Journal of Colloid and Interface Science 356 (2011) 100–106

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science

www.elsevier .com/locate / jc is

pH sensitive swelling and releasing behavior of nano-gels basedon polyaspartamide graft copolymers

Sunmi Kim, Ji-Heung Kim, Dukjoon Kim ⇑School of Chemical Engineering, Theranostic Macromolecules Research Center, Sungkyunkwan University, Suwon, Kyunggi 440-746, Republic of Korea

a r t i c l e i n f o

Article history:Received 2 November 2010Accepted 2 January 2011Available online 7 January 2011

Keywords:Nano-gelGraft copolymerpH sensitiveSwellingPolyaspartamideDrug release

0021-9797/$ - see front matter � 2011 Elsevier Inc. Adoi:10.1016/j.jcis.2011.01.003

⇑ Corresponding author. Fax: +82 31 290 7270.E-mail address: djkim@skku.edu (D. Kim).

a b s t r a c t

pH sensitive nano-gels based on polyaspartamide graft copolymers are prepared by UV crosslinking theself-assembled nano-aggregates in the presence of a series of hydrophobic and hydrophilic grafting seg-ments. While the physical nano-aggregates dissociate via ionization of the pH sensitive moiety, the nano-gels synthesized in this study swell instead. The chemical structure and morphology of the resultingnano-gels were analyzed using FTIR, 1H FTNMR, and TEM. The pH dependence of the particle size,120–250 nm, was investigated using a light scattering analyzer. The swelling behavior of the nano-coresunder acidic conditions triggered abrupt release of the drug; this pH dependence occurs reversibly andquickly. The nano-gels prepared may have endosomal rupturing characteristics, as their buffering capac-ity is as strong as that of uncrosslinked nano-aggregates. The nano-gels synthesized as such possesspotential application as sustained releasing drug carriers for intracellular delivery.

� 2011 Elsevier Inc. All rights reserved.

1. Introduction the external stimuli can control the on–off of drug release in a sus-

The introduction of nanotechnology into drug delivery systemsenables the development of a variety of nano-scaled diagnostic andtherapeutic systems in drug carriers, diagnostic reagents, and bio-medical sensors to maximize efficacy and minimize side effects [1–3]. In the case of drug carriers, it is often aimed at nano-particles orcapsules containing drug or bioactive molecules for selective trans-portation to a targeted site. Recently, great attention has been paidto the targeted intracellular drug delivery by nano-carriers to en-hance overall therapeutic effects. A number of bio- or syntheticpolymers have been studied for the preparation and applicationof suitable nano-carriers, possessing select prerequisite propertiesof biocompatibility, biodegradability, and chemical and biologicalnon-toxicities [4–18].

‘‘Nano-gels’’ are a few to hundreds of nanometer sized particlesor capsules with the characteristics of a hydrogel. Nano-drugcarriers are usually composed of a simple hydrophobic core andhydrophilic shell structure, such as self-assembled micelles ornano-aggregates. The micelles or nano-aggregates, self-assembledin a physical manner, generally do not establish sustained releasebecause of a high-rate release pattern from the extremely smallreservoir when the core structure is dissociated. Furthermore, thissituation is noticeable for the so-called stimuli sensitive carriersystems where drug release is triggered by changes in environ-mental conditions. For micelles or uncrosslinked nano-aggregates,

ll rights reserved.

tained fashion, but not the drug releasing rate.The pH sensitive polymers have been applied to a variety of

drug releasing systems as their drug releasing pattern is easily con-trolled by the pH change associated with the biological activitychange inside the body. One of the important applications in thepH sensitive nano-carrier is cancer targeted drug delivery, wheretargeting is accomplished by enhanced permeation and retention(EPR). It was recently reported that novel graft copolymers basedon aspartamide derivatives demonstrated pH sensitive behavior.The nano-aggregates were formed via self-assembly of the poly-meric systems with a series of hydrophobic and hydrophilicgrafting segments. Histidine is a well-known amino acid with apH sensitivity such that it is cationic under weak acidic conditions.As the imidazole group within the histidine has a pKa value near6.5, it has a hemolytic characteristic for endosomes and cancer cellmembranes. 1-(3-Aminopropyl)imidazole (API) has a chemicalstructure and properties very similar to histidine, with terminal,primary amine groups that allow for facile grafting to polysuccin-imide (PSI). Also, as it is more hydrophobic than histidine,formation of the hydrophobic core structure is more feasible byself-aggregation at a neutral pH. Introduction of polyethyleneglycol (PEG) into PSI prevents adsorption of protein, leading tolong-term circulation in blood vessels [19–23].

