported on the synthesis and characterization of PEG-P(DX-co-MeDX) block copolymers with methyl dioxanone (MeDX) content ranging from 0 to 20 % (Lochee et al. 2009). In this paper, we test the efficacy of diblock and triblock PEG-based copolymers in a broad range of MeDX content for the encapsulation of anti-inflammatory ketoprofen as model drug. To the best of our knowledge, very few papers have investigated the synthesis of triblock copolymers using JeffamineÒ ED series as macroinitiator (Tian et al. 2003; Deshayes et al. 2011). We present the synthesis of diblock and triblock PEG-based copolymers of dioxanone and MeDX with focus on their solution properties as well as their drug-loading efficiencies. It is expected that increasing the amounts of MeDX would lower crystallinity of the hydrophobic core as well as biodegradability characteristics of the copolymers and consequently influence solution properties and drugloading/release behavior.

Materials and methods Solvents were purchased from Aldrich Chemicals or Fischer and were subjected to purification prior to polymerization. 1,4-Dioxan-2-one (DX) and D,L-3methyl-1,4-dioxan-2-one (MeDX) were synthesized according to procedures previously reported (Lochee et al. 2010). Tin(II) octanoate was used as received

123

J Nanopart Res (2012) 14:1168

from Aldrich. a-Methoxy-x-hydroxy poly(ethylene glycol), MPEG, of molar mass 2000 was used as received from International Laboratory, USA. JeffamineÒ ED-2003 (Mw = 2,000 g mol-1) was used as received from Huntsman Corporation, USA. Phosphate buffered saline tablets were obtained from Sigma-Aldrich. Dialysis membrane of MWCO 3500 was purchased from Spectra/ProÒ, Spectrum Laboratories, Inc., USA. Synthesis of copolymers DX and MeDX in varying ratios were polymerized in bulk using MPEG or JeffamineÒ ED-2003 as macroinitiator in the presence of Sn(Oct)2 at 80 °C. A typical copolymerization of DX and MeDX with initial monomer feed ratio of 80 mol% of DX and 20 mol% of MeDX using MPEG/Sn(Oct)2 at 80 °C is as follows: MPEG (0.4 g, 0.2 mmol) and stannous octanoate solution (100 lL of 0.1 g Sn(Oct)2 in 5 ml in toluene) were mixed by successive addition under dry argon in a glove box. A minimum volume of dry toluene was added to dissolve the MPEG/Sn(Oct)2 initiating system. The mixture was allowed to stir at 80 °C for 24 h. DX (0.816 g, 640 lL, 8 mmol) and MeDX (0.232 g, 200 lL, 2 mmol) were then added to the initiating system and copolymerization was carried out at 80 °C. After the required copolymerization time, the tube was immersed in liquid N2 and opened to air. Percentage conversion was determined by 1H NMR. The crude sample was then dissolved in chloroform followed by precipitation in diethylether. The precipitate was washed several times with diethylether before drying under vacuum at room temperature. Preparation of drug-loaded block copolymer micelles To prepare drug-loaded nanoparticles, copolymer (10 mg) and drug (4 mg) were dissolved in acetone (2 ml). Deionised water (10 ml) was added dropwise under moderate stirring to the mixture in order to induce micelle formation. The resulting solution was left to stir for 24 h and weight loss was monitored at regular intervals to ensure maximum volatilization of acetone. The solution was dialyzed against deionised water for 48 h with a dialysis membrane of MWCO 3500 and lyophilized overnight. Drug-free

J Nanopart Res (2012) 14:1168

Page 3 of 13

nanoparticles were prepared in a similar fashion but without dialysis. Drug loading content The digestion method by Genta et al. (1997) was used to determine the drug loading content. 50 mg of drugloaded nanoparticles were digested with 20 ml of a mixture of 0.1 N HCl and ethanol (1:1 v/v) for 24 h in order to break up the micelles. Then the particles were separated by centrifugation at 10,000 rpm and the drug content in the supernatant was analyzed by UV spectrometry at 260 nm against empty nanoparticles which had been prepared and digested in a similar way. The weight of drug entrapped in the nanoparticles was calculated from a calibration curve. The drug loading content was calculated from Eq. 1. Drug loading content weight of drug in nanoparticles ¼  100 weight of nanoparticles

ð1Þ

Freeze-drying and reconstitution of samples and characterization Drug-loaded and drug-free micelles (10 ml) were frozen at -196 °C and lyophilized at room temperature. Lyophilized samples were reconstituted in deionised water or 5 % dextrose solution to obtain a copolymer concentration of 1 mg/ml. The size and size distribution of micelles before freezing and after reconstitution in water or redispersion in 5 % dextrose solution were measured by dynamic light scattering (DLS).

Degradation of copolymers and copolymer micelles The in vitro degradation studies were performed at 37 °C in PBS buffer solution (pH = 7.4). 0.5 g of copolymer or copolymer micelle was measured into test tubes containing the hydrolysis medium (PBS), which was in turn placed in an oil bath at 37 °C. At the end of each week, the test tubes were removed and the contents were filtered. Based on the initial and final weights of the dry filter paper, the weight loss of the sample was calculated. The pH change of the buffer solution was also monitored using a Jenway 3010 pH meter. The experiments were done in duplicate.

