Journal of Controlled Release 107 (2005) 310 319 www.elsevier.

com/locate/jconrel

Preparation and characterization of melittin-loaded poly (dl-lactic acid) or poly (dl-lactic-co-glycolic acid) microspheres made by the double emulsion method
Fude Cui *, Dongmei Cun, Anjin Tao, Mingshi Yang, Kai Shi, Min Zhao, Ying Guan
Department of Pharmaceutics, School of Pharmaceutical Science, Shenyang Pharmaceutical University, No. 103, Wenhua Road, Shenyang 110016, China Received 3 September 2004; accepted 6 July 2005

Abstract The water soluble peptide, melittin, isolated from bee venom and composed of twenty-six amino acids, was encapsulated in poly (dl-lactic acid, PLA) and poly (dl-lactic-co-glycolic acid, PLGA) microspheres prepared by a multiple emulsion [(W1 / O)W2] solvent evaporation method. The aim of this work was to develop a controlled release injection that would deliver the melittin over a period of about one month. The influence of various preparation parameters, such as the type of polymer, its concentration, stabilizer PVA concentration, volume of internal water phase and level of drug loading on the characteristics of the microspheres and drug release was investigated. It was found that the microspheres of about 5 Am in size can be produced in high encapsulation (up to 90%), and the melittin content in the microspheres was up to 10% (w/w). The drug release profiles in vitro exhibited a significant burst release, followed by a lag phase of little or no release and then a phase of constant melittin release. The type of polymer used was a critical factor in controlling the release of melittin from the microspheres. In this study, the rate of peptide release from the microspheres correlated well with the rate of polymer degradation. Moreover, melittin was released completely during the study period of 30 days, which agreed well with the polymer degradation rate. D 2005 Published by Elsevier B.V.
Keywords: Biodegradable microspheres; Melittin; W1 / O / W2 emulsion-solvent evaporation technique

1. Introduction With the development of biotechnology and genetic engineering, many pharmacologically active peptides and proteins are now good candidates for therapeutic drug treatment. However, their clinical applications involving oral administration have been

* Corresponding author. Tel.: +86 24 23986355; fax: +86 24 23986355. E-mail address: cundongmei@hotmail.com (F. Cui). 0168-3659/$ - see front matter D 2005 Published by Elsevier B.V. doi:10.1016/j.jconrel.2005.07.001

F. Cui et al. / Journal of Controlled Release 107 (2005) 310319

311

limited due to their higher molecular weight, which makes diffusion through biological membrane difficult, and their instability in the GI environment. Alternative administration options by frequent injection are also tedious and expensive [1]. For these reasons, priority is being given to developing controlled delivery systems with a greater parenteral dosing interval, which will increase patience acceptance, and improve drug management. Melittin, the predominant peptide isolated from honeybee venom (Apis mellifera), is a 26 amino acid peptide [2]. The peptide has a molecular weight of 2.86 kD and high aqueous solubility. It is a typical representative of biologically active peptide drugs with high therapeutic potential. Bee venom has been found to have a marked effect on the immune system [3], cardiovascular system [4] and blood [5] and it also exhibits anti-tumour activity [6]. Melittin is commonly used in the treatment of arthritic disorders, such as rheumatoid arthritis and osteoarthritis. The traditional therapy pattern for arthritis using melittin was direct stings by bees, which have a number of disadvantages such as the severe pain involved, no control over the exact dose, and the need to sting repeatedly due to the short half-life of melittin. So this often leads to poor patient compliance. It is necessary to design a suitable release system which provides a long-term and constant therapeutic effect. In recent years, microspheres of biodegradable polymers such as poly (dl-lactic acid, PLA), and poly (dl-lactic-co-glycolic acid, PLGA) have attracted much interest due to their ability to control the release of bioactive macromolecules, such as some peptides or proteins loaded microspheres have been used to improve patient compliance by eliminating the need for frequent injection [7,8]. PLA and PLGA are biocompatible, biodegradable and approved by the FDA for certain clinical uses. Their degradation times can be varied from days to years by altering the type of polymer, the polymer molecular weight, or the structure of the microspheres. Moreover, biodegradable PLGA microspheres have also been demonstrated to be an excellent drug carrier for arthritic lesions following radiopharmaceutical scintigraphic studies in rabbits [9]. The aim of this study was to develop and characterize melittin-loaded biodegradable PLA or PLGA