As previous studies focused mainly upon the pH sensitive phasetransition behavior of uncrosslinked self-aggregates of polyaspar-tamide derivatives, this study aims at the synthesis of nano-gelsby crosslinking aggregate cores to investigate their swelling anddrug releasing behaviors. The results were compared with uncross-linked systems.

S. Kim et al. / Journal of Colloid and Interface Science 356 (2011) 100–106 101

2. Materials and methods

2.1. Materials

L-Aspartic acid was purchased from Sigma (St. Louis, MO,USA). Phosphoric acid (85%, Aldrich, Milwaukee, WI, USA) andtriethylamine (TEA, Aldrich) were used as catalysts. N,N-dimeth-ylformamide (DMF, Aldrich), mesitylene (Aldrich), sulforane(Aldrich), N,N-dimethylacetamide (DMA, Aldrich), methanol(Daejung), and dimethyl sulfoxide (DMSO, Aldrich) were usedas solvents. Phosphate buffered saline (PBS pH 7.4, Aldrich)was used as a buffer solution. The following reagents were usedfor graft segments onto PSI: 1-(3-aminopropyl)imidazole (API,Aldrich) for the pH sensitive moiety; O-(2-aminoethyl)-O0-meth-ylpolyethylene glycol (MPEG, MW 5000, Fluka) for the hydro-philic segment; octadecylamine (C18, Aldrich) for thehydrophobic segment; methacrylic anhydride (MA, Aldrich) forthe crosslinking agent; ethanolamine (Aldrich) for the reactivefunctional group, respectively.

2.2. Synthesis of octadecylamine/O-(2-aminoethyl)-O0-methylpoly-ethylene glycol/1-(3-aminopropyl) imidazol/methacrylicanhydride-g-polyaspartamide (C18/MPEG/API/MA-g-PASPAM)

Polysuccinimide (PSI) was synthesized from L-aspartic acid inthe presence of an acid catalyst by the condensation polymeriza-tion method [24]. L-Aspartic acid (25.0 g) and phosphoric acid(15 mmol) were placed in a co-solvent composed of 70.0 g ofmesitylene and 30.0 g of sulforane. The polymerization reactionwas conducted at 170 �C for 8 h under a nitrogen gas environment.The condensed water was removed using a Dean–Stark trap. Theproduct was precipitated in methanol and then filtrated. The fil-trate was washed with distilled water several times and dried invacuum oven at 70 �C for 24 h.

The PSI (10.0 g) was dissolved in DMF (90 mL). To this solutionwas added a desired amount of C18, and then the resulting mixturestirred for 24 h at 70 �C under a nitrogen atmosphere. The finalproduct (C18-g-PSI) was precipitated out in methanol and the pre-cipitate filtrated and washed with distilled water before drying in avacuum oven at 70 �C for 24 h.

O-(2-Aminoethyl)-O0-methylpolyethylene glycol-g-poly(suc-cinimide) (C18-MPEG-g-PSI) copolymer was synthesized byintroducing MPEG into PSI. The introductory amount of MPEGwas 2 mol% to the repeating PSI unit. The MEPG solution inDMF was slowly added to the PSI solution (2.0 g PSI in 10 mLDMF) at 0 �C and then reacted for 48 h at 70 �C in a nitrogenatmosphere. A desired amount of API was then added to thepolymer solution and stirred for 48 h under the same conditions.Excess amounts of ethanolamine were introduced to substitutethe residual end groups with hydroxyl groups. The productwas dialyzed for a week against distilled water using a mem-brane (molecular weight cut-off = 10,000–12,000 g mol�1) to re-move residual MPEG, API, and ethanolamine; the product wasthen freeze-dried.