In vitro release studies A weighed amount of drug-loaded nanoparticles was suspended in PBS solution (5 ml) (0.1 M, pH = 7.4) and then transferred to a dialysis bag. The dialysis bag was sealed and immersed in 100 ml of PBS at 37 °C. At predetermined intervals, 1 ml of PBS was taken from the external medium and replaced by fresh PBS. The drug concentration was determined by measuring the absorbance at 260 nm by UV. By comparing the amount of the released drug and total drug loading, cumulative releases were obtained. Characterization 1

H and 13C NMR spectra were recorded on a 250 MHz Bruker Electrospin spectrometer in CDCl3 at room temperature. SEC analysis of polymers was performed using a Polymer Standard Systems apparatus with a refractive index detector. A PSS SDV PC column 1 (8 9 50 mm dimension) and PSS Gram linear column 2 (8 9 300 mm dimension) were used at a flow rate of 1 ml/min, pressure 10 bar, temperature 23 °C and THF as eluent. DSC was carried out using a DSC 200 F3 MaiaÒ thermal analyzer at a heating rate of 10 K/min from 20 to 110 °C. SEM (Zeiss EVO 50 XVP) was employed to characterize the morphology of the copolymers. Samples were mounted on an aluminum stub and sputter coated with gold for imaging. All micrographs were collected at a potential of 10.19 kV. Nanoparticle size and shape was further explored using cryo-TEM (JEOL 1200 EX, Tokyo, Japan, 120 keV). Samples were prepared by dispersing nanoparticles in ethanol, vortexing the dispersion and then placing it on a perforated carbon film copper grid with a micro-pipette. This was followed by evaporation and viewing at room temperature. A DLS particle size analyzer (90 Plus Particle Size Analyzer, Brookhaven Instruments Corporation) was used to characterize the particle size and distribution. Analysis was conducted in aqueous solution after filtration through a 0.45-lm pore size PTFE syringe filter to remove free drug/small polymer aggregates. Each analysis was performed at 25 °C with angle detection of 90°. All the measurements were repeated three times, and the average size and size distribution were determined. Values reported are the mean diameter ± SD for two replicate samples. The UV–Vis spectra were recorded on a Biochrom Libra S22 UV/

123

Page 4 of 13

Vis Spectrophotometer at 25 °C in a glass cuvette of 1 cm optical path length. The absorbance at peak 260 nm was used for ketoprofen concentration calculations. All drug loading and drug release experiments and characterization were carried out in triplicate. Determination of CMC The critical micelle concentration (CMC) of the copolymers was determined using a particle size analyzer. For a given copolymer, a stock solution of the copolymer in water was prepared and diluted to different concentrations and then analyzed by DLS. The logarithm of the intensity of the scattered light was plotted as a function of the concentration of the copolymer solutions. Two lines were fitted on the initial and final parts of the curve. The CMC was given by the intercept of these lines (Hussain et al. 2010; Nagy et al. 2009).

Results and discussion Synthesis and characterization of MPEG-b-P(DXco-MeDX) and P(DX-co-MeDX)-b-PEG-b-P(DXco-MeDX) MPEG-b-P(DX-co-MeDX) and P(DX-co-MeDX)-bPEG-b-P(DX-co-MeDX) copolymers were synthesized by ring-opening polymerization of dioxanone and MeDX with x-methoxypoly(ethylene glycol) (MPEG) or di-amino-terminated PEG (JeffamineÒ ED-2003) as macroinitiator in the presence of tin (II) octanoate at 80 °C. At higher temperatures, depolymerization reaction increased due to closeness with the ceiling temperature of PMeDX (Tc = 105 °C), thus lowering MeDX content in the copolymer (Lochee et al. 2009). Polymerization time was optimized to achieve maximum monomer conversions and generally longer polymerization time was required as MeDX content increased. A series of copolymers in the range 0–100 mol% of DX and MeDX were thus successfully synthesized and characterized (Tables 1, 2). Analysis of the purified diblock and triblock copolymers by 1H NMR (Figs. 1, 2) confirmed the presence of DX, MeDX and PEG units characterized by peaks at 4.15 ppm (s, CH2), 1.39–1.42 ppm (d, CH3) and 3.6 ppm (s, CH2), respectively. Moreover, the spectra indicated the absence of residual