microspheres in order to obtain a controlled release system to treat arthritic conditions. The influence of the type of polymer, its concentration, stabilizer PVA concentration, volume of internal water phase and level of drug loading on microsphere characteristics (encapsulation efficiency, particle size and surface morphology) and in vitro release profiles was studied to optimize the microsphere system.

2. Materials and methods 2.1. Materials Melittin was isolated from natural bee venom (A. mellifera) as described in the literature [10]. PLA with a weight average molecular weight (Mw) of 10,425 was supplied by Shan dong Medical Institute. PLGA with weight average molecular weights (Mw) of 10,050 and 9500, in which the copolymer ratio of d, l-lactide to glycolide was 75 : 25 and 50 : 50, respectively, were obtained from Chengdu Institute of Organic Chemistry, Chinese Academy of Sciences. The weight average molecular weight (Mw) of three kinds of polymer was determined by the supplier using gel permeation chromatography (GPC). Polyvinyl alcohol (PVA) was supplied by Shin-Etsu Chemical Co., Ltd. Japan. All other chemicals were of analytical grade. 2.2. Preparation of melittin-loaded microspheres A modified water-in oil-in water (W1 / O / W2) double-emulsion solvent evaporation method was employed to prepare microspheres containing melittin. An aqueous melittin solution was mixed with dichloromethane (DCM) containing polymer (PLA or PLGA), and emulsified in a homogenizer (FA 25 Fluko Germany) at 23,000 rpm for 2 min. The resultant first emulsion (W1 / O) was injected into 40 mL polyvinyl alcohol (PVA) solution (external aqueous phase) using a glass syringe with a needle under agitation, and emulsification continued at 10,000 rpm for 2 min to produce a W1 / O / W2 emulsion. In order to increase the diffusion rate of dichloromethane from the emulsion drops into the external phase, 160 mL of external water phase (PVA solution containing the same concentration of PVA) was added to the W1 /

312

F. Cui et al. / Journal of Controlled Release 107 (2005) 310319

O / W2 emulsion system under magnetic agitation. The system was stirred continuously for 4 h at room temperature and atmospheric pressure to evaporate the dichloromethane completely and prevent pore formation on the surface of the microspheres. The emulsion solidified gradually with the diffusion of the solvent from the emulsion droplets into the external solvent and the solvent escaped from the external phase. After the microspheres had formed, they were centrifuged, washed three times with distilled water and freeze-dried. The final product was stored in a desiccator at 20 8C. 2.3. Characterization of microspheres

2.5. In vitro polymer degradation Thirty milligrams of blank microspheres (n = 3) was incubated in 6 mL 1 / 30 M acetate sodium buffer (pH 4.0)5 mM SDS containing 0.01% sodium azide at 37.0 F 0.1 8C. At each time point, the recovered wet microspheres were dried for 48 h under vacuum at room temperature, and weighed accurately (dry weight, W d). The mass remaining percent (MR%) was calculated as follows: MR% Wd =W0 d100 Where, W 0 is the initial mass at zero time. 2.6. In vitro release profiles 1

