- Email: [email protected]
Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification
ACTBIO 3275
No. of Pages 10, Model 5G
3 July 2014 Acta Biomaterialia xxx (2014) xxx–xxx 1
Contents lists available at ScienceDirect
Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat 6 7
Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification
3 4 5 8
Q1
Feng Qi a,b,1, Jie Wu a,1, Tingyuan Yang a, Guanghui Ma a,c,⇑, Zhiguo Su a,c a State Key Laboratory of Biochemical Engineering, PLA Key Laboratory of Biopharmaceutical Production & Formulation Engineering, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, People’s Republic of China b University of the Chinese Academy of Sciences, Beijing 100049, People’s Republic of China c Collaborative Innovation Center of Chemical Science and Engineering (Tianjin), Tianjin 300072, People’s Republic of China
9 10 11 12 13 14
a r t i c l e
1 9 6 2 17 18 19 20 21
i n f o
Article history: Received 8 March 2014 Received in revised form 2 May 2014 Accepted 11 June 2014 Available online xxxx
22 23 24 25 26 27 28
Keywords: PLGA Controlled release Exenatide SPG membrane emulsification Microsphere
a b s t r a c t Poly(DL-lactic-co-glycolic acid) (PLGA) microspheres have been widely prepared by many methods, including solvent evaporation, solvent extraction and the co-solvent method. However, very few studies have compared the properties of microspheres fabricated by these methods. This is partly because the broad size distribution of the resultant particles severely complicates the analysis and affects the reliability of the comparison. To this end, uniform-sized PLGA microspheres have been prepared by Shirasu porous glass premix membrane emulsification and used to encapsulate exenatide, a drug for treating Type 2 diabetes. Based on this technique, the influences on the properties of microspheres fabricated by the aforementioned three methods were intensively investigated, including in vitro release, degradation and pharmacology. We found that these microspheres presented totally different release behaviors in vitro and in vivo, but exhibited a similar trend of PLGA degradation. Moreover, the internal structural evolution visually demonstrated these release behaviors. We selected for further examination the microsphere prepared by solvent evaporation because of its constant release rate, and explored its pharmacodynamics, histology, etc., in more detail. This microsphere when injected once showed equivalent efficacy to that of twice-daily injections of exenatide with no inflammatory response. Ó 2014 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46
47 48
1. Introduction
49
In the last few decades, microspheres based on biodegradable polymers such as poly(DL-lactic-co-glycolic acid) (PLGA) for sustained delivery of protein/peptide have been widely studied [1–3]. Some peptides, such as Lupron DepotÒ (leuprolide acetate) and Sandostatin LARÒ (octreotide acetate), have been incorporated into PLGA microspheres and commercialized as sustained-release systems [4]. Solvent evaporation, solvent extraction and co-solvent methods are the most popular approaches used to prepare the microspheres, because these processes are simple and convenient to control [5]. In the first two methods, an aqueous solution dissolving the drug is emulsified with an organic solution (oil phase)
50 51 52 53 54 55 56 57 58 59
⇑ Corresponding author at: State Key Laboratory of Biochemical Engineering, PLA
Q2
Key Laboratory of Biopharmaceutical Production & Formulation Engineering, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, People’s Republic of China. Tel./fax: +86 10 82627072. E-mail address: [email protected] (G. Ma). 1 These authors contributed equally to this work.
containing polymer. Methylene chloride (MC) is frequently used as the organic solvent in the evaporation method [6], and ethyl acetate (EA) is often used in the extraction method. For the co-solvent method, the drug is dissolved directly in two miscible organic solvents—typically MC and alcohol (methanol and ethanol)—without aqueous solution [7]. The use of different organic solvents results in microspheres with various characteristics. For instance, solvent removal rate affects the solidification, which is critical in determining the morphology, surface area and other aspects of microspheres [8]. Thus, discrepancies of the resultant particles in terms of, for example, release behavior and pharmacology will be generated. Studying this discrepancy can provide more important and general insights into the mechanisms on microsphere degradation and drug release in vitro/vivo, which is a key issue in developing long-term release systems. Although much effort has been devoted to microsphere preparations, relevant information about this discrepancy is scarce. This is partly because the broad size distribution resulting from conventional mechanical stirring will result in particles with poor reproducibility with respect to release
http://dx.doi.org/10.1016/j.actbio.2014.06.018 1742-7061/Ó 2014 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Qi F et al. Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2014.06.018
60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78
ACTBIO 3275
No. of Pages 10, Model 5G
3 July 2014 2 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100
F. Qi et al. / Acta Biomaterialia xxx (2014) xxx–xxx
behavior, drug efficacy, etc. [9–12]. Therefore, to control particle size, narrow down the size distribution and realize mass production, Shirasu porous glass (SPG) premix membrane emulsification has been employed [13]. Exenatide (synthetic exendin-4), a therapy for Type 2 diabetes mellitus (T2DM), was used as a model peptide in this study. It possesses 39 amino acids (H-His-Gly-Glu-Gly-Thr-Phe-Thr-SerAsp-Leu-Ser-Lys-Gln-Met-Glu-Glu-Glu-Ala-Val-Arg-Leu-Phe-Ile-GluTrp-Leu-Lys-Asn-Gly-Gly-Pro-Ser-Ser-Gly-Ala-Pro-Pro-Pro-Ser-NH2), shares about 53% homology with mammalian gut hormone (GLP1), and also possesses glucoregulatory actions including enhancement of insulin secretion, reduction of food intake, deceleration of gastric emptying and improvement of b-cell function [14,15]. Herein, based on SPG premix membrane emulsification, monodisperse exenatide-loaded PLGA microspheres were prepared by solvent evaporation, solvent extraction and co-solvent methods, respectively. Systematic research into the effect of preparation methods on the properties of microspheres was performed, including release behaviors, molecular weight (Mw) degradation, structural evolution, etc. Then, pharmacological aspects of the optical formulation were investigated, such as pharmacodynamics, immunohistochemistry analysis and inflammatory response.
101
2. Materials and methods
102
2.1. Materials
103
114
PLGA with a molar ratio of D,Llactide/glycolide 75/25 (Mw 13 kDa) was purchased from Lakeshore Biomaterials (Birmingham, AL, USA). Exenatide was provided by Hybio Pharmaceutical Co. Ltd. (Shenzhen, PR China). Poly(vinyl alcohol) (PVA-217, degree of polymerization 1700, degree of hydrolysis 88.5%) was provided by Kuraray (Tokyo, Japan). SPG membranes were purchased from SPG Technology Co. Ltd. (Miyazaki, Japan). The SPG premix membrane emulsification equipment (FMEM-500M) was designed by National Engineering Research Center for Biotechnology (NERCB, Beijing, PR China). Acetonitrile and trifluoroacetic acid (TFA) (both HPLC grade) were purchased from Dikma Co. Ltd. (Lake Forest, USA). All other reagents were analytical grade.
