微包纳

Controlled release of protein drugs from newly developed amphiphilic polymer-based microparticles composed of nanoparticlesYoshinori Kakizawa,Reiji Nishio,Taisuke Hirano,Yoichiro Koshi,Mio Nukiwa,Masakazu Koiwa,Junji Michizoe,Nobuo Ida ⁎New Frontiers Research Laboratories,Toray Industries,Inc.,6-10-1Tebiro,Kamakura,Kanagawa 248-8555,Japana b s t r a c ta r t i c l e i n f o Article history:Received 15June 2009Accepted 26September 2009Available online 3October 2009Keywords:Microparticle NanoparticleSustained releaseHuman growth hormone Protein drugA novel formulation of biodegradable microparticles was developed for the sustained release of peptide and protein drugs.The microparticles were formed by the aggregation of protein nanoparticles through water-in-oil (W/O)emulsion-lyophilization and subsequent solid-in-oil-in-water (S/O/W)emulsion –solvent evaporation.Amphiphilic copolymers were used as an emulsi fier in the W/O emulsion and matrix of the microparticles.Among the various copolymers investigated,poly(lactide-co-glycolide)-grafted dextran (Dex-g-PLGA)was chosen as the best candidate on the basis of the encapsulation ef ficiency and in vitro release pro file,the near zero-order release without a signi ficant initial burst,of human growth hormone (hGH).The release rate of hGH was controllable by changing the composition of Dex-g-PLGA.The in vivo release studies using normal mice revealed that the plasma concentration of hGH was maintained for 1week without a signi ficant initial burst.The enhancement of biological activity of hGH by sustained release was con firmed by measuring the IGF-1concentration and body weight of hypophysectomized mice.These results suggest the high potential of the newly developed microparticles for the sustained release of biopharmaceuticals.©2009Elsevier B.V.All rights reserved.1.IntroductionSustained-release formulations of peptide and protein drugs (biopharmaceuticals)have attracted considerable interest because the short half-life and instability of biopharmaceuticals in plasma necessitate frequent administration by injection.Many microparticle formulations of those drugs are under development in order to reduce the dosing frequency,thereby increasing patient compliance.How-ever,despite the advances in this technology,there are only a few products that have received regulatory approval.Several problems such as the initial burst release,dif ficulty in controlling the release period and denaturation of entrapped proteins have prevented the realization of commercially viable products [1–4].The release behavior and denaturation of proteins depend on the environment that the protein encounters during the particle forma-tion and the internal structure of the microparticles.Numerous studies have been conducted to control these factors.The formation processes utilize various emulsions including water-in-oil-in-water [5],water-in-oil-in-oil [6],solid-in-oil-in-oil-in-water [7],solid-in-oil-in-oil [8]and solid-in-oil-in-water (S/O/W)[9–17].In particular,the S/O/W emulsion methods are effective for the stable encapsulation of proteins in the microparticles.Solidi fied protein microparticles in theS/O/W emulsion may suppress the contact between the protein and water/oil interface,which is one of the major causes of denaturation.However,a severe initial burst is often observed in the case of the solid phase prepared by conventional methods.The objective of this study is to develop a novel formulation of biodegradable microparticles for protein and peptide drugs through the S/O/W emulsion.The solid phase of S/O/W emulsion was prepared by lyophilization of the water-in-oil (W/O)emulsion using amphi-philic copolymers as an emulsi fier.In this process,we aimed to suppress the contact between the protein and water/oil interface using the amphiphilic copolymers aligned at the interface.The obtained solid was dispersed in an organic solvent in the form of nanoparticles.The agglomerated forms of the nanoparticles were then produced by S/O/W emulsion –solvent evaporation method to form protein-loaded microparticles.In this paper,we describe the characteristics