2.3.1. Particle size analysis A small amount of freeze-dried microspheres was redispersed in distilled water and their size was measured by laser diffractometry using a Beckman Coulter LS32 laser sizer. The particle size was expressed as the volume mean diameter (Vmd) in micrometer. 2.3.2. Surface morphology of PLGA and PLA microspheres The shape and surface morphology of the microspheres were examined by a scanning electron microscope (SEM, S-5200, Hitachi Co., Japan). 2.4. Determination of melittin loading and encapsulation efficiency The encapsulation efficiency was determined by a modified solvent dissolution method with dimethylsulfoxide (DMSO) [11]. Briefly, freeze-dried microspheres were accurately weighed (10 to 15 mg) and were placed in a 10 mL flask to which 0.8 mL DMSO was added. After the micropsheres had dissolved, 0.1 N NaOH containing 0.5% SDS was added and gently mixed. After being allowed to stand at room temperature for 1 h, the resultant solution became clear. The melittin concentration in the mixed solution was measured by the Lowry method. From this result, the loading percentage (w/w, melittin content per dry microspheres) was determined. Each sample was assayed in triplicate. The percentage entrapment efficiency was expressed by comparing the actual melittin loading with the theoretical melittin loading.

In vitro release testing of the microspheres was performed as follows. About 10 mg melittin-loaded microspheres was quantitatively transferred to a testtube and incubated with 2 mL 1 / 30 mM acetate sodium buffer (pH 4.0) containing 5 mM SDS and 0.01% sodium azide. The temperature in the waterbath incubator was maintained at 37.0 F 0.1 8C with continuous agitation at about 50 rpm. At each sampling time, the supernatant was withdrawn after centrifugation at 3000 rpm for 5 min, and replaced with an equal volume of release medium. The amount of melittin released was evaluated by the Lowry method. Experiments were preformed in triplicate. Data are presented as average with the standard deviations. Comparisons were made between different preparations for the quality of released melittin in vitro using one way analysis of variance. P b 0.05 was considered as the significant level.

3. Results and discussion This study was carried out to investigate the feasibility of preparing biodegradation melittin microspheres using the double-emulsion evaporation technique. Generally, the characteristics of the microspheres such as their particle size, morphology, and drug content were affected by the different experimental conditions. In the present study, several parameters such as the type of polymer, its concentration, the volume of inner water phase, the concentration of PVA in the external water phase and peptide loading

F. Cui et al. / Journal of Controlled Release 107 (2005) 310319 Table 1 Preparation parameters for melittin microspheres Formulation ID A B C D E F G H I J K Polymer PLA PLGA PLGA PLGA PLGA PLGA PLGA PLGA PLGA PLGA PLGA Polymer conc. (%, w/v) 7.5 7.5 7.5 15. 3.75 7.5 7.5 7.5 7.5 7.5 7.5 PVA conc. (%, w/v) 1 1 1 1 1 2 5 1 1 1 1 Volume of DCM (mL) 4 4 4 4 4 4 4 4 4 4 4 Volume of inner water phase (mL) 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.4 0.8 0.2 0.2

313

Theoretical loading (%) 14.3 14.3 14.3 14.3 14.3 14.3 14.3 14.3 14.3 9.1 3.2

(75 : 25) (50 : 50) (50 : 50) (50 : 50) (50 : 50) (50 : 50) (50 : 50) (50 : 50) (50 : 50) (50 : 50)

were investigated to determine their effect on the characteristics of the melittin-loaded biodegradable microspheres. The preparation parameters used in this study are listed in Table 1. 3.1. Effect of preparation conditions on size, morphology and melittin entrapment of the microspheres 3.1.1. The effect of the polymer type In order to select a biodegradable polymer for the expected controlled release system, three kinds of polymer were used to prepare microspheres to investigate their effect on the characteristics of the melittinloaded microspheres. As shown in Table 2 (formulation A, B and C), it was found that the melittin encapsulation efficiency was increased and the average particle size was decreased on increasing the GA / LA ratio. Similar results have also been reported in the literature [12,13]. Fig. 1(ad) shows the surface structure of the microspheres prepared with the three kinds of polymers. It can be seen that their surface morphology differs sig-