115
2.2. Preparation of microspheres
116
Microspheres prepared by solvent evaporation, solvent extraction and the co-solvent method are abbreviated as EVM, EXM and COM, respectively. Before preparation, SPG membrane 50.2 lm in size was installed in the equipment (Fig. 1a). For preparation of EVM
104 105 106 107 108 109 110 111 112 113
(Fig. 1b), 1 ml exenatide aqueous solution (3%, w/v, W1) was emulsified with 8 ml organic solvent (MC, O) containing PLGA (10%, w/ v) by homogenization (T18, IKA, Germany) at 18,000 rpm for 60 s to form W1/O. Next, the W1/O was stirred at 250 rpm for 1 min with external aqueous phase (W2) containing PVA (2%, w/v) and NaCl (0.5%, w/v) to form coarse W1/O/W2 emulsions. These were then poured into a premix reservoir and extruded through the SPG membrane by N2 pressure at 5 kPa to achieve uniform-sized droplets. After that, they were solidified at room temperature at 250 rpm for 5 h. Finally, the microspheres were collected and washed with distilled water five times by centrifugation for 3 min at 300 g. The washed microspheres were stored in 70 °C overnight, then lyophilized, and obtained after 48 h. The conditions for lyophilization were as follows: ice condenser 80 °C; vacuum 31 °C, 0.34 mbar. For preparation of EXM (Fig. 1b), EA was used as oil phase (O). The preparation process was the same as above. In addition, the uniform-sized droplets achieved by extrusion through SPG membrane were poured quickly into solidification solution with a large volume (1.6 l containing 0.9% (w/v) NaCl) under magnetic stirring at 250 rpm for 4 h to solidify the microspheres. The microspheres were obtained in the same way as above For preparation of COM (Fig. 1b), the exenatide powder was dissolved in mixed organic solvent (volume ratio of MC and methanol 6:2) containing PLGA (10%, w/v). The other steps were same as those for EVM.
121
2.3. Characterization of microspheres
147
2.3.1. Surface morphology observation and size distribution measurement The shape and surface morphology of PLGA microspheres were observed by scanning electron microscopy (SEM) with a JSM-6700F (JEOL, Japan). The particle size and size distribution was measured with a Mastersizer 2000 (Malvern, UK). The size distribution was referred as the Span value and was calculated as follows:
148
122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146
149 150 151 152 153 154 155
156
117 118 119 120
Span ¼
Dv;90% Dv;10% ; Dv;50%
where Dv,90%, Dv,50% and Dv,10% are volume size diameters at 90%, 50% and 10% of the cumulative volume, respectively. The smaller the Span value, the narrower the size distribution.
Fig. 1. (a) Schematic depiction of the SPG premix membrane emulsification equipment; (b) schemes of the EVM, EXM and COM preparation.
Please cite this article in press as: Qi F et al. Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2014.06.018
158
159 160 161
ACTBIO 3275
No. of Pages 10, Model 5G
3 July 2014 F. Qi et al. / Acta Biomaterialia xxx (2014) xxx–xxx 162 163 164 165 166 167 168 169 170 171 172 173 174
2.3.2. Drug-loading capacity and encapsulation efficiency measurement Microspheres (5 mg) were dissolved in a mixed solution composed of 150 ll acetonitrile and 850 ll 0.01 M HCl. The concentration of exenatide was determined by a reverse-phase HPLC (RP-HPLC) system at room temperature with a C18 (250 mm 4.6 mm 5 lm, Syncronis, Thermo) chromatographic column. The chromatography was performed with a linear gradient elution from 20% to 60% acetonitrile in ultrapure water containing 0.1% TFA for 16 min. The flow rate was 1 ml min1, and the UV absorbance was set at 214 nm. The loading capacity and encapsulation efficiency of the microspheres were calculated by the following equations:
175 177
Loadingcapacityðlg=mgÞ ¼
Mass of drug in microspheres Mass of microspheres
3
2.4.2. Pharmacodynamics study Male kk-ay mice (9–10 weeks old), T2DM model mice, were assigned into three groups (n = 6 per group): exenatide solution group (twice daily injection for 14 days, 50 lg kg1), EVM (single subcutaneous injection of EVM, 40 mg kg1 containing 1.44 mg kg1 exenatide, equivalent to the dose of twice-daily injection of exenatide solution for 2 weeks) and saline group (single subcutaneous injection, control). The blood samples were obtained after fasting 4 h. A one-touch blood glucose meter (ACCU-CHEKÒ Performa) was used to measure the blood glucose. The area under the concentration–time curve (AUC) was estimated by the linear trapezoidal method. The in vivo results were analyzed as mean ± SE. Statistics were performed using GraphPad Prism 5 (San Diego, CA). Comparisons between groups were analyzed by two-way ANOVA analysis followed by Tukey’s test, and P < 0.05 was considered statistically significant.
220
2.4.3. Histological and immunohistochemistry studies Histological study was used to evaluate the safety of EVM in vivo. Subcutaneous tissues were removed from the injection sites. The tissue samples were fixed in 10% formalin, and sections were immersed in paraffin and cut using a microtome. The tissues were stained with hematoxylin and eosin. Insulin immunohistochemistry study was performed to confirm the insulinotropic bioactivity of the drug. Tissue samples were washed with PBS, dehydrated by graded alcohols and xylene, and embedded in paraffin wax. The embedded tissues were sectioned at 4 lm, and attached to polylysine-coated slides. Endogenous peroxidase blocking process was performed after 20 min pretreatment at room temperature. Sections were reacted with anti-insulin antibody diluted at a ratio of 1/800 as a primary antibody and incubated at 4 °C overnight. After washing the sections with PBS three times, secondary antibody was added and incubated for 4 h at room temperature before washing with PBS. Finally, sections were stained with DAB, counterstained in real hematoxylin and 0.3% ammonia water. Proliferating cell nuclear antigen (PCNA) immunohistochemistry was performed by the same steps. Photomicrographs of the histological and immunohistochemical slices were taken by an Olympus microscope (Tokyo, Japan). The images were analyzed by Image J version 1.48 (NIH Image, Maryland, MD).