of the microparticles and in vitro release behavior of an encapsulated protein drug,human growth hormone (hGH).The in vivo release was studied using a normal mouse and the biological activity of the released hGH was investigated using hypophysectomized mice.2.Materials and methods 2.1.MaterialsRecombinant hGH,methanol,chloroform,t -butanol and dimethyl-carbonate were purchased from Wako Pure Chemical (Osaka).DextranJournal of Controlled Release 142(2010)8–13⁎Corresponding author.Tel.:+81467328974;fax:+81467328363.E-mail address:Nobuo_Ida@nts.toray.co.jp (N.Ida).0168-3659/$–see front matter ©2009Elsevier B.V.All rights reserved.doi:10.1016/j.jconrel.2009.09.024Contents lists available at ScienceDirectJournal of Controlled Releasej ou r n a l h o m e pa g e :ww w.e l s ev i e r.c o m/l o c a t e /j c o n re l(molecular weight(MW)15,000–20,000)was purchased from Nacalai Tesque(Kyoto).DL-Lactide and hexamethyldisilazane were purchased from Tokyo Chemical(Tokyo).Glycolide was purchased from Poly-sciences,Inc.(Warrington,PA).Potassium t-butoxide(t BuOK),acetic acid and tetrahydrofuran(THF)were purchased from Kanto Chemical (Tokyo).Trifluoroacetic acid(TFA)was purchased from Kishida Chemical(Osaka).Pluronic F68was purchased from MP Biomedicals (Morgan Irvine,CA).Fluorecein-labeled dextran(MW40000,FD40) was purchased from Sigma Chemical Co.(St.Louis,MO).Anti-human growth hormone(hGH)monoclonal antibody,biotinylated anti-hGH polyclonal antibody and an ELISA kit for hGH quantitation were purchased from R&D Systems Inc.(Minneapolis,MN).Anti-mouse insulin-like growth factor1(IGF-1)monoclonal antibody and biotinylated anti-mouse IGF-1polyclonal antibody were purchased from R&D Systems Inc.(Minneapolis,MN).DL-lactide was purified by recrystallization and sublimation before use.Glycolide was purified by recrystallization.2.2.Synthesis of poly(lactide-co-glycolide)-grafted dextran(Dex-g-PLGA) graft copolymersDex-g-PLGA(LA:GA=50:50)was synthesized by graft polymer-ization of DL-lactide and glycolide on trimethylsilyl dextran(TMS-Dex)with t BuOK as an initiator,using the method reported by Ohya et al.with modification[18].Dextran was silylated using hexam-ethyldisilazane to obtain TMS-Dex with a silylation yield of82–92%. TMS-Dex and t BuOK were dissolved in anhydrous ctide and glycolide monomers dissolved in anhydrous THF were added to TMS-Dex solution by canula.The polymerization reaction was carried out for5min at room temperature under argon atmosphere.The reaction was quenched by adding acetic acid.The products were purified by chloroform/methanol precipitation.The deprotection of the OH group of TMS-Dex-g-PLGA was carried out by treatment with TFA in chloroform to yield Dex-g-PLGA.The molecular weight and graft number of the PLGA segment of Dex-g-PLGA was determined by 1H-NMR spectroscopy.GPC measurement was performed using a Tosoh GPC-8020system(eluent:DMF,column:TSK-GELα-5000×2, detector:refractive index)to check the purity of Dex-g-PLGA and to determine Mw/Mn.2.3.Microparticle preparationA typical microparticle preparation is as follows:An amphiphilic copolymer(5mg)was dissolved in100μL of dimethylcarbonate(DMC) containing20%(v/v)t-butanol.A protein solution(1or10mg/mL for in vitro or in vivo study,respectively,50μL)was added to the copolymer solution.After mixing using a vortex mixer,the obtained W/O emulsion was lyophilized.The lyophilized powder was suspended in200μL of DMC to yield S/O suspension(25mg(polymer)/mL).This S/O suspen-sion was added slowly to10%(w/v)aqueous Pluronic F68solution (2mL)and vortexed for30s or mixed using a stirrer at500,1000and 1600rpm for5min to form the S/O/W emulsion.The organic solvent was removed under reduced pressure.In the size control experiment, O/W ratios of the S/O/W emulsion were varied in the range from0.025 to1,keeping the polymer content and volume of water phase constant (2mL).The resulting microparticles were lyophilized and stored at −20°C.2.4.Particle characterizationThe morphologies of the nano-and microparticles were investi-gated by the scanning electron microscopy(SEM,Hitachi-S4800, Hitachi Ltd.,Japan).The particles suspended in water