nificantly. The PLA microspheres possessed a higher porous surface and had a dsponge-likeT porous structure. However, a smoother, nonporous surface was formed on increasing the GA / LA ratio. The microspheres prepared with PLGA (50 / 50) had a smoother, denser and nonporous surface compared with the other microspheres. Evidently, the morphology of the microspheres was affected by the type of polymer. This phenomenon was due to the difference in crystallinity and hydrophobicity between the polymers. The microspheres prepared from a polymer with a higher lactic acid content hardened more quickly during the preparation procedure. The reason for the porous surface of PLA and PLGA (75 : 25), especially PLA, was that the solvent had evaporated too quickly from the microspheres. Vaporization of the solvent inside the microspheres will cause disruption of the polymer film. 3.1.2. Effect of the polymer concentration in the oil phase Table 2 (formulation C, D and E) shows that the microsphere size and melittin encapsulation efficiency

Table 2 Characteristics of melittin-loaded microspheresa Formulation ID Particle size (Am) Melittin content (%) Encapsulation efficiency (%)
a

A 21.4 8.7 60.8

B 9.0 10.6 74.1

C 7.0 11.5 80.4

D 24.7 12.5 87.4

E 4.3 10.0 69.9

F 6.2 12.2 85.3

G 4.1 13.3 93.0

H 7.4 11.8 82.5

I 9.8 12.1 84.6

J 5.7 7.7 84.6

K 5.0 3.0 93.8

For sample characteristics in Table 1.

314

F. Cui et al. / Journal of Controlled Release 107 (2005) 310319

Fig. 1. Typical SEM images of the melittin-loaded microspheres. See Table 1 for the details of the formulations: (a) formulation C; (b) formulation B; (c) and (d) formulation A; (e) formulation H; (f) formulation I.

were increased when the PLGA (50 / 50) concentration in DCM was increased. It was demonstrated that the size of emulsion droplets depends on the balance between stirring shear force and droplet cohesion. It was considered that the higher concentration of PLGA, at a fixed stirring shear force, results in a higher viscosity of the oil phase, which makes it difficult for small w / o / w emulsion droplets to form and become larger particles. The increased melittin entrapment efficiency can also be attributed to the increase in both oil phase viscosity and the larger size of the oil droplets, which prevents leakage of

the inner aqueous phase through the oil phase into the outer aqueous phase [13]. 3.1.3. Effect of PVA concentration in the external water phase PVA concentration in the external water phase is known to be a key factor affecting the size of microspheres [12]. In the present work, 1%, 2% and 5% PVA solutions were used as the external water phase to examine the effect of PVA concentration on the characteristics of the microspheres. The results are summarized in Table 2 (formulation C, F and G).

F. Cui et al. / Journal of Controlled Release 107 (2005) 310319

315

The size of microspheres fabricated at 1%, 2% and 5% PVA concentration was 7.0, 6.2 and 4.1 Am, respectively. Evidently, a significant decrease in particle size can be achieved by increasing the concentration of PVA in the continuous phase. A higher PVA concentration could increase the stability of emulsion droplet formed during homogenization because the increased viscosity of the external water phase would prevent emulsion droplets from coalescence, resulting in smaller emulsion droplets. These emulsion droplets gradually hardened to form microspheres as the solvent in the emulsion droplets continued to evaporate. Therefore, the size of the microparticles was dependent on the size of the emulsion droplets formed during homogenization. Moreover, a significant increase in melittin encapsulation efficiency from 80.4% to 93.0% occurred as the concentration of PVA increased in the external water phase. This phenomenon can be explained as follows. A higher PVA concentration increased the viscosity of the external water phase, which decreased the rate of melittin diffusing from the inner water phase to the outer water phase, i.e. hindered the mass transfer of melittin to the surrounding region. Thus, the drug can be distributed more evenly within the interior of the microspheres resulting in the increased encapsulation efficiency of melittin [14]. 3.1.4. The effect of the internal aqueous phase volume Table 2 (formulation C, H and I) lists the properties of the microspheres produced with different internal aqueous phase volumes, in which the amount of melittin remained unchanged, at a fixed volume of oil phase. Fig. 1(a), (e) and (f) illustrates the corresponding surface morphology. It can be seen that the internal aqueous phase volume has less effect on the encapsulation efficiency, but has a significant impact on the surface morphology of the resulting microspheres [12,15]. It was observed that: (1) the surface of the microspheres was more porous when the microspheres were prepared at an increased water phase (i.e. decreasing the oil / water volume ratio); (2) the average particle size and melittin entrapment efficiency were increased slightly on increasing the internal aqueous phase volume. The size of the inner water droplets was determined during production of the W / O emulsions, and the increased volume of inner water phase led