236
3. Results and discussion
260
221 222 223 224 225 226 227 228 229 230 231 232 233 234 235
178
Encapsulationefficiencyð%; w=wÞ ¼ 180
181 182 183
184 185 186 187 188 189 190 191 192
Loading capacity 100% Theoretical loading capacity
2.3.3. Circular dichroism (CD) measurement A detailed description of method used for CD measurement is included in the Supporting Information. 2.3.4. In vitro drug release and polymer degradation studies PLGA microspheres (10 mg) were dispersed in 1 ml 10 mM phosphate buffer saline (PBS) medium (pH 7.4) and incubated under agitation at 37 °C. At each time interval, supernatants were collected by centrifugation for 3 min at 300 g and replaced with fresh buffer of equal volume. The concentration of exenatide in the supernatant was determined by Micro BCA Protein Assay Kit. The method for studying polymer degradation is shown in the Supporting Information.
237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253
200
2.3.5. Observation of drug distribution inside microspheres Super Fluor 488 SE (Fanbo Biochemicals, Beijing, PR China) was used to label exenatide to observe drug distribution within microspheres during incubation. The microsphere preparation and in vitro release experiment were the same as above. At predetermined times, a small amount of microspheres were removed from the samples and observed by confocal laser scanning microscopy (CLSM, TCS SP 5, Leica) at an excitation wavelength of 488 nm.
201
2.4. Animal experiments and statistics
3.1. Characteristics of microspheres
261
202
The animal experiments were consistent with the guidelines set by the National Institutes of Health (NIH Publication No. 85-23, revised 1985) and were approved by the Experimental Animal Ethics Committee, Beijing. The animal experiments were performed at Beijing Dingguo Changsheng Biotechnology Co. Ltd.
Uniform-sized exenatide-loaded PLGA microspheres were successfully prepared by SPG premix membrane emulsification combined with solvent evaporation, solvent extraction and co-solvent methods. As shown in Fig. 2a–c, all the microspheres had smooth external surfaces with a few pores, attributed to the drug diffusion from particle core to the surface during solidification [16]. The sizes of the three resultant microspheres were 20.63, 22.38 and 21.50 lm with small ranges (Table S1 in the Supporting Information), suggesting a narrow size distribution (Fig. 2d). EVM and EXM were prepared as double emulsions (W/O/W) with different organic solvents, but the encapsulation efficiency of the former (92.51 ± 3.41%, MC as oil phase) was higher than the later (84.42 ± 3.67%, EA as oil phase), as was the loading capacity (34.23 ± 1.31 vs 31.64 ± 1.43 lg mg1). The reason was that due to volumes of solidification solution added in emulsion droplets of EXM, the hydrophilic drug diffused into the external water phase more easily during solidification. The highest encapsulation efficiency and loading capacity (98.48 ± 4.94% and 35.21 ± 1.82 lg mg1) was achieved by co-solvent method (O/W), because
262
193 194 195 196 197 198 199
203 204 205 206
207 208 209 210 211 212 213 214 215 216 217 218 219
2.4.1. Pharmacokinetics study Male SD rats (250–300 g) were divided into four groups (n = 6 per group): exenatide solution, EVM, EXM and COM groups. The rats in the exenatide solution group were subcutaneously injected with exenatide at a dose of 36 lg per rat. The other groups were subcutaneously injected with microspheres containing exenatide at a dose of 1 mg per rat (equivalent to the dose of a twice-daily injection of exenatide solution for 2 weeks). Plasma was obtained at the same sampling time as in vitro release. The establishment of in vitro/in vivo correlation is included in the Supporting Information. The exenatide concentration in plasma was analyzed with an Exendin-4 EIA kit (Phoenix Pharmaceuticals, CA, USA) according to the manufacturer’s instructions.
Please cite this article in press as: Qi F et al. Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2014.06.018
254 255 256 257 258 259
263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280
ACTBIO 3275
No. of Pages 10, Model 5G
3 July 2014 4
F. Qi et al. / Acta Biomaterialia xxx (2014) xxx–xxx
Fig. 2. SEM images of (a) EVM, (b) EXM and (c) COM; (d) size distribution. Scale bar: 10 lm.
283
methanol, miscible with MC, could dissolve the drug, making it distribute uniformly in the oil phase, and leading to a fast polymer precipitation favorable for high encapsulation efficiency.
284
3.2. Drug stability
285
Using organic solvent may lead to denaturation of encapsulated peptide/protein during preparation, owing to formation of an oil– water interface. For example, MC had a denaturing effect on the stability of protein C, and the effect would be worse if EA were used [17]. EVM and EXM were prepared using MC and EA as the organic solvent, respectively. In COM preparation, MC and methanol were both used. To study the stability of exenatide during preparation, CD spectroscopy was used to monitor its secondary structure changes. The peptide was extracted from the microspheres after lyophilization and compared with the native one. As shown in Fig. S1 (Supporting Information), all the spectra show identical curves with two negative minima around 208 and 222 nm, indicating the predominant a-helix and b-structure. Thus, the stability of exenatide was not influenced by the preparation method used.
281 282
286 287 288 289 290 291 292 293 294 295 296 297 298
299
3.3. In vitro release
300
It is well known that drug-release behavior is significantly affected by preparation method [18,19]. Similarly, in our study, all the release profiles were completely different (Fig. 3). EVM exhibited a low burst initial release within 24 h (25%), followed by constant release until the end. EXM showed a plateau with the highest burst (45%), and after 5 weeks the release rate increased. On the other hand, COM exhibited a similar release behavior as EXM but with the slowest burst (13%). To further investigate the mechanisms, studies of the morphological evolution and polymer degradation of microspheres were undertaken as follows.
301 302 303 304 305 306 307 308 309 310
Fig. 3. In vitro release profiles (n = 3, mean ± SD).
3.4. Structural evolution and polymer degradation
311
PLGA microspheres release drug via gradual erosion of their matrix after hydration and cleavage of hydrolytically sensitive bonds in the polymer chains [20]. Therefore, the morphological and inner structural evolution can reflect the microsphere degradation states, and the molecular weight decrease can quantitatively characterize the polymer degradation.