were centri-fuged at130×g for10min and the supernatant was removed.The precipitated particles were resuspended in water.SEM samples were prepared by placing a droplet of this suspension on a silicon wafer.The samples were dried in a vacuum desiccator and sputter-coated with platinum.SEM images were obtained at1.0–5.0kV.The sizes of nano-and microparticles were determined using SEM, the dynamic light scattering technique(DLS)with ELSZ(Otsuka Electronics Co.Ltd.,Japan)and the laser diffraction technique(LD) with Microtrac FRA(Nikkiso Co.Ltd.,Japan).DLS was employed for the particles with diameters smaller than1μm and LD was employed for the particles with diameters larger than1μm.The samples for DLS and LD were prepared by suspending dry particles(5mg)in water (15ml).In the SEM analysis,the size of microparticles were determined by measuring30to120particles in the SEM images and presented as mean±SD.2.5.Confocal laser scanning microscopyThe microparticles loaded withfluorecein-labeled dextran(MW 40000)as a model compound were suspended in phosphate-buffered saline(PBS)and placed in a glass-bottom96-well plate(BD Falcon). The microparticles werefixed with highly viscous polysaccharide solution.Confocal images were acquired using an Olympus laser scanning microscope(FV1000)equipped with a100×oil-immersion objective.The excitation wavelength was488nm(argon laser).2.6.Drug loading efficiencyThe encapsulation efficiency(EE)of hGH was determined as follows:the lyophilized microparticles were suspended in PBS (pH7.4)containing0.1%bovine serum albumin,0.1%Pluronic F-68 and sodium azide.The suspension was centrifuged at18,800×g for 10min and the supernatant was carefully removed.This procedure was performed thrice.The protein concentrations of the supernatants were determined by sandwich ELISA using monoclonal anti-hGH as a capture antibody and biotinylated polyclonal anti-hGH as a detection antibody.EE was determined by subtracting the amount of hGH in the supernatants from the feed amount.2.7.In vitro release studiesThe microparticles were washed thrice then suspended in PBS. The suspensions were incubated at37°C with continuous rotation at 2rpm.At different time points,an aliquot was centrifuged at 18,800×g for10min and the hGH concentration of the supernatant was determined by ELISA.2.8.In vivo release studiesThe in vivo release of hGH from the microparticles of Dex-g-PLGA (MW(Dex)17500,MW(PLGA)1822,number of graft chains38, Mw/Mn=1.20)was assessed using male BALB/c mice(10weeks old). The animals were immunosuppressed by intraperitoneal treatment with tacrolimus(Astellas Pharma)[13].The microparticles were washed thrice then suspended in PBS.The suspensions or free hGH were injected subcutaneously at a dose of28mg(hGH)/kg using 29-gauge needles.Blood samples were collected via the tail vein.After centrifugation,the heparinized plasma was collected and analyzed for hGH using the ELISA kit(R&D Systems).2.9.Biological activity of released hGHThe biological activity of released hGH was assessed using hypophysectomized male ICR mice(7weeks old).The microparticles of Dex-g-PLGA(MW(Dex)17500,MW(PLGA)1822,number of graft chains38,Mw/Mn=1.20)were washed thrice then suspended in PBS.The suspensions or free hGH were injected subcutaneously at a dose of28mg(hGH)/kg using29-gauge needles.The animals were weighed and blood samples were collected via the tail vein for9Y.Kakizawa et al./Journal of Controlled Release142(2010)8–132weeks.After centrifugation of the blood samples,the plasma was collected and analyzed for IGF-1by sandwich ELISA using monoclonal anti-mouse IGF-1as a capture antibody and biotinylated polyclonal anti-mouse IGF-1as a detection antibody.All the animal experiments were conducted in accordance with the Guidelines for Animal Experiments,Research&Development Divi-sion,Toray Industries,Inc.3.Results and discussion3.1.Microparticle preparationFig.1shows the fabrication of the microparticles entrapping protein or peptide drugs.In thefirst step,a protein aqueous solution is added to an organic solvent containing an amphiphilic copolymer as an emulsifier to form a water-in-oil(W/O)emulsion.The