to the formation of larger emulsion droplets, followed by the formation of larger microspheres. Moreover, the internal phase water can be eliminated from microspheres by two routes: (1) during the preparation process, part of the inner water in the W / O / W emulsion droplets diffuses through the polymeroil phase into the external phase; (2) during the drying process, residual water evaporates from the microspheres. The removal of water within the microspheres could leave pores on the surface of the microspheres. Consequently, the more water that was introduced into the preparation system, the greater the number of porous or empty spaces that were generated. 3.1.5. The effect of melittin loading The effect of melittin loading on the characteristics of the microspheres is shown in Table 2 (formulation C, J and K). As the melittin / polymer ratio was increased by increasing the initial weight of melittin dissolved in the inner aqueous phase, the efficiency of melittin entrapment was reduced. There was also a small increase in average particle size. It was clear that increasing the melittin concentration in the emulsion droplets increased the concentration gradient of melittin between the internal water phase and the external water phase, and the drug diffusion from the emulsion droplets into the external water phase was promoted [16,17]. 3.2. In vitro release studies 3.2.1. The effect of the type of polymer The release profiles of microspheres prepared with different types of polymer are shown in Fig. 2(a). It was found that a significant initial bburst releaseQ profile was obtained for all of three kinds of polymer over the first day. Especially, in the case of the PLA microspheres, virtually 80% of the entrapped melittin was released over the first day of the study. It was assumed that the burst release phenomenon was caused by the release of poorly entrapped or surface-associated melittin and melittin diffusion occurring through porous channels. The surface morphology (seen in Fig. 1) of the microspheres demonstrated that the PLA microspheres possess a polyporous surface and dspongelikeT structure. So, melittin diffused readily through the channels and pores. The initial release of melittin from PLGA (75 / 25) microspheres or the PLGA (50 /

316

F. Cui et al. / Journal of Controlled Release 107 (2005) 310319

(a)
100

Cumulative release(%)

80 60 40 20 0 0 5 10 15 20 25 30

glycolic acid (50 : 50) accelerated the degradation rate and presumably water was taken up more readily compared with the lower content of glycolic acid (75 : 25) [19]. It was evident that the co-polymer composition was a very important factor in controlling the release profile of melittin from microspheres. 3.2.2. The effect of the concentration of polymer (PLGA 50 / 50) in the oil phase The influence of the concentration of polymer (PLGA 50 / 50) on the release characteristics was shown in Fig. 3(a). The PLGA microspheres showed a typical three-phase in vitro release. The release profiles were characterized by the initial burst release followed by a lag period of about 6 days until the polymer looses sufficient mass to initiate erosion-con-

Time(days)

(b)
100

Mass remaining(%)

80 60 40 20 0 0 5 10 15 20 25 30 35

(a)
100

Cumulative release(%)

80 60 40 20 0 0 5 10 15 20 25 30

Time(days)
Fig. 2. (a) The effect of the type of polymer used on the melittin release characteristics from microspheres: PLGA (50 / 50) (E), PLGA (75 / 25) ( ) and PLA (n) (n = 3); (b) mass loss of biodegradable microspheres in 1 / 30 M acetate sodium buffer (pH 4.0) at 37 8C: PLGA (50 / 50) (E), PLGA (75 / 25) ( ) and PLA (n) (n = 3).