To the intermediate C18/MPEG/API-g-PASPAM product wasintroduced methacrylic anhydride (MA) to produce C18/MPEG/API/MA-g-PASPAM. The synthesized C18/MPEG/API-g-PASPAMcopolymer (1.0 g) was then dissolved in DMA (20 mL). The MAand TEA were added in a mol ratio of 2 to the C18/MPEG/API-g-PASPAM copolymer solution. The amount of MA added was at amol ratio of 0.5 to PHEA. For grafting, the mixture was reacted at40 �C for 48 h under nitrogen gas. The final product was obtainedafter 4 d of dialysis, followed by freeze-frying. The syntheticscheme for the C18/MPEG/API/MA-g-PASPAM copolymer is shownin Fig. 1.

2.3. Preparation of nano-gels

C18/MPEG/API/MA-g-PASPAM, 0.05 g, was dispersed in 20 mLdistilled water to prepare the nano-aggregates by self-assembling.Ultrasonic treatment (Branson Ultrasonics, Sonifier 250, Danbury,CT, USA) was applied when necessary. The aggregate cores werecrosslinked by UV radiation to produce a gel structure under con-tinuous stirring in the presence of argon gas. This was followed by4 d of dialysis and freeze-drying. Fig. 2 shows the schematics of thenano-gel preparation procedure.

2.4. Molecular weight

An Ubbelohde viscometer was used to measure the viscosity ofthe polymer solution prepared in DMF at a concentration of0.5 g dL�1. The viscosity measured at 25 �C was used to calculatethe viscosity average molecular weight using empirical Eq. (1), pro-vided by Neri et al. [25]:

n ¼ 3:52� g1:56r ð1Þ

where n and gr indicate the degree of polymerization and reducedviscosity of the polymer solution, respectively.

2.5. Chemical structure

The chemical structures of the polyaspartamide derivatives syn-thesized were identified using Fourier transform nuclear magneticresonance spectroscopy (FTNMR, Uniy Inova 500, Varian, SantaClara, CA, USA) and Fourier transform infrared spectroscopy (FTIR,Perkin Elmer Model SPECTRUM 2000, Boston, MA, USA). For the 1HNMR experiments, the samples were prepared by dissolving thepolymers in DMSO4-d6. For the FTIR experiments, the copolymersynthesized was mixed with KBr at a 1:100 weight ratio.

2.6. Particle size and shape

Dynamic light scattering analysis (ELS-Z, OTSUKA, Japan) wasemployed to measure the size distribution and mean size of thenano-particles dispersed in water or phosphate buffer solution atpH 7.4. Prior to measurement, the samples were ultrasonicatedfor good dispersion.

Transmission electron microscopy (HR-TEM, JEOL 300 kV,JEM2100F, Japan) was employed to analyze the morphology, size,and shape of the prepared self-aggregates. A polymer aqueoussolution (1.0 mg mL�1) was dropped on a copper-coated TEM grid.After dye treatment, a microscopic image was taken at room tem-perature after drying.

2.7. Buffering test

Acid–base titration was used to investigate the pH buffering ef-fects of the polyaspartamide copolymers. The sample (20.0 mg)was dissolved in aqueous 150 mM NaCl (35 mL). After addition of100 lL of aqueous 1.0 N NaOH, the basic solution was titrated withaqueous 0.1 N HCl. The pH of the titrated solution was measuredusing a pH meter (Orion 430A, Korea).

2.8. Cell viability test

The 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bro-mide (MTT) method was applied to test cell toxicity of the polymeraggregates. The L929 cells were cultivated in a flask at 37 �C in a 5%CO2 bath (Forma, Costa Mesa, CA, USA). For cell cultivation, Dul-becco’s modification of Eagle’s Medium (DMEM) was applied,including 10% FBS and 1% penicillin/streptomycin (100 unit mL�1).

Fig. 1. Synthetic scheme of C18/MPEG/API/MA-g-PASPAM.

Fig. 2. Crosslinking of C18/MPEG/API/MA-g-PASPAM.