123

J Nanopart Res (2012) 14:1168

monomers. In the case of the copolymers obtained from JeffamineÒ ED-2003, 1H and 13C NMR together with COSY experiments were run for peak assignment (Fig. 2). There is a general enrichment of DX units, which is even more pronounced at high DX mole percent in the initial feed, which could be explained by the elimination of oligomers richer in MeDX (Tables 1, 2). The triblock gave much higher DX incorporation as the bi-functionalionalized JeffamineÒ ED-2003 is able to initiate the ROP of DX and MeDX at both ends. The hydrophobic block chain length has been kept comparable for both diblock and triblock copolymers. A linear relationship between percentage of DX in the feed against percentage of DX incorporated in the copolymers was found for the MPEG-b-P(DX-co-MeDX) copolymers (Fig. 3). A high initiation efficiency of the PEG macroinitiator is observed, and this is also supported by relatively low polydispersity indices of the copolymers as determined by SEC (Fig. 4). The SEC traces of the crude copolymer extracts in THF exhibited a monomodal distribution, thus ruling out the formation of homopolymers. A plot of initial DX mole percent against incorporated DX for the triblock copolymers showed a nonlinear relationship (Fig. 3). The copolymers showed enhanced solubility in common organic solvents such as CHCl3, CH3OH as the MeDX content increased. The morphology of MPEG-b-PDX, MPEG-bPMeDX, and MPEG-b-P(DX-co-MeDX) of varying ratio of DX and MeDX units were investigated by scanning electron microscopy (Fig. 5). It is apparent that as the percentage of MeDX units increases in the copolymer, phase mixing occurs with a shift from a porous crystalline to a more amorphous sheet homogeneous phase. The incorporation of MeDX units in the copolymer disrupts the ordering of the chains and reduces crystallinity. This is confirmed by DSC measurements as listed in Table 3. In the case of diblock and triblock copolymers, bimodal peaks are usually observed if the respective blocks are of comparable lengths. However, our MPEG-b-P(DXco-MeDX) and P(DX-co-MeDX)-b-PEG-b-P(DX-coMeDX) copolymers exhibited only one thermal transition. Indeed, the presence of only one thermal transition has been reported previously for triblock MPEG-PCL-MPEG (Cuong et al. 2010) and PPDOco-PCL-b-PEO-b-PPDO-co-PCL (Bahadur et al. 2007). A single peak was indicative of complete mixing of the

J Nanopart Res (2012) 14:1168

Page 5 of 13 of Sn(Oct)2 (250 lL of 0.1 g Sn(Oct)2 in 5 ml toluene), [Sn(Oct)2] = 2,000 ppm with respect to total no of moles of monomers

Table 1 Bulk copolymerization of DX/MeDX (no of moles of DX ? MeDX = 25 mmol) using MPEG (2,000 g mol-1) (1.00 g, 0.5 mmol) as macroinitiator at 80 °C in the presence Initial mole ratio of monomers (%)

Time (h)

Conversiona (%)

Mol% of units in purified copolymera

DX

MeDX

DX

MeDX

M1H n

NMR a

(g mol-1)

MSEC n

b

(g mol-1)

Ic

DX

MeDX

100

–

44

89

–

–

–

6000

–d

–d

75

25

72

72

36

84

16

5500

2500

1.19

70 60

30 40

120 168

74 78

52 53

82 74

18 26

5300 6300

2300 1400

1.23 1.24

50

50

236

72

54

60

40

5100

2100

1.47

40

60

216

69

49

45

55

4400

2400

1.18

–

100

308

46

–

–

–

4100

3200

1.25

a

Determined by 1H NMR in CDCl3

b

Crude samples were taken from the reaction mixture and analyzed by SEC before purification using THF as eluent and polystyrene as standards c

I polydispersity index

d

Not determined

presence of Sn(Oct)2 (250 lL of 0.1 g Sn(Oct)2 in 5 ml toluene), [Sn(Oct)2] = 2,000 ppm with respect to total no of moles of monomers

Table 2 Bulk copolymerization of DX/MeDX (no of moles of DX ? MeDX = 25 mmol) using JeffamineÒ ED-2003 (2,000 g mol-1) (1.00 g, 0.5 mmol) as macroinitiator at 80 °C for 46 h in the Initial mol% of monomers

Conversiona (%)

Mol% of units in purified copolymera

DX

MeDX

DX

MeDX

DX

100

–

82

–

100

0

6390

75

25

76

51

92

8

5050

50

50

75

52

80

20

5240

25

75

72

50

43

57

4280

a

M1H n

NMR a

(g mol-1)

MeDX

1

Determined by H NMR in CDCl3

polymer segments within the copolymer. Cuong et al. (2010) further attributed a single peak to a longer PEG block and shorter PCL chain length. Trimaille et al. (2006) also found that complete mixing occurred and a lone melting temperature was observed in the case of MPEG-poly(hexyl-substituted lactides) (MPEGPHLAs). The observed decrease in Tm corresponded to a less pronounced crystallinity in the PEG block indicating interaction between both polymer chains. Increasing the hydrophobic chain length of the MPEG-PHLAs leads to greater chain mobility and interaction thereby decreasing Tm. As can be noted from Table 3, the MeDX content significantly influences the melting temperature of both di and triblock copolymers. At low MeDX content, the melting temperature was close to the predominating

PDX-co-PMeDX hydrophobic segment (%DX:%Me DX = 100:0 and 84:16 in the case of diblock copolymers and 92:8 and 80:20 in the case of triblock copolymers). Increasing amount of MeDX leads to decreasing crystallinity of the hydrophobic segment till it becomes totally amorphous in which case the melting transition observed is that of the hydrophilic segment, that is, MPEG or JeffamineÒ. Tm values decrease to *25 °C, indicative of enhanced interaction between hydrophobic and hydrophilic segments. In addition, a diblock copolymer (%DX:%MeDX = 84:16, Tm = 60.8 °C) exhibits a lower Tm than a triblock (%DX: %MeDX = 80:20, Tm = 79.0 °C), again probably due to greater interaction between hydrophilic and hydrophobic chain segments.