Time(days)

(b)
50) microspheres was apparently restricted compared with that from PLA microspheres. After the initial burst release, there was a lag time of little or no release, followed by a phase of constant melittin release. The lag time of the melittin release profile depended on the degradation rate of the polymer [18]. As shown in Fig. 2(b), there was no change in PLA weight during the lag period in the degradation study. Consequently, the melittin release from the PLA microspheres reached a plateau after 72 h. Comparing the release profiles of the microspheres prepared from PLGA (75 / 25) and PLGA (50 / 50), it was found that the melittin release rate from PLGA (50 : 50) was faster than from PLGA (75 : 25), although the surface of PLGA (75 : 25) was more porous. Furthermore, the release profile from PLGA (75 : 25) microspheres exhibited a prolonged lag time. This may be the reason why the higher content of
100

Cumulative release(%)

80 60 40 20 0 0 5 10 15 20 25 30

Time(days)
Fig. 3. (a) The effect of the concentration of polymer on the melittin release characteristics from microspheres: 3.75% PLGA (E), 7.5% PLGA ( ) and 15% PLGA (n) (n = 3); (b) the effect of the concentration of PVA in external water phase on the release rate of melittin from PLGA microspheres: 1% PVA (E), 2% PVA ( ) and 5% PVA (n) (n = 3).

F. Cui et al. / Journal of Controlled Release 107 (2005) 310319

317

100

80 60 40 20 0 0 5 10 15 20 25 30

Time(days)
Fig. 4. The effect of the volume of the internal water phase on the release rate of melittin from PLGA microspheres: 0.2 mL (E), 0.4 mL ( ) and 0.8 mL (n) (n = 3).

3.2.4. The effect of the volume of the internal water phase The microspheres were prepared using 7.5% PLGA (50 / 50) in the oil phase with different volumes of internal water phase. The melittin release profiles from these preparations were shown in Fig. 4. The initial burst release increased significantly ( p b 0.05) on increasing the internal water volume. The internal water volume had a significant impact both on the surface and the inner morphology of the microspheres, i.e., more internal water produced matrix microspheres characterized by a greater number of pores [15]. Thus, there was a correlation between drug release and the porosity of the microspheres. 3.2.5. The effect of drug loading Drug loading was also a significant factor influencing drug release ( p b 0.05). Fig. 5 showed the effect of different degrees of melittin loading on the cumulative release of microspheres made with PLGA (50 / 50). It was obvious that a higher concentration of melittin embedded in the microspheres structured in the PLGA matrix led to a higher concentration gradient between the microspheres and dissolution medium. In addition, at a high loading level, more melittin may be distributed over the surface of the microspheres, so that a higher drug loading always leads to a greater initial burst release [19,20]. Higher encapsulation efficiency and lower initial release are the most important parameters in the development of sustained release microspheres containing water-soluble drugs. Microspheres containing melittin,

trolled release. After the initial lag, a nearly linear and continuous release was observed over 1013 days in vitro; a continuous release followed, up to 100% at day 20. As shown in Fig. 3(a), the initial burst release was highly dependent on the formulation parameters. When the concentration of polymer solution increased from 3.75% to 15%, the initial burst release decreased from 49% to 17%. The concentration of polymer had a significant impact ( p b 0.05) on the initial burst release of melittin from microspheres. This may be due to the higher polymer concentration leading to the formation of a dense polymer matrix structure in the microspheres, resulting in smaller pores and a more tortuous structure. 3.2.3. The effect of the PVA concentration in the external water phase Release profiles of melittin-loaded microspheres prepared with PLGA (50 / 50) at 1%, 2%, and 5% of PVA concentrations were shown in Fig. 3(b). The release behavior of the microspheres prepared with 1% and 2% PVA in the external phase exhibited a slightly lower initial burst release compared with microspheres prepared with 5% PVA in the external phase. The burst release was considered to be due to the diffusion of the surface-associated melittin or the larger diffusion area. It was found that the increase in the concentration of PVA in the external water phase leads to a higher viscosity of the phase. Consequently, smaller microspheres with a greater surface area were formed, resulting in microspheres having more surface-bound melittin.