102 S. Kim et al. / Journal of Colloid and Interface Science 356 (2011) 100–106

The 292 cells (2.0 � 105 L) were dispersed and cultivated in a 24-well dish for 24 h, and the polymers added for further cultivationfor 48 h. To 1.0 mL cultivated medium was added 100 lg of a

MTT solution (5.0 mg mL�1). After 4 h, the cultivate solution wasdiscarded and the formazan crystals inside the cells dissolved byaddition of DMSO (200 lg) to each well. An ELISA plate reader

Fig. 3. 1H NMR spectra of: (a) PSI and (b) C18/MPEG/API/MA-g-PASPAM.

S. Kim et al. / Journal of Colloid and Interface Science 356 (2011) 100–106 103

(Bio-Rad Laboratories, Hercules, CA, USA) was used to measure theabsorption at 570 nm. The average value of three measurementswas taken for determination.

2.9. Drug loading and releasing experiment

The solubility of PTX in an aqueous N,N-diethylnicotinamide(DENA) solution was much higher than acetonitrile or aqueous eth-anol solution. For example, its solubility was near 1.0 mg mL�1 inaqueous 2.0 M DENA [26,27]. In this study, the 0.2 M DENA PBSsolution (solubility 0.001 mg mL�1) was prepared and used forexternal releasing media for drug release experiments.

Table 1Substitution degree (SD) of C18, MPEG, API, and MA in C18/MPEG/API/MA-g-PASPAMs.

Sample name Molar feed ratio ofC18a

SD of C18b

(%)Molar feed ratio ofMPEGa

SD o(%)

C18/MPEG/API/MA100

0.1/1.0 8 0.02/1.0 1.5

C18/MPEG/API/MA200

0.1/1.0 8 0.02/1.0 1.5

C18/MPEG/API/MA300

0.1/1.0 8 0.02/1.0 1.5

a Feed C18, MPEG, API, and MA mole/succinimide unit mole.b SD (mol%) was determined by 1H NMR.

In order to investigate the crosslinking effects on the releasingbehavior, two types of samples were used in the releasing experi-ment, uncrosslinked C10-MPEG2-API300-MA and crosslinked C10-MPEG2-API300-MA nano-aggregates. The loading amount of thePTX in the polymer sample was 50%; the releasing experimentswere conducted at 37 �C and pH 7.4 or 5.0. The PTX release contentwas determined using high-performance liquid chromatography(HPLC, Waters, Burnsville, MN, USA) each time. UV/Vis detectionat 227 nm was analyzed using a reversed-phase column (ZORBAX300SB-C18). All PTX concentrations were obtained using a calibra-tion curve prepared for PTX dissolved in the DENA aqueoussolution.

3. Results and discussion

3.1. Synthesis and chemical identification of the C18/PEG/API/MA-g-PASPAM copolymer

The PSI was synthesized from L-lactic acid by a condensationreaction. As the reduced viscosity of the PSI synthesized was26.9, the molecular weight of the PSI determined from Eq. (1)was 59,000 g mol�1.

Fig. 3a and b shows the 1H NMR spectra of PSI and the synthe-sized C18/PEG/API/MA-g-PASPAM copolymer, respectively. In the1H NMR spectrum of PSI, the methine and methylene protons inPSI were well illustrated at 5.1–5.3 and 3.2–3.5 ppm, and 2.4–2.5 ppm, respectively (Fig. 3a). The chemical structure of the C18/PEG/API/MA-g-PASPAM copolymer is also well identified inFig. 3b. Each proton peak was correspondingly assigned to the pro-ton located in each segment of the C18/PEG/API/MA-g-PASPAMcopolymer molecule. As the characteristic proton peaks of theC18, MPEG, API, and MA graft groups were clearly illustrated inthe NMR spectra, the grafting reaction was assured to be well con-ducted. Calculation of proton peak intensities in the C18, MPEG, AP,and MA segments, taking into account the ratio of each to that ofthe main chain, gave rise to the substitution degree (SD) of eachgraft segment, with the results summarized in Table 1. As shownin Table 1, the experimentally measured SD of each functionalgroup was slightly smaller than the theoretical (ideal); the exper-imental SD of C18 was 8 mol% when fed at 10 mol% to the PSIrepeating unit.