W/O emulsion is then quickly cooled using liquid nitrogen and lyophilized to remove water and organic solvent.The aim of this process is to obtain protein,which is present in the water phase of the W/O emulsion,as nanoparticles surrounded by the amphiphilic copoly-mers.Then,an organic solvent is added to the lyophilized powder to obtain a solid-in-oil(S/O)suspension.The S/O suspension is added to water containing Pluronic F68as an emulsifier to form a solid-in-oil-in-water(S/O/W)emulsion.Finally,the organic phase is evaporated to form microparticles with protein dispersed in the biodegradable polymer matrix.There are several studies in which the S/O/W emulsion–solvent evaporation method is used to prepare protein-loaded microparticles[9–17].In these studies,the solid phases of the protein typically have diameters in the range of2to10μm.In contrast, the size of the solid phase in our process depends on the size of the W/O emulsion formed in thefirst step.By properly choosing the condition of the emulsion formation,we can obtain the solid phase containing protein as nanoparticles.Amphiphilic copolymers play a critical role as a matrix of the microparticles and the stabilizer of W/O emulsion.According to Bancroft's rule,an oil-soluble surfactant is suitable for stabilization of a W/O emulsion[19].We synthesized more thanfifty different amphiphilic copolymers with sufficient oil solubility using this rule as a guideline and examined the particle formation capability,encapsu-lation efficiency and in vitro release behavior of a protein drug using hGH as a model(data not shown).On the basis of the screening assay results,we chose Dex-g-PLGA as the candidate polymer and further investigated the characteristics of microparticles composed of this polymer.3.2.Morphology and internal structure of microparticles of Dex-g-PLGAFig.2A shows the SEM image of the particles in the S/O suspension that was formed by the addition of the organic solvent to the lyophilized powder of the W/O emulsion.The morphologies of the microparticles obtained by solvent evaporation of the S/O/W emulsion are shown in Fig.2B.The microparticles were spherical and the surface was smooth,suggesting that the free polymer that existed in the organic phase of the S/O/W suspension forms the polymer matrix in which the nanoparticles are dispersed.The distribution of the encapsulated molecules in the microparticles was examined by confocal laser scanning microscopy usingfluorescein-labeled dextran(FD40)as a model compound(Fig.2C and D). Punctuate distribution offluorescence of FD40was observed in the microparticles,which is consistent with the proposed nanoparticle-aggregated structure of the microparticles.3.3.Size control of microparticlesThe size of the biodegradable microparticles has a profound impact on the release rate of protein[20,21].In addition,particle size is important with respect to the phagocytosis of particles by macro-phages and the needle diameter used for administration.To control the size of the microparticles,we investigated various parameters of the microparticle fabrication.Among the parameters studied,the stirring speed and ratio of the volume of the organic phase to that of the water phase(O/W ratio)for the formation of S/O/W emulsion have a considerable effect on the size of the particles.We changed the stirring speed during the formation of the S/O/W emulsion from500 to1600rpm and measured the diameter of the particles observed by SEM(Fig.3).As the stirring speed was decreased,the mean diameter of the microparticles increased from24±12to59±12μm.As for the O/W ratio,the increase in the ratio resulted in a decrease in the diameter,which was determined by SEM,DLS and LD(Figs.3and4). In these experiments,the concentration of the solid phase ofS/O Fig.1.Scheme of preparation of protein-loadedmicroparticles.Fig.2.Scanning electron micrographs of(A)nanoparticles in the S/O suspension and(B)microparticles of Dex-g-PLGA.