312
3.4.1. Morphological evolution As illustrated in Fig. S2 (Supporting Information), irrespective of the preparation method used, we observed the general trend that the external surfaces of the microspheres became coarse with large pores, and more pores were formed by day 14. Moreover, the pores disappeared gradually, but all of them still maintained their integrity up to 60 days. The reason was that the reduced glass transition temperature (Tg) caused by decreased Mw, promoted the mobility of degraded oligomers and accelerated their rearrangement by plasticization [21]. For the same reason, the microspheres did not collapse even up to 100 days. However, they all became smaller,
318
Please cite this article in press as: Qi F et al. Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2014.06.018
313 314 315 316 317
319 320 321 322 323 324 325 326 327 328
ACTBIO 3275
No. of Pages 10, Model 5G
3 July 2014 F. Qi et al. / Acta Biomaterialia xxx (2014) xxx–xxx 329 330 331 332 333 334 335 336 337 338 339 340
341 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393
and some were aggregated and had coarse surfaces. The maintenance of microsphere integrity might be one of the reasons for the incomplete release. In conclusion, no significant differences of the external surfaces were observed. 3.4.2. Polymer degradation Polymer degradation is the result of hydrolysis of ester bonds through the whole polymer matrix [22]. As shown in Fig. S3 (Supporting Information), surprisingly, all of the polymers followed the similar Mw degradation profile. The polymer Mw decreased almost linearly with time, suggesting pseudo-first-order kinetics for the polymer degradation during incubation. The Mw decrease was calculated as follows:
Mw ðtÞ ¼ Mw0 expðkd tÞ
ð1Þ
where Mw0 is the initial Mw of polymer and kd is the pseudo-firstorder degradation rate constant of the polymer [23]. The values of kd calculated by Eq. (1) from the experimental results were almost identical (Fig. S3). This indicated that the degradations proceeded via the same mechanism, and were not strongly affected by the preparation method. 3.4.3. Inner structural evolution The inner structural changes that could affect in vitro release behavior [8,24] were reflected in the drug distribution variations visualized by CLSM. In order to explore the mechanism of internal microsphere degradation, labeled exenatide was encapsulated in PLGA microspheres. As illustrated in Fig. 4, EVM exhibited non-uniform drug distribution with many large non-homogeneous aggregations, as the peptide was dissolved in the inner aqueous droplets within the matrix of particle. EXM also presented the same distribution, but it was obvious that there were more drugs on or near the surfaces. This was because that when the uniform-sized emulsion droplets were added to the solidification solution, they were exposed to volumes of water, resulting in the peptide diffusing near or on the particle surface. This also explained the reason for its highest initial burst release. As for COM, exenatide was uniformly distributed inside these microsphere, since the drug was dissolved in the oil phase. During incubation, the microspheres degraded gradually. Simultaneously, the drug domains changed into aqueous pores showing no green color [11,25]. As incubation proceeded, more pores were generated inside EVM, and contacted the external phase. Thus, water and degraded oligomers diffused into and out of the matrix easily, leading to a constant release rate. For EXM, the drug near the surface quickly diffused into medium, resulting in a high burst initial release. In addition, the internal structural changes were almost the same as those in EVM. In COM, however, at an early phase (day 14), the drug distribution was still relatively uniform except for generation of small pores near the surfaces, indicating the drug nearby was released first. At day 30, some drug remained in the core of microspheres; hence, the release rate was very slow in this phase. Afterwards, the drug in the core began to diffuse out of the microspheres showing a green margin, resulting in a high release rate. Compared with the corresponding morphological evolution, the internal structure clearly evolved, because the pH inside the particle decreased due to local polymer degradation. Whereas pH at the surface remained unchanged, the morphology could only be changed by Tg caused by degradation of the whole polymer [26]. The internal structural evolution reflected by drug distribution clearly demonstrated and explained the differences in release behaviors. Above all, it was found that although the drug release was controlled by polymer degradation, the internal structural changes of microspheres rather than the decrease of polymer Mw played the most important role.
5
3.5. Pharmacokinetics
394
Pharmacokinetics using SD rats was evaluated to further explore the differences between the microsphere preparations in vivo (Fig. 5, Table S2). EVM achieved maximal plasma drug concentration (136.16 ± 19.11 ng ml1) within 2 h after injection, which was significantly lower than the exenatide solution group (239.83 ± 15.48 ng ml1). Furthermore, it showed constant release within 14 days, and then leveled off. The EXM group presented a quick release with high concentration in 1 h, and achieved the highest value (190.09 ± 16.04 ng ml1) within 2 h. Moreover, its plasma drug level was lower than that of EVM almost throughout the entire investigations period except the initial hours, in accordance with the in vitro release behavior. On the other hand, COM showed its highest value (60.12 ± 12.62 ng ml1) at 1 h. Despite this, its maximal value was the lowest among the three. In addition, its drug concentration presented the lowest level within the first 14 days. Afterwards, however, it became very high until the end. The AUC01d value of the exenatide solution group (25.01 ± 1.66 ng day ml1) was significantly higher than the group Q3 COM (9.46 ± 0.59 ng day ml1), showed no difference compared with EVM (20.14 ± 2.29 ng day ml1), and was significantly lower than EXM (33.65 ± 3.12 ng day ml1). The AUC030d values of EVM (188.89 ± 19.45 ng day ml1) and COM (255.01 ± 72.60 ng day ml1) were significant higher than EXM (149.74 ± 13.85 ng day ml1). According to the above results, EVM was the best formulation for a long-term effective system, because it possessed both the high AUC030d value and constant release rate both in vitro or in vivo. In the following experiments, EVM was employed to further investigate the relevant pharmacology. The in vitro/in vivo correlations of EVM, EXM and COM were established by the profiles of cumulative release in vitro and in vivo (Fig. 5b), exhibiting a good linear regression correlation. Based on these results, it was feasible to predict the release in vivo by the established correlation according to the in vitro release.
395
3.6. The pharmacological study of EVM
430
3.6.1. Hypoglycemic efficacy Kk-ay mice were used as a T2DM model to evaluate the pharmacodynamics. After injection, the blood glucose level in exenatide solution and in the EVM group were significantly lower than that in control (saline group), and remained lower for first 14 days (Fig. 6a). However, when the mice in the solution group stopped being injected from day 15, the blood glucose returned to the control level gradually. In contrast, the EVM group still presented significantly low blood glucose throughout the entire experimental period. The trend of hypoglycemic efficacy was well explained by the pharmacokinetics of EVM, i.e. the drug concentration in blood was high for the first 14 days, leading to a significantly low blood glucose level. Afterwards, the release leveled off, and the efficacy gradually became weak. The AUCblood glucose 014d values of exenatide solution (186.73 ± 40.35 mmol l1 day1) and EVM (154.47 ± 75.90 mmol l1 day1) group were significantly smaller than that of control (296.84 ± 41.40 mmol l1 day1), indicating the former two groups both presented hypoglycemic efficacy. Furthermore, there was no significant difference between them, suggesting they offered equivalent efficacy.