123

Page 6 of 13 Fig. 1 1H NMR spectrum of purified MPEG44-b[Poly(DX11-co-MeDX9)] in CDCl3

J Nanopart Res (2012) 14:1168

O

g H3CO h

O

g

c

O

a

O

d O

e

n

f

m

c

b

O i

H O p

g

g

e+f

a+b

i d

f

h

CDCl3

Fig. 2 1H NMR spectrum of purified [(DX)11-co(MeDX)4]-b-PEG-b[(DX)11-co-(MeDX)4] in CDCl3 (where x = 39 and y = 4)

Solution properties and drug loading The morphology of the diblock and triblock copolymers, determined by TEM, revealed that they all exist as regular spherical micelles (Fig. 6). The CMC of the copolymers were determined in water by DLS according to the methods used previously by Hussain et al. (2010) and Nagy et al. (2009) (Fig. 7). The CMC values of the diblock and triblock

123

copolymers are listed in Table 4. The values of the orders of 10-4–10-5 M are comparable with CMC values of amphiphilic PEG-based block copolymers (Han et al. 2009; Lukyanov et al. 2002; Yan et al. 2010). Moreover, a decrease in the value of CMC was noted with increasing MeDX ratio in the copolymer. The presence of methyl groups increases the hydrophobic character of the chain and consequently CMC decreased. Trimaille et al. (2006) also observed a

Fig. 3 DX feed against incorporated DX in MPEGb-P(DX-co-MeDX) diblock copolymers (filled diamond) and P(DX-co-MeDX)-bPEG-b-P(DX-co-MeDX) triblock copolymers (filled square)

Page 7 of 13

Mole % of incorporated DX in purified copolymer

J Nanopart Res (2012) 14:1168

Mole % DX in feed Fig. 4 SEC traces of (a) MPEG, (b) MPEG-bPMeDX, (c) MPEG-b[P(DX-co-MeDX)] (DX:MeDX = 45/55), (d) MPEG-b-[P(DX-coMeDX)] (DX:MeDX = 60/ 40), (e) MPEG-b-[P(DX-coMeDX)] (DX:MeDX = 74/ 26), (f) MPEG-b-[P(DX-coMeDX)] (DX:MeDX = 82/ 18), and (g) MPEG-b[P(DX-co-MeDX)] (DX:MeDX = 84/16)

decrease in CMC upon increasing the number of hexyl groups on the polyester block of MPEG–PHLA copolymers. Such a trend has also been found previously (Lin et al. 2010). The size of the diblock copolymer micelles was found to be in the range of 230–300 nm and the triblock copolymer micelles afforded smaller sizes in the range of 120–160 nm (Table 4). Particle size of block copolymer micelles increased with weight percent of the hydrophobic segment (Qiu and Bae 2006). In general, micelle size is affected by the hydrophobic block chain length and the interactions between the various chemical groups of the polymer chains. The smaller size of the triblock is attributed to

the higher hydrophobic ratio compared to the diblock as reported previously (Lin et al. 2006). In our case, an increase in micelle size was observed as the MeDX content in the core increased for both di and triblock copolymers. The state of the micelle core is determined by the melting temperatures of the core forming block. The presence of the methyl group of the MeDX units increases the amorphous character of the hydrophobic core and thus chain mobility. As a result micelle size increased with MeDX content. The relatively large diameter of the micelles, above the expected 100 nm range, could be accounted for (i) by the aggregation of nanoparticles into clusters as

123

Page 8 of 13

J Nanopart Res (2012) 14:1168

Fig. 5 SEM micrographs: a MPEG44-b-PDX39, b MPEG-b-[P(DX-co-MeDX)], (DX:MeDX = 74/26), c MPEG-b-[P(DX-coMeDX)] (DX:MeDX = 45/55), d MPEG44-b-PMeDX18 at 5009 and 30009 magnification

explained previously for the PS/PEO diblock copolymer micellar system or for the poly(caprolactone)– poly(ethylene oxide)–polylactide nanoparticle system (Bhattarai et al. 2003; Xu et al. 1991), (ii) by the concentration dependence of the micellar solutions prepared. Bhattarai et al. (2003) noted a clear trend of

123

increasing particle size with copolymer concentration from DLS results. The copolymeric micelles were loaded with ketoprofen using the acetone volatilization technique. The loading/encapsulation of ketoprofen in the nanoparticles was confirmed by UV–visible spectroscopy. After

J Nanopart Res (2012) 14:1168 Table 3 Thermal properties of diblock and triblock copolymers

Page 9 of 13

Co/polymer

Mol% of DX:MeDX

Tm (°C)

PDX-co-PMeDX

100:0

100–111a

92:8

91b

87:13

83b

85:15

72b

–

53.3

MPEG(2000) MPEG-b-P(DX-co-MeDX)

100:0

84.1

84:16

60.8

60:40 30:70

32.6 26.5

JeffamineÒ ED-2003

–

38.3

P(DX-co-MeDX)-b-Jeffamine-b-P(DX-co-MeDX)