3.2. Particle size and shape

The copolymers were self-aggregated to form a particulatestructure in pH 7.4 PBS solution. The pH dependence of the phasetransition and the mean particle size of the C18/MPEG/API/MA-g-PASPAM copolymers with different API content are shown inFig. 4a and b. As the pKa value of the imidazole group in API is6.7, the copolymer is ionized below that point. As the pH depen-dence of this ionization was very sensitive, its phase transition oc-curred very quickly, even at a slight pH change near that value.

f MPEGb Molar feed ratio ofAPIa

SD of APIb

(%)Molar feed ratio ofMAa

SD of MAb

(%)

1.0/1.0 77 0.5/1.0 13.5

2.0/1.0 85 0.5/1.0 5.5

3.0/1.0 90 0.5/1.0 0.5

0

50

100

150

200

250

Parti

cle

size

(nm

)

pH

uncrosslinked C10-PEG2-API100-MA uncrosslinked C10-PEG2-API200-MA uncrosslinked C10-PEG2-API300-MA

2 3 4 5 6 7 8 9 10

2 3 4 5 6 7 8 9 100

100

150

200

250

300

Parti

cle

size

(nm

)

pH

crosslinked C10-PEG2-API100-MA crosslinked C10-PEG2-API200-MA crosslinked C10-PEG2-API300-MA

(a)

(b)

Fig. 4. pH dependent mean diameter of crosslinked and uncrosslinked aggregatesprepared from: (a) uncrosslinked C18/MPEG/API-g-PASPAM copolymer and (b)crosslinked C18/MPEG/API-g-PASPAM copolymer, respectively.

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75150

160

170

180

190

200

210

220

230

240

250

Mea

n pa

rticl

e si

ze (n

m)

Time (min)

Fig. 5. pH sensitive reversible swelling and deswelling behavior of C18/MPEG/API/MA-g-PASPAM.

104 S. Kim et al. / Journal of Colloid and Interface Science 356 (2011) 100–106

Because of this ionization, the uncrosslinked C18/MPEG/API-g-PASPAM copolymer was dissolved completely in buffer mediumbelow pH 6.7, but formed nano-aggregates of 150–200 nm com-posed of C18 hydrophobic cores and MPEG hydrophilic shellsabove pH 6.7. This ionization at the low pH occurred betweenthe ionic repulsion among the imidazole groups, which are muchstronger than the hydrophobic interactions among the C18 groups.

Unlike the uncrosslinked copolymers, no dissociation was ob-served below pH 6.7, even down to pH 2.0 for nano-gels, due tothe chemically crosslinked core structure. The mean diameter ofthese gels at pH 7.4 decreased from 200 to 150 nm with increasingAPI content, because the core gel structure becomes denser withincreasing hydrophobicity. The mean particle size of the cross-linked nano-gel was nearly the same as that of the uncrossed gelat pH 7.4. The mean diameter of the nano-gels abruptly increasedwhen the pH changed below pH 6.7, because of swelling of thecores associated with the ionization of the API segments, whichis a significant difference from those uncrosslinked. This swellingratio change was more prominent when greater levels of pH sensi-tive moieties (API) were involved, with nearly 20% for the C18/MPEG/API100-g-PASPAM system, 35% for the C18/MPEG/API200-g-PASPAM system, and 45% for the C18/MPEG/API300-g-PASPAMsystem, respectively.

Fig. 5 shows the swelling/deswelling behavior of the C18/MPEG/API-g-PASPAM system triggered by pH changes. The sam-ples were reversibly swollen and de-swollen by lowering and

increasing the pH below and above 6.7. The fast swelling/deswell-ing kinetics ensured the high pH sensitivity of the present copoly-mer systems.

Fig. 6 shows the TEM microphotograph of the C10-PEG2-API300-MA system. The particles are sized from 150 to 200 nmand spherical in shape.