Cumulative release(%)

Cumulative release(%)

100 80 60 40 20 0 0 5 10 15 20 25 30

Time(days)
Fig. 5. The effect of loading on the rate of release of melittin entrapped in microspheres: 11.5% w/w melittin (n), 7.7% w/w melittin ( ) and 3.0% w/w melittin (E) (n = 3).

318

F. Cui et al. / Journal of Controlled Release 107 (2005) 310319

a peptide with a low molecular weight and high aqueous solubility, tend to have a lower encapsulation efficiency and higher burst release. Although, in our study, the formulations were prepared without any surfactants or stabilizing agents in the primary emulsion, the encapsulation of melittin can reach up to 90% under optimum formulation conditions, and the burst release can also fall below 20%. In this case, the fact that melittin has 26 amino acid residues, 6 of which are positively charged, and there are no negative charges [21] over the pH range studied, was important. Owing to this property of melittin, an electric interaction between negatively charged COO in PLGA or PLA + and positively charged NH4 in melittin would have occurred, resulting in the formation of a barrier. It was assumed that this barrier efficiently prevented melittin leaving or diffusing from the microspheres during the preparation process. For the same reason, the melittin release profile from the micropsheres appeared primarily due to an erosion mechanism after burst release. Consequently, there was a lag time between the burst release and the erosion-controlled release phase. Moreover, the entrapped melittin could be released completely when the polymer was degraded completely, i.e., the melittin release reflected the polymer degradation rate, and complete release was achieved, as desired, over 30 days. This suggests that the interaction between negative and positive charges has no influence on the stability of melittin within the microspheres.

initial burst release was due to the fast diffusion of surface-located peptide. The lag phase was usually associated with a separation between the initial diffusion and the erosion-controlled release period. The length of the lag period and the release rate of melittin during the course of continuous release mainly depend on the rate of polymer degradation. The differences in the formulations could lead to differences in the structure, porosity, size and morphology of the microspheres, so that the initial burst release was influenced by these formulation parameters. Melittin-loaded PLGA (50 : 50) microspheres released melittin completely over 30 days following degradation of the polymer, as was the intention.

References
[1] V.R. Sinha, A. Trehan, Biodegradable microspheres for protein delivery, J. Control. Release 90 (2003) 261 280. [2] N. Angelika, A.T. David, Self-association and membrane-binding behavior of melittins containing trifluoroleucine, J. Am. Chem. Soc. 123 (2001) 7407 7413. [3] M.N. Jutel, W.J. Pichler, D. Skribic, A. Urwyler, C. Dahinden, Bee venom immunotherapy results in decrease of IL-4 and IL-5 and increase of IFN-gamma secretion in specific allergen-stimulated T cell cultures, J. Immunol. 154 (1995) 4187 4194. [4] G.R. Thomas, C.R. Hiley, Cardiovascular effect of intracerebroventricular bradykinin and melittin in the rat, J. Pharm. Pharmacol. 40 (1988) 721 723. [5] E.M. Pogliani, E. Cofrancesco, Thrombotic thrombocytopenic purpura, Haematologica 68 (1983) 546 557. [6] L. Xing, C. Dawei, X. Liping, Z. Rongqing, Effect of honey bee venom on proliferation of K1736M2 mouse melanoma cell in-vitro and growth of murine B16 melanomas in-vivo, J. Pharm. Pharmacol. 54 (2002) 1 8. [7] Y. Ogawa, H. Yamamoto, H. Okada, T. Yashiki, T. Shimamoto, A new technique to efficiently entrap leuprolide acetate into microcapsules of polylactic acid or copoly(lacide/glycolic)acid, Chem. Pharm. Bull. 5 (1988) 1095 1103. [8] J.L. Cleland, O.L. Johnson, S. Putney, A.J.S. Jones, Recombinant human growth hormone poly (lactide-co-glycolic acid) microsphere formulation development, Adv. Drug Deliv. Rev. 28 (1997) 71 84. [9] M. Tuncay, S. Calis, H.S. Kas, M.T. Ercan, R. Dalal, G.D. Green, H.B. Haimes, P.A. Jimenez, A.A. Hincal, Dicrofenac sodium incorporated PLGA (50 : 50) microspheres: formulation considerations and in vitro/in vivo evaluation, Int. J. Pharm. 195 (2000) 179 188. [10] K. Rader, A. Wildfeuer, F. Wintersberger, P. Bossinger, H.W. Mucke, Characterization of bee venom and its main compo nents by high-performance liquid chromatography, J. Chromatogr. 408 (1987) 341 348.