92:8

87.9

80:20

79.0

43:57

26.5

a

Depending on molecular weight

b

Lochee et al. (2009)

Fig. 6 TEM image of P(DX-co-MeDX)-b-PEG-b-P(DX-coMeDX) (DX:MeDX = 92/8) at 20,0009 magnification

substraction of the UV spectrum of empty copolymer micelles, the drug-loaded spectrum shows bands in the range of 238–296 nm. The peak at 260 nm corresponding to the p–p* transition of the aromatic ring of ketoprofen was used to estimate the percentage drug encapsulated (Latere Dwan’Isa et al. 2008). Drug loading content was calculated using Eq. 1. Drug loading in the range of 43–83 % and 36–46 % were obtained for the di and triblock copolymers, respectively. The higher loading efficiency was attributed to the larger micelle size of the diblock compared to the triblock. In general, drug loading was found to decrease with increasing MeDX content even though the size of micelles increased. The efficacy of drug

Fig. 7 Logarithm of light scattering intensity versus copolymer concentration for MPEG-b-P(DX-co-MeDX) (DX:MeDX = 45/55) in water

loading depends among others on the interaction between drug and the hydrophobic core. The presence of the methyl group may hinder drug from being encapsulated inside the inner core of the micelles and could account for a decreased drug loading as reported previously for PEG-b-PLA (Lin et al. 2010). We observe a decrease in micelle size upon drug entrapment for both diblock and triblock micelles. The size of polymeric micelles generally increases upon encapsulation of drug as compared to drug-free micelles (Li et al. 2009; Cuong et al. 2010). However, Zweers et al. (2006) found that the diameter of drug-loaded PEO-b-

123

Page 10 of 13

J Nanopart Res (2012) 14:1168

Table 4 CMC, drug loading content, and effective diameter of diblock and triblock copolymer nanoparticles Length of hydrophobic block

CMCa 9 10-4 (M)

Drug loading content (%)

Diameter of empty micelle (nm)a

Diameter of loaded micelle (nm)a

PDIb

0.329

MPEG-b-P(DX-co-MeDX) [(DX)26-co-(MeDX)9]

1.17

86

232 ± 8

229 ± 7

[(DX)16-co-(MeDX)11]

1.00

67

253 ± 3

261 ± 3

0.491

[(DX)11-co-(MeDX)9]

0.84

43

300 ± 3

277 ± 3

0.332

P(DX-co-MeDX)-b-PEG-b-P(DX-co-MeDX) [(DX)22]2

1.20

46

120 ± 3

72 ± 2

0.111

[(DX)11-co-(MeDX)4]2

1.15

40

130 ± 3

110 ± 3

0.141

[(DX)12-co-(MeDX)3]2

0.94

36

154 ± 3

112 ± 3

0.172

a

Each value is the mean (±SD) of three experiments (n = 3)

b

Polydispersity index of the micelles in aqueous solution

PLGA nanoparticles is smaller than unloaded ones while Sutton et al. (2007) have concluded that loading of doxorubicin into PEG-b-PCL or PEG-b-PLA micelles does not noticeably affect micelle size. The decrease in size can be accounted for by a stabilization of the polymeric micelles upon drug encapsulation.

Micelle stability Polymeric micelles in aqueous form are usually not physically stable, undergoing substantial aggregation and are not practical for storage or transportation. To improve their physical stability, one possible solution is to lyophilize polymeric micelles into powders. However, upon reconstitution of the freeze-dried samples into water, the size of the micelles is found to increase considerably. Researchers have investigated the addition of cryoprotectant such as sugars or amino acids in the aqueous medium prior to freezedrying (Di Tommaso et al. 2010; Moretton et al. 2011). In our case, the lyophilization of the polymeric nanoparticles was done without any cryoprotectant, but the redispersion of the freeze-dried samples was investigated in both water and 5 % dextrose solution. As an example, for MPEG44-bP[(DX)16-co-(MeDX)11], reconstitution in water gives rise to aggregation leading to size greater than 1000 nm, but in 5 % dextrose solution at room temperature, the size (280 nm) is only slightly larger

123

than that before freeze-drying (261 nm). Similar trends were observed for all diblock and triblock copolymers.

In vitro drug release The drug release profile was recorded for the 70:30, 60:40, and 45:55 DX:MeDX diblock copolymers in PBS at 37 °C (Fig. 8). The release profiles do not show a burst effect. Indeed, after 144 h, the % drug release lies between 30 and 58 %, which is comparable to PEG-b-PCL system where only 40 % of the anticancer agent ellipticine was released within 150 h (Liu et al. 2004). Diffusion of the drug and hydrolytic cleavage of the hydrophobic polymer segments are the dominant mechanisms for drug release as reported for PEG-b-PLAs micelles (Liu et al. 2006). We also note an increase in the release rate with increasing MeDX content. The faster degradation of the 45:55 copolymer is accounted for by an increase in the amorphous character of the micelle core resulting in enhanced hydrolysis and greater diffusion of the drug (Liu et al. 2006). Another factor that influences the release rate of PEG-b-polyester micelles is the percentage of drug loaded (Lim Soo et al. 2002; Ryu et al. 2000). In accordance with the literature, the higher the drug loading, the slower is the drug release which corresponds here to the 70:30 copolymer (Venkatraman et al. 2005).