431
3.6.2. Insulin immunohistochemistry analysis After 14 days of treatment, the mice pancreatic tissues were removed to analyze the insulinotropic bioactivity of exenatide by insulin immunohistochemistry, which reflected the insulin secretion in pancreas islets visually [27,28].
451
Please cite this article in press as: Qi F et al. Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2014.06.018
396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429
432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450
452 453 454 455
ACTBIO 3275
No. of Pages 10, Model 5G
3 July 2014 6
F. Qi et al. / Acta Biomaterialia xxx (2014) xxx–xxx
EVM
EXM
COM
Day 0
Day 14
Day 30
Day 60
Fig. 4. CLSM images for inner structural (drug distribution) evolutions of EVM, EXM and COM at different incubation times (day 0, 14, 30 and 60). Scale bar: 10 lm.
Fig. 5. (a) Plasma exenatide concentration vs. time profiles of EVM, EXM and COM (n = 6); the insert presents the plasma exenatide concentrations vs. time within 24 h. (b) In vitro/in vivo correlations of EVM, EXM and COM.
456 457 458 459 460 461 462
As shown in Fig. 7a–c, in the cytoplasm of the endocrine islet, positive signals (dark yellow) with larger area were obvious in the exenatide solution and EVM groups, indicative of higher insulin secretion and more b-cells. Semiquantitative analysis of the ratios (b-cell area to islet area) (Fig. 7d) also demonstrated that insulin secretions in these two groups were significantly higher than in control, in line with the result of hypoglycemic efficacy. This result
also suggested that EVM achieved the same insulinotropic activity as twice-daily injection of exenatide solution for 2 weeks.
463
3.6.3. PCNA immunohistochemistry analysis In addition to hypoglycemic efficacy and insulinotropic activity, exenatide is also capable of increasing b-cell mass through stimulation of b-cell neogenesis, proliferation and suppression of b-cell
465
Please cite this article in press as: Qi F et al. Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2014.06.018
464
466 467 468
ACTBIO 3275
No. of Pages 10, Model 5G
3 July 2014 F. Qi et al. / Acta Biomaterialia xxx (2014) xxx–xxx
7
Fig. 6. (a) Blood glucose vs. time profiles of exenatide solution, EVM and saline group. (b) AUC values of blood glucose concentration in three groups (0–14 days) (AUCblood Statistical differences were tested using two-way ANOVA followed by Tukey’s test. Differences were termed statistically significant at ⁄P < 0.05, ⁄⁄P < 0.01 vs. saline group. Data are mean ± SE. n = 6.
glucose 014d).
Fig. 7. Insulin immunohistochemistry of pancreas tissues from kk-ay mice in (a) exenatide solution, (b) EVM and (c) saline group after 14 days of treatment, the dark yellow areas indicate the insulin-positive signals. (d) Percentages of insulin-positive b-cell area to the islet area (n = 4). Scale bar: 200 lm. Statistical differences were tested using two-way ANOVA followed by Tukey’s test. Differences were termed statistically significant at ⁄P < 0.05 vs. saline group. Data are mean ± SE.
469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488
apoptosis, which is the biggest difference compared with chemical drugs such as metformin and sulfonylurea treatment for T2DM [29]. Because PCNA is distinguished by its apparent association with cell division [30], its immunohistochemistry can be used to indicate cell neogenesis including b-cells in pancreas islets. As illustrated in Fig. 8a,b, proliferative cells (dark brown points) were distributed sporadically within pancreas islets (in frame) in the exenatide solution and EVM groups. However, the control group shows few dark brown points (Fig. 8c). Semiquantitative analysis of the ratios (PCNA-positive cells to all cells) (Fig. 8d) indicated that the numbers of proliferative cells in the exenatide solution and EVM groups were significant more than that in the saline group, and that some of them were b-cells. Furthermore, the area of islet was smaller than with the other two groups. Although PCNA immunohistochemistry indicates no direct b-cell neogenesis, it was suggested that the pre-existing b-cell in the islets might proliferate slowly [31], and there were more in the former two groups than in the control. As we discussed above, in Fig. 7, the positive areas indicated not only the insulin secretion, but also the number of b-cells. Combined with the results of insulin and PCNA immuno-
histochemistry, it was concluded that the exenatide and EVM group both improved pancreatic function.
489
3.7. Histological study
491
After administration to SD rats, the injection sites were removed periodically to perform the histological study. As shown in Fig. 9, compared with control, no inflammatory response was observed in the tissue-injected exenatide solution. Similarly, the EVM group also presented no significant inflammatory reaction in any phase (day 7–30). This is an exciting result compared with other works that also used PLGA as material to encapsulate drugs, because accumulation of acidic oligomers degraded by PLGA might lead to an inflammatory response [32,33]. One of the possible reasons was that the high LA/GA (75/25) ratio caused slow degradation PLGA, generating lower levels of acidic oligomers. Meanwhile, the oligomers would be quickly transported to further adjust the local acidity. Another reason was that a only a small number of microspheres needed to be injected to achieve an efficacious dose. Thus, the inflammatory response was very weak or
492
Please cite this article in press as: Qi F et al. Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2014.06.018
490
493 494 495 496 497 498 499 500 501 502 503 504 505 506
ACTBIO 3275
No. of Pages 10, Model 5G
3 July 2014 8
F. Qi et al. / Acta Biomaterialia xxx (2014) xxx–xxx
(b)
(a)
(d)
(c)
Fig. 8. PCNA immunohistochemistry of pancreas tissues from kk-ay mice in (a) exenatide solution, (b) EVM and (c) saline group after 14 days of treatment; (d) percentages of PCNA-positive cells to all cells in islet (n = 4). The dark brown points which the arrow points to in the frame are PCNA-positive signals. Statistical differences were tested using two-way ANOVA followed by Tukey’s test. Differences were termed statistically significant at ⁄P < 0.05 vs. saline group. Scale bar: 100 lm.
Control
Exenatide solution
Exenatide 7d
14 d
30 d
Fig. 9. Histology of injection sites after s.c. administration with exenatide solution and EVM. Top panel shows normal tissue (control) and tissue injected with exenatide solution. Bottom panel shows tissues injected with EVM after 7, 14 and 30 days. Scale bar: 200 lm.