(C)Confocal microscopy image and(D)differentialinterference micrograph of Dex-g-PLGA microparticles containing a model compound(FD40).10Y.Kakizawa et al./Journal of Controlled Release142(2010)8–13suspension varies with the O/W ratio.The diameter of the resulting particles is presumably affected by the emulsion size and concentra-tion of the S/O suspension.The diameter reached the minimum of 270±120nm at the O/W ratio of 0.25.The particles above this O/W ratio are likely to correspond to the single nanoparticle present in the S/O suspension.It is noteworthy that the range of diameters of the particles depends on the properties of the organic phase.Similar results were obtained for ethyl acetate used as the organic phase of the S/O/W emulsion.On the other hand,the diameters were in the range from 1.1±0.1to 1.2±0.2μm when chloroform was used to form the S/O/W emulsion in the range from 0.025to 0.25in O/W ratio (data not shown).Taken together,we can control the diameter of theparticles by over two orders of magnitude by properly choosing the stirring speed,O/W ratio and type of organic phase.3.4.Effects of polymer compositions on in vitro release of hGH The in vitro release of hGH from the microparticles was investigated with particular focus on the molecular weight of Dex-g-PLGA.Table 1shows the parameters of Dex-g-PLGA polymers and microparticles used for the in vitro release study.The EE of hGH was more than 94%and the diameter of all the microparticles was approximately 5μm.Fig.5shows the release pro file from the microparticles composed of Dex-g-PLGA with three different compositions.In these experi-ments,the molecular weight (MW)of dextran segments is fixed at 17500and those of the PLGA segments were varied from 2790to 4330.Because the graft numbers of the polymers are almost constant,the ratio of total MW of PLGA segments to the MW of dextran is proportional to the MW of PLGA.The microparticles composed of Dex-g-PLGA with low-molecular-weight PLGA segments released hGH at a high release rate.The time periods at which 50%of hGH was released from microparticles composed of polymer A,B and C were 7,11and 20days,respectively.It is reported that the release rate is higher for the particles composed of block copolymers with low hydrophobic segment content [22,23].As for the microparticles composed of Dex-PLGA with low PLGA content,the water molecules would diffuse inside the particles more easily because of the more hydrophilic interior.3.5.In vivo release of hGH from subcutaneously injected microparticles Fig.6shows the change in the plasma hGH concentration after subcutaneous injection to normal mice.The concentration of hGH was determined by ELISA.In the case of free drug,theplasmaFig. 3.Scanning electron micrographs of microparticles formed under various conditions.O/W ratio of the formation of S/O/W emulsion is 0.1and stirring speed during the formation of the S/O/W emulsion is (A)500,(B)1000and (C)1600rpm.The S/O/W emulsion was formed using a vortex mixer and O/W ratio is (D)0.1(E)0.175and (F)0.5.Fig.4.Microparticle size as a function of O/W ratio of the S/O/W emulsion determined by (▲)SEM,(□)DLS and (○)LD.Table 1Characteristics of Dex-g-PLGA used in in vitro release study.Polymer MW(Dex)MW(PLGA)a Number of graft chains a TotalMW b Mw/Mn c EE/%dA 17,50027902792,830 1.1794.2±0.5B 17,50030002798,500 1.1595.7±0.1C17,500433033160,390 1.1797.5±0.1a Determined by 1H-NMR.b Calculated by the equation:MW(dextran)+MW(PLGA)×number of graft chains.c Determined by GPC.dMean ±SD (n =3).Fig.5.In vitro release pro files of hGH from Dex-g-PLGA microparticles composed of various molecular weights of PLGA segments in PBS at 37°C.(▲)polymer A,(●)polymer B and (■)polymer C.Characteristics of the polymers are described in Table 1.Values are the mean and standard deviation of three experiments.11Y.Kakizawa et al./Journal of Controlled Release 142(2010)8–13concentration increased rapidly and decreased to the control level within 3days.On the other hand,the rapid increase after adminis-tration was not observed and the constant plasma level was maintained for a week in the case of the microparticles.Fourteen days after the microparticle administration,hGH was no longer detectable in the plasma;this may be due to the production of anti-hGH antibody,even though we used an