J Nanopart Res (2012) 14:1168

Cumulative release (%)

Fig. 8 Ketoprofen release profiles from MPEG-b[P(DX-co-MeDX)], (DX:MeDX = 70/30, 60/40, and 45/55) in PBS at 37 °C

Page 11 of 13

Time (hours)

Table 5 Percentage weight loss of diblock copolymers and copolymer micelles after 7 and 14 days Copolymer DX incorporation (%)

MeDX incorporation (%)

Weight loss after 7 days (%)

Weight loss after 14 days (%)

Copolymer

Copolymer

Copolymer micelle

Copolymer micelle

100

0

10

3

18

9

70

30

28

17

77

34

50

50

33

21

80

56

30

70

35

25

84

67

Degradation of copolymers and copolymer micelles Table 5 shows the percentage of weight loss of the samples at the end of the first and second weeks. It is interesting to note that the presence of MeDX brought about a drastic decrease in weight as compared to a 100 % PDX copolymer. Weight loss results are for 7 and 14 weeks only. (We can see a loss of 28–35 % after 7 days and 70–84 % after 14 days depending on %MeDX). The formation of acidic products is known to catalyze the degradation of the polymers’ own hydrolysis products (Schindler and Pitt 1982). As is also established, drug release operates by two mechanisms: degradation of the inner micelle core and diffusion of the drug through the inner core. We have correlated the percentage weight loss of empty copolymer micelles with percentage drug release from micelles both in PBS for the same period of time. While the percentage release is in the range of 30–60 % after 7 days (Fig. 8), a weight loss of only 17–21 % is observed for the same duration. This

implies that both diffusion and degradation are operating with a predominance of the former.

Conclusions In this paper, a series of MPEG-b-P(DX-co-MeDX) and P(DX-co-MeDX)-b-PEG-b-P(DX-co-MeDX) copolymers were successfully synthesized and characterized. These copolymers self-assembled in solution to form spherical micelles with sizes ranging from 120 to 300 nm. We have shown that the crystalline-amorphous character of the hydrophobic segment of the copolymer depended on the MeDX content. As a result, micelle size increased with increasing percentage of MeDX, while drug loading was found to decrease. Higher drug loading was achieved with diblock than triblock copolymers. Drug release could also be controlled by varying the composition of the hydrophobic core. Enhanced drug release was achieved with copolymers having a larger MeDX content and for all

123

Page 12 of 13

copolymers, no burst effect was observed. The degradability of the hydrophobic core can be adjusted to maintain the drug concentration within the therapeutic window. Due to the tunable drug loading/release PEGb-poly(ester–ether)s could be interesting candidates for controlled drug delivery apart from the classical PEG-b-polyesters. Acknowledgments We thank the Tertiary Education Commission, Mauritius, for PhD fellowship to RJ. We are most grateful to Prof G Bowlin (Virginia Commonwealth University, USA) and Prof V Pillay (WITS University, South Africa) for assistance in SEM and TEM analyses.

References Bahadur RKC, Bhattarai SR, Aryal S, Khil MS, Dharmaraj N, Kim HY (2007) Novel amphiphilic triblock copolymer based on PPDO, PCL, and PEG: Synthesis, characterization, and aqueous dispersion. Colloids Surf A Physicochem Eng Asp 292:69–78. doi:10.1016/j.colsurfa.2006.06.009 Bhattarai N, Bhattarai SR, Khil MS, Lee DR, Kim HY (2003) Aqueous solution properties of amphiphilic triblock copolymer poly(p-dioxanone-co-L-lactide)-block-poly(ethylene glycol). Eur Polym J 29:1603–1608. doi:10.1016/S0014-3057 (03)00057-0 Bhattarai SR, Kim SY, Jang KY, Yi HK, Lee YH, Bhattarai N, Nam SY, Lee DY, Kim HY, Hwang PH (2007) Amphiphilic triblock copolymer poly(p-dioxanone-co-L-lactide)block-poly(ethylene glycol), enhancement of gene expression and inhibition of lung metastasis by aerosol delivery. Gene Ther 14:476–483. doi:10.1038/sj.gt.3302876 Cuong NG, Hsieh MF, Chen YT, Liau I (2010) Synthesis and characterization of PEG–PCL–PEG triblock copolymers as carriers of doxorubicin for the treatment of breast cancer. J App Polym Sci 117:3694–3703. doi:10.1002/app.32266 Danhier F, Lecouturier N, Vroman B, Je´roˆme C, MarchandBrynaert J, Feron O, Pre´at V (2009) Paclitaxel-loaded PEGylated PLGA-based nanoparticles: in vitro and in vivo evaluation. J Control Release 133:11–17. doi:10.1016/ j.jconrel.2008.09.086 Derakhshandeh K, Soheili M, Dadashzadeh S, Saghiri R (2010) Preparation and in vitro characterization of 9-nitrocamptothecin-loaded long circulating nanoparticles for delivery in cancer patients. Int J Nanomed 5:463–471. doi: 10.2147/IJN.S11586 Deshayes G, Delcourt C, Verbruggen I, Trouillet-Fonti L, Touraud F, Fleury E, Degee P, Destarac M, Willem R, Dubois P (2011) Eur Polym J 47:98–110. doi:10.1016/j.eur polymj.2010.07.005 Di Tommaso C, Como C, Gurny R, Mo¨ller M (2010) Investigations on the lyophilisation of MPEG-hexPLA micelle based pharmaceutical formulations. Eur J Pharm Sci 40: 38–47. doi:10.1016/j.ejps.2010.02.006 Genta I, Conti B, Perugini P, Pavanetto F, Sparado A, Puglisi G (1997) Bioadhesive microspheres for ophthalmic administration of acyclovir. J Pharm Pharmacol 49: 737–742. doi:10.1111/j.2042-7158.1997.tb06103.x