507 508
even negligible. In conclusion, the EVM was acceptable and safe for administration.
509
3.8. Morphological evolution of microspheres in vivo
510
It is well known that after injection, large PLGA microspheres will remain in place under the skin and degrade slowly affected by biological action [35]. During this degradation, exenatide is released continuously into circulation as a result of polymer erosion [36–38], a process that can be mirrored by the biodegradation state of the microspheres. Thus, the microspheres injected were removed periodically to observe any morphological changes by SEM. After injection of EVM, a nodule was formed under the skin at the injection site. At day 7, the EVM remained as individuals, although they were no longer spherical (Fig. 10). After 14 and
511 512 513 514 515 516 517 518 519 520
22 days, the microspheres agglomerated and interspersed among surrounding connective tissues. At the end of 1 month, from the exterior, no subcutaneous nodule could be detected in any rat (figure not shown). Internal observation, however, showed that the microspheres kept interspersing, invaded deeper into tissue, and could not be distinguished clearly. This was because in the physiological environment, the area of microspheres injected into the body decreased owing to degradation, and the surrounding tissues integrated with the microspheres and reorganized [34]. As the injection time increased, the phenomenon was more obvious. At the later stage, the microspheres became submerged by the connective tissue and degraded further. This study showed the biodegradation states of EVM with a size of 20 lm in vivo. According to our observations, it might not be a significant burden for the body in the late phase of injection.
Please cite this article in press as: Qi F et al. Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2014.06.018
521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536
ACTBIO 3275
No. of Pages 10, Model 5G
3 July 2014 F. Qi et al. / Acta Biomaterialia xxx (2014) xxx–xxx
(a)
(b)
(c)
(d)
9
Fig. 10. Morphological changes of EVM in vivo: (a) day 7, (b) day 14, (c) day 22 and (d) day 30. Scale bar: 20 lm.
537
4. Conclusion
Appendix B. Supplementary data
566
538
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.actbio.2014. 06.018.
567
References
570
554
Exenatide-loaded PLGA microspheres were prepared by SPG premix membrane emulsification combined with solvent evaporation, solvent extraction and co-solvent methods. All of the microspheres had a narrow size distribution and an average size of 20 lm. EVM presented a constant release, and EXM showed the highest burst release. As for COM, it exhibited a slow release, followed by a fast release in the later phases. The preparation methods did not influence the polymer Mw degradation, but affected the internal structural evolution that played an important role in the in vitro release behaviors. These relevant studies provide a new guidance for microsphere preparation in pharmaceutics. EVM was the best formulation due to its constant release rate in vivo over 14 days. The pharmacodynamics showed that the EVM could effectively reduce the blood glucose with no inflammatory response, and also possessed the same function in the pancreas as exenatide solution. Above all, it is feasible for a single injection of EVM to replace a twice-daily injection of exenatide for 2 weeks.
555
Acknowledgements
[1] Aubert-Pouessel A, Venier-Julienne MC, Clavreul A, Sergent M, Jollivet C, Montero-Menei CN, et al. In vitro study of GDNF release from biodegradable PLGA microspheres. J Control Release 2004;95:463–75. [2] Jiang WL, Gupta RK, Deshpande MC, Schwendeman SP. Biodegradable poly(lactic-co-glycolic acid) microparticles for injectable delivery of vaccine antigens. Adv Drug Deliv Rev 2005;57:391–410. [3] Eswaramoorthy R, Chang CC, Wu SC, Wang GJ, Chang JK, Ho ML. Sustained release of PTH(1–34) from PLGA microspheres suppresses osteoarthritis progression in rats. Acta Biomater 2012;8:2254–62. [4] Ahn JH, Park EJ, Lee HS, Lee KC, Na DH. Reversible blocking of amino groups of octreotide for the inhibition of formation of acylated peptide impurities in poly(lactide-co-glycolide) delivery systems. AAPS PharmSciTech 2011;12: 1220–6. [5] Ye ML, Kim S, Park K. Issues in long-term protein delivery using biodegradable microparticles. J Control Release 2010;146:241–60. [6] Elkharraz K, Ahmed AR, Dashevsky A, Bodmeier R. Encapsulation of watersoluble drugs by an o/o/o-solvent extraction microencapsulation method. Int J Pharm 2011;409:89–95. [7] Lassalle V, Ferreira ML. PLA nano- and microparticles for drug delivery: an overview of the methods of preparation. Macromol Biosci 2007;7:767–83. [8] Yang YY, Chung TS, Bai XL, Chan WK. Effect of preparation conditions on morphology and release profiles of biodegradable polymeric microspheres containing protein fabricated by double-emulsion method. Chem Eng Sci 2000;55:2223–36. [9] Mao SR, Xu J, Cai CF, Germershaus O, Schaper A, Kissel T. Effect of WOW process parameters on morphology and burst release of FITC-dextran loaded PLGA microspheres. Int J Pharm 2007;334:137–48. [10] Sah HK, Toddywala R, Chien YW. Biodegradable microcapsules prepared by a W/O/W technique—effects of shear force to make a primary W/O Emulsion on their morphology and protein release. J Microencapsul 1995;12:59–69. [11] Yan CH, Resau JH, Hewetson J, West M, Rill WL, Kende M. Characterization and morphological analysis of protein-loaded poly(lactide-co-glycolide) microparticles prepared by water-in oil-in-water emulsion technique. J Control Release 1994;32:231–41. [12] Wang LY, Gu YH, Zhou QZ, Ma GH, Wan YH, Su ZG. Preparation and characterization of uniform-sized chitosan microspheres containing insulin
571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606
539 540 541 542 543 544 545 546 547 548 549 550 551 552 553
556 557 558 559 560
We thank the NSFC/RGC Joint Research Scheme sponsored by the Research Grants Council of Hong Kong and the National Natural Q4 Science Foundation of China (No. 21161160555) and National NatQ5 ural Science Foundation of China (No. 21336010) for the financial support provided.
561
Appendix A. Figures with essential color discrimination
562
Certain figures in this article, particularly Figs. 1–9 are difficult to interpret in black and white. The full color images can be found in the on-line version, at http://dx.doi.org/10.1016/ j.actbio.2014.06.018.