immunosuppressive reagent in this experiment.The antibody production is attributed to the fact that human growth hormone is a foreign protein for a mouse [24].These results demonstrate the potential of our microparticles to overcome the problems associated with commercially available hGH-loaded microparticles,such as the initial burst and relatively short release period.3.6.In vivo study of biological activity of released proteinTo con firm the biological activity of the released hGH,the plasma concentration of insulin-like growth factor-1(IGF-1),a cytokine induced by the growth hormone activity,and change in body weight,also a marker of growth hormone action,were measured using hypophysectomized mice injected with hGH microparticles [25,26].Fig.7shows the change in the IGF-1concentration.Following free drug administration,the IGF-1level became highest (61and 75ng/mL)1day after the administration and decreased to the base level after 4days.On the other hand,following microparticle administration,the IGF-1concentration reached 108and 167ng/mL after 4days and remained high for about 2weeks.As for the body weight change,the increments at 2weeks after administration were 9%and 20%for free drug and microparticles,respectively.These results demonstrate that the hGH released from the microparticles maintained its biological activity and,because of the sustained release,the hGH encapsulated in the microparticle showed better effects than free drug.4.ConclusionsA new formulation of protein-loaded microparticles was developed through S/O/W emulsion using amphiphilic copolymers as a matrix.The near zero-order release without a signi ficant initial burst was achieved using Dex-g-PLGA.In addition,it was possible to control the size of the microparticles from 270nm to 59μm by changing the parameters including the O/W ratio of the S/O/W emulsion and the stirring speed.The release rate of hGH is dependent on the composition of Dex-g-PLGA.The in vivo release studies using normal mice revealed that the plasma concentration of hGH was maintained for 1week without a signi ficant initial burst.The biological activity of hGH and the enhancement ofactivity by sustained release were con firmed.These results suggest the high potential of the newly developed microparticles for the sustained release of biopharmaceuticals.AcknowledgementsThe authors express their great appreciation to Drs.Yuichi Ohya,Masahiro Goto and Noriho Kamiya for helpful discussion.This research was financially supported by the Japan Science and Technology Agency (JST).References[1]R.T.Bartus,M.A.Tracy,D.F.Emerich,S.E.Zale,Sustained delivery of proteins fornovel therapeutic products,Science 281(1998)1161–1162.[2]S.D.Putney,P.A.Burke,Improving protein therapeutics with sustained-releaseformulations,Nat.Biotechnol.16(1998)153–157.[3]J.L.Cleland,O.L.Johnson,S.Putney,A.J.S.Jones,Recombinant human growthhormone poly(lactic-co-glycolic acid)microsphere formulation development,Adv.Drug Deliv.Rev.28(1997)71–84.[4]F.Wu,T.Jin,Polymer-based sustained-release dosage forms for protein drugs,challenges,and recent advances,AAPS PharmSciTech 9(4)(2009)1218–1229.[5]J.L.Cleland,E.Duenas,A.Daugherty,M.Marian,J.Yang,M.Wilson,A.C.Celniker,A.Shahzamani,V.Quarmby,H.Chu,V.Mukku,A.Mac,M.Roussakis,N.Gillette,B.Boyd,D.Yeung,D.Brooks,Y.F.Maa,C.Hsu,A.J.S.Jones,Recombinant human growth hormone poly(lactic-co-glycolic acid)(PLGA)microspheres provide a long lasting effect,J.Control.Release 49(1997)193–205.[6]M.K.Yeh,A.G.A.Coombes,P.G.Jenkins,S.S.Davis,A novel emulsi fication-solventextraction technique for production of protein loaded biodegradable micropar-ticles for vaccine and drug delivery,J.Control.Release 33(1995)437–445.Fig.6.hGH concentrations in plasma of normal mice treated with (○)free hGH and (●)hGH entrapped in Dex-g-PLGA microparticles.Values are the mean and standard deviation of threemice.Fig.7.(A)IGF-1concentrations in plasma and (B)body weight of hypophysectomized mice treated with (Δ)saline,(□)free hGH and (●)hGH entrapped in Dex-g-PLGA microparticles.Values of two independent experiments at each point are presented.12Y.Kakizawa et al./Journal of Controlled Release 142(2010)8–13[7]W.Yuan,F.Wui,M.Guo,T.Jin,Development of protein delivery microspheresystem by a novel S/O/O/W multi-emulsion,Eur.J.Pharm.Sci.36(2009)212–218.