123

J Nanopart Res (2012) 14:1168 Han X, Liu J, Xie C, Zhan C, Gu B, Liu Y, Feng L, Lu W (2009) 9-NC-loaded folate-conjugated polymer micelles as tumor targeted drug delivery system: preparation and evaluation in vitro. Int J Pharm 372:125–131. doi:10.1016/j.ijpharm. 2008.12.035 Hu Y, Jiang X, Ding Y, Zhang L, Yang C, Zhang J, Chen J, Yang Y (2003) Preparation and drug release behaviors of nimodipineloaded poly(caprolactone)–poly(ethylene oxide)–polylactide amphiphilic copolymer nanoparticles. Biomaterials 24: 2395–2404. doi:10.1016/S0142-9612(03)00021-8 Hu K, Li J, Shen Y, Lu W, Gao X, Zhang Q, Jiang X (2009) Lactoferrin-conjugated PEG-PLA nanoparticles with improved brain delivery: In vitro and in vivo evaluations. J Control Release 134:55–61. doi:10.1016/j.jconrel.2008. 10.016 Hussain H, Tan BH, Mya K, Liu Y, He CB, Davis TP (2010) Synthesis, micelle formation, and bulk properties of poly(ethylene glycol)-b-poly(pentafluorostyrene)-g-polyhedral oligomeric silsesquioxane amphiphilic hybrid copolymers. J Polym Sci Part A 48:152–163. doi:10.1002/pola.23773 Kalangos A (2005) Polydioxanone polymers for annuloplasty in valve repair. Macromol Symp 231:81–83. doi:10.1002/masy. 200590027 Kumar N, Ravikumar MN, Domb AJ (2001) Biodegradable block copolymers. Adv Drug Del Rev 53:23–44. doi: 10.1016/S0169-409X(01)00219-8 Latere Dwan’Isa JP, Rouxhet L, Brewster ME, Pre´at V, Arie¨n A (2008) Spontaneously self-assembled micelles from poly(ethylene glycol)-b-poly(e-caprolactone-co-trimethylene carbonate) for drug solubilization. Pharmazie 63: 235–240. doi:10.1691/ph.2008.7324 Li MX, Zhuo RX, Qu FQ (2003) Study on the preparation of novel functional poly(dioxanone) and for the controlled release of protein. React Funct Polym 55:185–195. doi: 10.1016/S1381-5148(02)00246-8 Li Y, Qi XR, Maitani Y, Nagai T (2009) PEG–PLA diblock copolymer micelle-like nanoparticles as all-trans-retinoic acid carrier: in vitro and in vivo characterizations. Nanotechnology 20:1–10. doi:10.1088/0957-4484/20/5/ 055106 Lim Soo P, Luo L, Maysinger D, Eisenberg A (2002) Incorporation and release of hydrophobic probes in biocompatible polycaprolactone-block-poly (ethylene oxide) micelles: implications for drug delivery. Langmuir 18:9996–10004. doi:10.1021/la026339b Lin WJ, Juang LW, Wang CL, Juang LW (2006) Characterization and comparison of diblock and triblock amphiphilic copolymers of poly(delta-valerolactone). J Appl Polym Sci 100:1831–1836. doi:10.1002/app.22580 Lin WJ, Juang LW, Wang CL, Chen YC, Lin CC, Chang KL (2010) Pegylated polyester polymeric micelles as a nanocarrier: synthesis, characterization, degradation, and biodistribution. J Exp Clin Med 2:4–10. doi:10.1016/S18783317(10)60002-2 Liu J, Xiao Y, Allen C (2004) Polymer-drug compatibility: a guide to the development of delivery systems for the anticancer agent, ellipticine. J Pharm Sci 93:132–143. doi: 10.1002/jps.10533 Liu J, Lee H, Allen C (2006) Formulation of drugs in block copolymer micelles: drug loading and release. Curr Pharm Des 12:4685–4701. doi:10.2174/138161206779026263