563 564 565
Please cite this article in press as: Qi F et al. Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2014.06.018
568 569
ACTBIO 3275
No. of Pages 10, Model 5G
3 July 2014 10 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646
[13]
[14]
[15]
[16]
[17]
[18] [19]
[20]
[21]
[22] [23]
[24] [25]
F. Qi et al. / Acta Biomaterialia xxx (2014) xxx–xxx by membrane emulsification and a two-step solidification process. Colloids Surf B 2006;50:126–35. Qi F, Wu J, Fan QZ, He F, Tian GF, Yang TY, et al. Preparation of uniform-sized exenatide-loaded PLGA microspheres as long-effective release system with high encapsulation efficiency and bio-stability. Colloids Surf B 2013;112:492–8. Drucker DJ, Buse JB, Taylor K, Kendall DM, Trautmann M, Zhuang DL, et al. Exenatide once weekly versus twice daily for the treatment of type 2 diabetes: a randomised, open-label, non-inferiority study. Lancet 2008;372:1240–50. Zhou J, Cai Z-H, Li L, Kou C, Gao Y-F. Preparation and PEGylation of exendin-4 peptide secreted from yeast Pichia pastoris. Eur J Pharm Biopharm 2009;72:412–7. Ibrahim MA, Ismail A, Fetouh MI, Gopferich A. Stability of insulin during the erosion of poly(lactic acid) and poly(lactic-co-glycolic acid) microspheres. J Control Release 2005;106:241–52. Zambaux MF, Bonneaux F, Gref R, Dellacherie E, Vigneron C. Preparation and characterization of protein C-loaded PLA nanoparticles. J Control Release 1999;60:179–88. Sinha VR, Trehan A. Biodegradable microspheres for protein delivery. J Control Release 2003;90:261–80. Miyajima M, Koshika A, Okada J, Kusai A, Ikeda M. The effects of drug physicochemical properties on release from copoly (lactic/glycolic acid) matrix. Int J Pharm 1998;169:255–63. Samadi N, Abbadessa A, Di Stefano A, van Nostrum CF, Vermonden T, Rahimian S, et al. The effect of lauryl capping group on protein release and degradation of poly(D, L-lactic-co-glycolic acid) particles. J Control Release 2013;172:436–43. Wei GB, Pettway GJ, McCauley LK, Ma PX. The release profiles and bioactivity of parathyroid hormone from poly(lactic-co-glycolic acid) microspheres. Biomaterials 2004;25:345–52. Gopferich A. Mechanisms of polymer degradation and erosion. Biomaterials 1996;17:103–14. Siepmann J, Faisant N, Akiki J, Richard J, Benoit JP. Effect of the size of biodegradable microparticles on drug release: experiment and theory. J Control Release 2004;96:123–34. Freiberg S, Zhu X. Polymer microspheres for controlled drug release. Int J Pharm 2004;282:1–18. Yushu H, Venkatraman S. The effect of process variables on the morphology and release characteristics of protein-loaded PLGA particles. J Appl Polym Sci 2006;101:3053–61.
[26] Schliecker G, Schmidt C, Fuchs S, Kissel T. Characterization of a homologous series of D, L-lactic acid oligomers; a mechanistic study on the degradation kinetics in vitro. Biomaterials 2003;24:3835–44. [27] Hahn SK, Kong JH, Oh EJ, Chae SY, Lee KC. Long acting hyaluronate—exendin 4 conjugate for the treatment of type 2 diabetes. Biomaterials 2010;31:4121–8. [28] Chae SY, Jin CH, Shin JH, Son S, Kim TH, Lee S, et al. Biochemical, pharmaceutical and therapeutic properties of long-acting lithocholic acid derivatized exendin-4 analogs. J Control Release 2010;142:206–13. [29] Xu G, Kaneto H, Lopez-Avalos MD, Weir GC, Bonner-Weir S. GLP-1/exendin-4 facilitates beta-cell neogenesis in rat and human pancreatic ducts. Diabetes Res Clin Pract 2006;73:107–10. [30] Mathews MB, Bernstein RM, Franza BR, Garrels JI. Identity of the proliferating cell nuclear antigen and cyclin. Nature 1984;309:374–6. [31] Fujisawa H, Zhang Z, Sun W, Huang MS, Kobayashi J, Yasuda H, et al. Histopathological changes in the pancreas from a spontaneous hyperglycemic cynomolgus monkey. J Toxicol Pathol 2012;25:215–9. [32] Yang JA, Kim H, Park K, Hahn SK. Molecular design of hyaluronic acid hydrogel networks for long-term controlled delivery of human growth hormone. Soft Matter 2011;7:868–70. [33] Zolnik BS, Burgess DJ. Evaluation of in vivo-in vitro release of dexamethasone from PLGA microspheres. J Control Release 2008;127:137–45. [34] Kim BS, Oh JM, Hyun H, Kim KS, Lee SH, Kim YH, et al. Insulin-loaded microcapsules for in vivo delivery. Mol Pharm 2009;6:353–65. [35] Blanco MD, Sastre RL, Olmo R, Teijon C, Muniz E, Teijon JM. 5-Fluorouracil plasma levels and biodegradation of subcutaneously injected drug-loaded microspheres prepared by spray-drying poly(D, L-lactide) and poly(D, L-lactideco-glycolide) polymers. Int J Pharm 2007;338:180–90. [36] DeYoung MB, MacConell L, Sarin V, Trautmann M, Herbert P. Encapsulation of exenatide in poly-(D, L-Lactide-Co-Glycolide) microspheres produced an investigational long-acting once-weekly formulation for Type 2 diabetes. Diabetes Technol Ther 2011;13:1145–54. [37] Wysham C, Grimm M, Chen S. Once weekly exenatide: efficacy, tolerability and place in therapy. Diabetes Obes Metab 2013;15:871–81. [38] Chang LC, Yang CY, Chua ACN, Lin YJ, Lai SM. Sustained release of transgenic human factor IX: preparation, characterization, and in vivo efficacy. Mol Pharm 2011;8:1767–74.