[8]W.Jiang,S.P.Schwendeman,Stabilization and controlled release of bovine serumalbumin encapsulated in poly(D,L-lactide)and poly(ethylene glycol)microsphere blends,Pharm.Res.18(6)(2001)878–885.[9]J.L.Cleland, A.J.S.Jones,Stable formulations of recombinant human growthhormone and interferon-γfor microencapsulation in biodegradable microspheres, Pharm.Res.13(10)(1996)1464–1475.[10]O.L.Johnson,W.Jaworowicz,J.L.Cleland,L.Bailey,M.Charnis,E.Duenas,C.Wu,D.Shepard,S.Magil,st, A.J.S.Jones,S.D.Putney,The stabilization and encapsulation of human growth hormone into biodegradable microspheres, Pharm.Res.14(6)(1997)730–735.[11]M.Iwata,T.Tanaka,Y.Nakamura,J.W.McGinity,Selection of the solvent systemfor the preparation of poly(D,L-lactic-co-glycolic acid)microspheres containing tumor necrosis factor-alpha(TNF-α),Int.J.Pharm.160(1998)145–156.[12]T.Morita,Y.Sakamura,Y.Horikiri,T.Suzuki,H.Yoshino,Protein encapsulationinto biodegradable microspheres by a novel S/O/W emulsion method using poly (ethylene glycol)as a protein micronization adjuvant,J.Control.Release69 (2000)435–444.[13]S.Takada,Y.Yamagata,M.Misaki,K.Taira,T.Kurokawa,Sustained release ofhuman growth hormone from microcapsules prepared by a solvent evaporation technique,J.Control.Release88(2003)229–242.[14]I.J.Castellanos,K.Griebenow,Improvedα-chymotrypsin stability upon encapsu-lation in PLGA microspheres by solvent replacement,Pharm.Res.20(11)(2003) 1873–1880.[15]I.J.Castellanos,R.Crespo,K.Griebenow,Poly(ethylene glycol)as stabilizer andemulsifying agent:a novel stabilization approach preventing aggregation and inactivation of proteins upon encapsulation in biodegradable polyester micro-spheres,J.Control.Release88(2003)135–145.[16]J.W.Wang,K.M.Chua,C.H.Wang,Stabilization and encapsulation of humanimmunoglobulin G into biodegradable microspheres,J.Colloid Interface Sci.271 (2004)92–101.[17]garce,E.Garcion,N.Faisant,O.Thomas,P.Kanaujia,P.Menei,J.P.Benoit,Development and characterization of interleukin-18-loaded biodegradable micro-spheres,Int.J.Pharm.314(2006)179–188.[18]T.Ouchi,T.Saito,T.Kontani,Y.Ohya,Encapsulation and/or release behavior ofbovine serum albumin within and from polylactide-grafted dextran microspheres, Macromol.Biosci.4(2004)458–463.[19]J.Raynaud,B.Choquenet,E.Marie,E.Dellacherie,C.Nouvel,J.-L.Six,A.Durand,Emulsifying properties of biodegradable polylactide-grafted dextran copolymers, Biomacromolecules9(2008)1014–1021.[20]C.Berkland,M.King,A.Cox,K.Kim,D.W.Pack,Precise control of PLG microspheresize provides enhanced control of drug release rate,J.Control.Release82(2002) 137–147.[21]J.Panyam,M.M.Dali,S.K.Sahoo,W.Ma,S.S.Chakravarthi,G.L.Amidon,R.J.Levy,V.Labhasetwar,Polymer degradation and in vitro release of a model protein from poly(D,L-lactide-co-glycolide)nano-and microparticles,J.Control.Release92 (2003)173–187.[22]J.M.Bezemer,R.Radersma,D.W.Grijpma,P.J.Dijkstra,C.A.van Blitterswijk,J.Feijen,Microspheres for protein delivery prepared from amphiphilic multiblock copolymers2.Modulation of release rate,J.Control.Release67(2000)249–260.[23]J.M.Bezemer,R.Radersma,D.W.Grijpma,P.J.Dijstra,J.Feijen,C.A.van Blitterswijk,Zero-order release of lysozyme from poly(ethylene glycol)/poly(butylene terephthalate)matrices,J.Control.Release64(2000)179–192.[24]K.D.F.Vlugt-Wensink,R.de Vrueh,M.G.Gresnigt,C.M.Hoogerbrugge,S.C.vanBuul-Offers,L.G.J.de Leede,L.G.W.Sterkman,D.J.A.Crommelin,W.E.Hennik,R.Verrijk,Preclinical and clinical in vitro and in vivo correlation of an hGH dextran microsphere formulation,Pharm.Res.24(12)(2007)2239–2248.[25]L.S.Mathews,G.Norstedt,R.D.Palmiter,Regulation of insulin-like growth factor Igene expression by growth hormone,Proc.Natl.Acad.Sci.U.S.A.83(1986) 9343–9347.[26]M.Søndergaard,F.Dagnæs-Hansen,A.Flyvbjerg,T.G.Jense,Normalization ofgrowth in hypophysectomized mice using hydrodynamic transfer of the human growth hormone gene,Am.J.Physiol.:Endocrinol.Metab.285(2003)427–432.13Y.Kakizawa et al./Journal of Controlled Release142(2010)8–13。