4. Conclusions In this study, the natural small peptide melittin was successively encapsulated into PLA or PLGA microspheres with high drug loading and encapsulation efficiency using the w / o / w emulsion solvent evaporation technique. By modifying the preparation conditions (such as the type of polymer, its concentration, PVA (stabilizer) concentration, volume of internal water phase and level of drug loading), it was possible to obtain melittin-loaded biodegradable microspheres with a high encapsulation efficiency, spherical shape and smooth surface. For the polymer of PLGA (50 / 50), the release profile of melittin was characterized by a small initial burst release and a lag period followed by a nearly constant release. The

F. Cui et al. / Journal of Controlled Release 107 (2005) 310319 [11] H. Sah, A new strategy to determine the actual protein content of poly (lactide-co-glycolide) microspheres, J. Pharm. Sci. 86 (1997) 1315 1318. [12] H. Jeffery, S.S. Davis, D.T. OHagan, The preparation and characterization of poly (lactide-co-glycolide) microspheres. II. The entrapment of a model protein using a (water-in-oil)in-water emulsion solvent evaporation technique, Pharm. Res. 10 (1993) 362 368. [13] Y.Y. Yang, T.S. Chung, N.P. Ng, Morphology, drug distribution, and in vitro release profiles of biodegradable polymeric microspheres containing protein fabricated by double-emulsion solvent extraction/evaporation method, Biomaterials 22 (2001) 231 241. [14] Y.Y. Yang, T.S. Chung, X.L. Bai, W.K. Chan, Effect of preparation conditions on morphology and release profiles of biodegradable polymeric microspheres containing protein fabricated by double-emulsion method, Chem. Eng. Sci. 55 (2000) 2223 2236. [15] G. Crotts, T.G. Park, Preparation of porous and nonporous biodegradable polymeric hollow microspheres, J. Control. Release 35 (1995) 91 105. [16] R.S. Pradhan, R.C. Vasavada, Formulation and in vitro release study on poly(d,l-lactide) microspheres containing hydrophilic

319

[17]

[18]

[19]

[20]

[21]

compounds: glycine homopeptides, J. Control. Release 30 (1994) 143 154. Y. Li, M. Ogris, J. Pelisek, W. Roedl, Stability and release characteristics of poly (d,l-lactide-co-glycolide) encapsulated CaPi-DNA coprecipitation, Int. J. Pharm. 269 (2004) 61 76. L.M. Sanders, J.S. Kent, G.I. McRae, B.H. Vickery, T.R. Tice, D.H. Lewis, Controlled release of a luteininzing hormonereleasing hormone analogue from poly(dl-lactide-co-glycolide) biodegradable polymers, J. Pharm. Sci. 82 (1993) 326 329. D.T. OHagan, H. Jeffery, S.S. Davis, The preparation and characterization of poly(lactide-co-glycolide) microcapsules: III. Microparticle/polymer degradation rates and the in vitro release of a model protein, Int. J. Pharm. 103 (1994) 37 45. P.T. Homayoun, D. Mandal, Landry, H. Komiskey, Controlled release of anti-cocaine catalytic antibody from biodegradable polymer microspheres, J. Pharm. Pharmacol. 55 (2003) 933 938. J. Bello, H.R. Bello, E. Granados, Conformation and aggregation of melittin: dependence on pH and concentration, Biochemistry 21 (1982) 461 465.