3.3. Buffering effect

Acid–base titration of C18/MPEG/API/MA-g-PASPAM nano-gelsresulted in the pH buffering behavior shown in Fig. 7, in whichthe buffering effect was more profound when higher levels of APIpH sensitive moieties were introduced. As the pH buffering prop-erty was well illustrated around pH 6.7, this polymer has the intra-cellular delivery application associated with endosomal rupturingcharacteristics.

3.4. In vitro cell viability

Nearly 60% of the cells were viable for PEI (M.W. 25,000 Da) at10.0 lg lL�1. Cell survival% is illustrated in Fig. 8 for the two drugloading systems, Taxol and polymeric type. In the first type, thePTX was dissolved in a Cremophor El and ethanol mixture (1:1(v/v)); for the second, the PTX was loaded into the crosslinkedC10-MPEG2-API300-MA nano-aggregates. As the PTX concentra-tion increased, the cell toxicity increased in both cases, but the cellviability was a little higher for the polymeric system than for theTaxol system. From these behaviors, it was concluded that cell tox-icity evolved as the PTX was released into the surrounding media.As approximately 60% of the cells were viable for PEI (M.W.25,000 Da) at 10.0 lg lL�1, the cell viability of the present poly-meric system seemed better than that of a typical PEI system.

3.5. In vitro PTX releasing behavior

Fig. 9a and b show the drug releasing behavior of the uncross-linked and crosslinked polymeric systems. For the uncrosslinkedC10-MPEG2-API300-MA system, approximately 30% of the drugwas released in 5 h, followed by slow release at pH 7.4. At pH 5,however, the releasing rate was so high that nearly all drug inthe carriers were released within 5 h, as the core parts were dis-solved by ionization of the API imidazole group. For the crosslinked

Fig. 6. TEM images of crosslinked C18/MPEG/API/MA-g-PASPAM.

0 500 1000 1500 2000 25000

2

4

6

8

10

12

14

pH

Consumed 0.1N HCl solution

NaCl only crosslinked C10-PEG2-API300 crosslinked C10-PEG2-API200 crosslinked C10-PEG2-API100

Fig. 7. Acid–base titration curves for crosslinked C18/MPEG/API/MA-g-PASPAMsystems.

Fig. 8. In vitro cytotoxicity of PTX in cremophor EL/ethanol solution (1:1 v:v) andPTX-loaded crosslinked C18/MPEG/API/MA-g-PASPAM.

Fig. 9. pH dependent cumulative PTX release patterns for: (a) uncrosslinked C10-PEG2-API300-MA and (b) and crosslinked C10-PEG2-API300-MA, respectively.

S. Kim et al. / Journal of Colloid and Interface Science 356 (2011) 100–106 105

system, the nano-gels maintained their aggregate structure, evenat an acidic pH, resulting in different releasing patterns. Ionizationof the API group led to swelling of the nano-gels and thus increasedthe releasing rate at pH 5. Compared to uncrosslinked samples, amuch more sustained release pattern was observed, even at pH 5.

4. Conclusions

pH sensitive polyaspartamide derivatives were synthesize bygrafting C18, MPEG, and API onto PSI. Introduction of a MA groupenabled nano-gel preparation by crosslinking cores dispersed inaqueous media. The nano-gels prepared from C18/MPEG/API/MA-g-PASPAM showed a buffering effect within a pH 5–7 range by ion-ization of the imidazole group in the API component. Uncross-linked C18/MPEG/API/MA-g-PASPAM systems showed verysensitive self-assembling and dissembling behavior according to

106 S. Kim et al. / Journal of Colloid and Interface Science 356 (2011) 100–106

variations of pH near 6.7. Above pH 6.7, 150–250 nm sized aggre-gates were formed according to the content of the API component,but easily dissociated below it. C18/MPEG/API/MA-g-PASPAMnano-gels, however, showed very stable shapes, even below pH2, due to the crosslinked core structure. Those nano-gels werenot dissociated, but swollen when the pH was reduced below6.7; their swelling and deswelling behavior was reversible and fast.In association with such dissociation and swelling behavior be-tween uncrosslinked and crosslinked systems, the uncrosslinkedC18/MPEG/API-g-PASPAM systems showed very fast drug releasingpatterns below pH 6.7, but those crosslinked sustained a releasingpattern. The nano-gels prepared in this study are biocompatible,biodegradable, and highly pH sensitive, and thus those are ex-pected to be useful as novel drug delivery carriers for intracellularapplications.