J Nanopart Res (2012) 14:1168 Lochee Y, Bhaw-Luximon A, Jhurry D, Kalangos A (2009) Novel biodegradable poly(ester–ether)s: copolymers from 1,4-dioxan-2-one and D,L-3-methyl-1,4-dioxan-2-one. Macromolecules 42:7285–7291. doi:10.1021/ma900747q Lochee Y, Jhurry D, Bhaw-Luximon A, Kalangos A (2010) Biodegradable poly(ester–ether)s: ring-opening polymerization of D,L-3-methyl-1,4-dioxan-2-one using various initiator systems. Polym Int 59:1310–1318. doi:10.1002/pi.2871 Lukyanov AN, Gao Z, Mazzola L, Torchilin VP (2002) Polyethylene glycol-diacyllipid micelles demonstrate increased acculumation in subcutaneous tumors in mice. Pharm Res 19:1424–1429. doi:10.1023/A:1020488012264 Moretton MA, Chiappetta DA, Sosnik A (2011) Cryoprotectionlyophilization and physical stabilization of rifampicinloaded flower-like polymeric micelles. J R Soc Interface 9:487–502. doi:10.1098/rsif.2011.0414 Nagy M, Szo¨llo¨si L, Ke´ki S, Faust R, Zsuga M (2009) Poly (vinyl alcohol)-based amphiphilic copolymer aggregates as drug carrying nanoparticles. J Macromol Sci Part A 46:331–338. doi:10.1080/10601320902719370 Nishiyama N, Kataoka K (2006) Current state, achievements, and future prospects of polymeric micelles as nanocarriers for drug and gene delivery. Pharmacol Ther 112:630–648. doi:10.1016/j.pharmthera.2006.05.006 Owens DE, III Peppas NA (2006) Opsonization, biodistribution, and pharmacokinetics of polymeric nanoparticles. Int J Pharm 307:93–102. doi:10.1016/j.ijpharm.2005.10.010 Park J, Fong PM, Lu J, Russell KS, Booth CJ, Saltzmann WM, Fahmy TM (2009) PEGylated PLGA nanoparticles for the improved delivery of doxorubicin. Nanomedicine 5: 410–418. doi:10.1016/j.nano.2009.02.002 Qiu LY, Bae YH (2006) Polymer architecture and drug delivery. Pharm Res 23:1–30. doi:10.1007/s11095-005-9046-2 Ryu J, Jeong YI, Kim IS, Lee JH, Nah JW, Kim SH (2000) Clonazepam release from core-shell type nanoparticles of poly(epsilon-caprolactone)/poly(ethylene glycol)/poly (epsilon-caprolactone) triblock copolymers. Int J Pharm 200:231–242. doi:10.1016/S0378-5173(00)00392-6 Schindler A, Pitt CG (1982) Biodegradable elastomeric polyesters. Polym Prep Am Chem Soc 23:111–112

Page 13 of 13 Sutton D, Wang S, Nasongkla N, Gao J, Dormidontova EE (2007) Doxorubicin and b-lapachone release and interaction with micellar core materials: experiment and modeling. Exp Biol Med 232:1090–1099. doi:10.3181/0702-RM-31 Takeishi M, Hirase Y, Kojima T (1992) Microsurgical use of polydioxanone (PDS) suture: an experimental report. Microsurgery 13:268–272. doi:10.1002/micr.1920130515 Tian W, Chen Q, Yu C, Shen J (2003) Amino-terminated poly(ethylene glycol) as the initiator for the ring-opening polymerization of 3-methylmorpholine-2,5-dione. Europ Polym J 39:1935–1938. doi:10.1016/S0014-3057(03) 00091-0 Trimaille T, Mondon K, Gurny R, Moller M (2006) Novel polymeric micelles for hydrophobic drug delivery based on biodegradable poly(hexyl-substituted lactides). Int J Pharm 319:147–154. doi:10.1016/j.ijpharm.2006.03.036 Venkatraman SS, Jie P, Min F, Freddy BYC, Leong-Huat G (2005) Micelle-like nanoparticles of PLA–PEG–PLA triblock copolymer as chemotherapeutic carrier. Int J Pharm 298:219–223. doi:10.1016/j.ijpharm.2005.03.023 Xu RL, Winnik MA, Hallett FR, Gerrard R, Croucher MD (1991) Light-scattering study of the association behavior of styrene-ethylene oxide block copolymers in aqueous solution. Macromolecules 24:87–93. doi:10.1021/ma0000 1a014 Yan Q, Yuan J, Kang Y, Yin Y (2010) Dynamic supramacromolecular self-assembly: deformable polymer fabricated nanostructures through a host-controlled approach. Polym Chem 1:423–425. doi:10.1039/C0PY00024H Yang KK, Wang XL, Wang YZ (2002) Poly(p-dioxanone) and its copolymers. J Macromol Sci Part C 42:373–398. doi: 10.1081/MC-120006453 Yokoyama M, Kwon GS, Okano T, Sakurai Y, Seto Y, Kataoka K (1992) Preparation of micelle-forming polymer–drug conjugates. Bioconjug Chem 2:295–301. doi:10.1021/ bc00016a007 Zweers ML, Engbers GH, Grijpma DW, Feijen J (2006) Release of anti-restenosis drugs from poly (ethylene oxide)poly(DL-lactic-co-glycolic acid) nanoparticles. J Control Release 114:317–324. doi:10.1016/j.jconrel.2006.05.021

123