Acknowledgment

This work was supported by Basic Science Research Programthrough the National Research Foundation of Korea (NRF) Grantfunded by the Korea government (MEST) (2010-0027955).

References

[1] D. Drummond, M. Zignani, J. Leroux, Prog. Lipid Res. 39 (2000) 409.[2] M. Durfresne, D. Garrec, V. Sant, J. Leroux, M. Ranger, Int. J. Pharm. 277 (2004)

81.

[3] O. Gerasimov, J. Boomer, M. Qualls, D. Thompson, Adv. Drug Delivery Rev. 38(1999) 317.

[4] G.S. Khang, M.S. Kim, S.H. Cho, H.B. Lee, J.H. Chang, K.J. Kim, Polym. Sci.Technol. 13 (2003) 431.

[5] I.C. Kwon, K.M. Kim, S.W. Kim, H.S. Chung, S.Y. Jeong, Polym. Sci. Technol. 15(2004) 396.

[6] N. Nishiyama, K. Kataoka, Pharmacol. Ther. 112 (2006) 630.[7] N. Rapoport, Prog. Polym. Sci. 32 (2007) 962.[8] R. Savic, L. Luo, A. Eisenberg, D. Maysinger, Science 300 (2003) 615.[9] A. Kabanov, V. Alakhov, Crit. Rev. Ther. Drug 19 (2002) 1.

[10] R. Savic, A. Eisenberg, D. Maysinger, J. Drug Targeting 14 (2006) 343.[11] K. Kataoka, A. Harada, Y. Nagasaki, Adv. Drug Delivery Rev. 47 (2001) 113.[12] S.M. Moghimi, A.C. Hunter, J.C. Murray, Pharmacol. Rev. 53 (2001) 283.[13] Y. Bae, N. Nishiyama, S. Fukushima, H. Koyama, M. Yasuhiro, K. Kataoka,

Bioconjugate Chem. 16 (2005) 122.[14] K.Y. Lee, J.-H. Kim, I.C. Kwon, S.Y. Jeong, Colloid Polym. Sci. 278 (2000) 1216.[15] Y. Bae, S. Fukushima, A. Harada, K. Kataoka, Angew. Chem. Int. Ed. 42 (2003)

4640.[16] J. Panyam, V. Labhasetwar, Mol. Pharm. 1 (2004) 77.[17] K. Kataoka, T. Matsumoto, M. Yokoyama, T. Okano, Y. Sakurai, S. Fukushima, K.

Okamoto, G.S. Kwon, J. Controlled Release 64 (2000) 143.[18] Y. Zhang, J. Zhang, J. Colloid Interface Sci. 283 (2005) 352.[19] P. Midoux, A. Kichler, V. Boutin, J. Maurizot, M. Monsigny, Bioconjugate Chem.

9 (1998) 260.[20] P. Midoux, M. Monsigny, Bioconjugate Chem. 10 (1999) 406.[21] C. Pichon, C. Goncalves, P. Midoux, Adv. Drug Delivery Rev. 53 (2001) 75.[22] D. Pack, D. Putnam, R. Langer, Biotechnol. Bioeng. 67 (2000) 217.[23] D. Putnam, A. Zeliikin, V. Izumrudov, R. Langer, Biomaterials 24 (2003) 4425.[24] M. Tomida, T. Nakato, Polymer 38 (1997) 4733.[25] P. Neri, G. Antoni, F. Benvenuti, F. Colola, G. Gazzei, J. Med. Chem. 16 (1972)

893.[26] N. Baek, J. Lee, K. Park, J. Biomater. Sci. Polym. Ed. 15 (2004) 527.[27] S.C. Kim, J. Yu, J.W. Lee, E.-S. Park, S.-C. Chi, J. Pharm. Biomed. Anal. 39 (2005)

170.