Please cite this article in press as: Qi F et al. Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2014.06.018
647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683
No. of Pages 10, Model 5G
3 July 2014 Acta Biomaterialia xxx (2014) xxx–xxx 1
Contents lists available at ScienceDirect
Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat 6 7
Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification
3 4 5 8
Q1
Feng Qi a,b,1, Jie Wu a,1, Tingyuan Yang a, Guanghui Ma a,c,⇑, Zhiguo Su a,c a State Key Laboratory of Biochemical Engineering, PLA Key Laboratory of Biopharmaceutical Production & Formulation Engineering, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, People’s Republic of China b University of the Chinese Academy of Sciences, Beijing 100049, People’s Republic of China c Collaborative Innovation Center of Chemical Science and Engineering (Tianjin), Tianjin 300072, People’s Republic of China
9 10 11 12 13 14
a r t i c l e
1 9 6 2 17 18 19 20 21
i n f o
Article history: Received 8 March 2014 Received in revised form 2 May 2014 Accepted 11 June 2014 Available online xxxx
22 23 24 25 26 27 28
Keywords: PLGA Controlled release Exenatide SPG membrane emulsification Microsphere
a b s t r a c t Poly(DL-lactic-co-glycolic acid) (PLGA) microspheres have been widely prepared by many methods, including solvent evaporation, solvent extraction and the co-solvent method. However, very few studies have compared the properties of microspheres fabricated by these methods. This is partly because the broad size distribution of the resultant particles severely complicates the analysis and affects the reliability of the comparison. To this end, uniform-sized PLGA microspheres have been prepared by Shirasu porous glass premix membrane emulsification and used to encapsulate exenatide, a drug for treating Type 2 diabetes. Based on this technique, the influences on the properties of microspheres fabricated by the aforementioned three methods were intensively investigated, including in vitro release, degradation and pharmacology. We found that these microspheres presented totally different release behaviors in vitro and in vivo, but exhibited a similar trend of PLGA degradation. Moreover, the internal structural evolution visually demonstrated these release behaviors. We selected for further examination the microsphere prepared by solvent evaporation because of its constant release rate, and explored its pharmacodynamics, histology, etc., in more detail. This microsphere when injected once showed equivalent efficacy to that of twice-daily injections of exenatide with no inflammatory response. Ó 2014 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46
47 48
1. Introduction
49
In the last few decades, microspheres based on biodegradable polymers such as poly(DL-lactic-co-glycolic acid) (PLGA) for sustained delivery of protein/peptide have been widely studied [1–3]. Some peptides, such as Lupron DepotÒ (leuprolide acetate) and Sandostatin LARÒ (octreotide acetate), have been incorporated into PLGA microspheres and commercialized as sustained-release systems [4]. Solvent evaporation, solvent extraction and co-solvent methods are the most popular approaches used to prepare the microspheres, because these processes are simple and convenient to control [5]. In the first two methods, an aqueous solution dissolving the drug is emulsified with an organic solution (oil phase)
50 51 52 53 54 55 56 57 58 59
⇑ Corresponding author at: State Key Laboratory of Biochemical Engineering, PLA
Q2
Key Laboratory of Biopharmaceutical Production & Formulation Engineering, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, People’s Republic of China. Tel./fax: +86 10 82627072. E-mail address: [email protected] (G. Ma). 1 These authors contributed equally to this work.
containing polymer. Methylene chloride (MC) is frequently used as the organic solvent in the evaporation method [6], and ethyl acetate (EA) is often used in the extraction method. For the co-solvent method, the drug is dissolved directly in two miscible organic solvents—typically MC and alcohol (methanol and ethanol)—without aqueous solution [7]. The use of different organic solvents results in microspheres with various characteristics. For instance, solvent removal rate affects the solidification, which is critical in determining the morphology, surface area and other aspects of microspheres [8]. Thus, discrepancies of the resultant particles in terms of, for example, release behavior and pharmacology will be generated. Studying this discrepancy can provide more important and general insights into the mechanisms on microsphere degradation and drug release in vitro/vivo, which is a key issue in developing long-term release systems. Although much effort has been devoted to microsphere preparations, relevant information about this discrepancy is scarce. This is partly because the broad size distribution resulting from conventional mechanical stirring will result in particles with poor reproducibility with respect to release
http://dx.doi.org/10.1016/j.actbio.2014.06.018 1742-7061/Ó 2014 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Qi F et al. Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2014.06.018
60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78
ACTBIO 3275
No. of Pages 10, Model 5G
3 July 2014 2 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100
F. Qi et al. / Acta Biomaterialia xxx (2014) xxx–xxx
behavior, drug efficacy, etc. [9–12]. Therefore, to control particle size, narrow down the size distribution and realize mass production, Shirasu porous glass (SPG) premix membrane emulsification has been employed [13]. Exenatide (synthetic exendin-4), a therapy for Type 2 diabetes mellitus (T2DM), was used as a model peptide in this study. It possesses 39 amino acids (H-His-Gly-Glu-Gly-Thr-Phe-Thr-SerAsp-Leu-Ser-Lys-Gln-Met-Glu-Glu-Glu-Ala-Val-Arg-Leu-Phe-Ile-GluTrp-Leu-Lys-Asn-Gly-Gly-Pro-Ser-Ser-Gly-Ala-Pro-Pro-Pro-Ser-NH2), shares about 53% homology with mammalian gut hormone (GLP1), and also possesses glucoregulatory actions including enhancement of insulin secretion, reduction of food intake, deceleration of gastric emptying and improvement of b-cell function [14,15]. Herein, based on SPG premix membrane emulsification, monodisperse exenatide-loaded PLGA microspheres were prepared by solvent evaporation, solvent extraction and co-solvent methods, respectively. Systematic research into the effect of preparation methods on the properties of microspheres was performed, including release behaviors, molecular weight (Mw) degradation, structural evolution, etc. Then, pharmacological aspects of the optical formulation were investigated, such as pharmacodynamics, immunohistochemistry analysis and inflammatory response.
101
2. Materials and methods
102
2.1. Materials
103
114
PLGA with a molar ratio of D,Llactide/glycolide 75/25 (Mw 13 kDa) was purchased from Lakeshore Biomaterials (Birmingham, AL, USA). Exenatide was provided by Hybio Pharmaceutical Co. Ltd. (Shenzhen, PR China). Poly(vinyl alcohol) (PVA-217, degree of polymerization 1700, degree of hydrolysis 88.5%) was provided by Kuraray (Tokyo, Japan). SPG membranes were purchased from SPG Technology Co. Ltd. (Miyazaki, Japan). The SPG premix membrane emulsification equipment (FMEM-500M) was designed by National Engineering Research Center for Biotechnology (NERCB, Beijing, PR China). Acetonitrile and trifluoroacetic acid (TFA) (both HPLC grade) were purchased from Dikma Co. Ltd. (Lake Forest, USA). All other reagents were analytical grade.
115
2.2. Preparation of microspheres
116
Microspheres prepared by solvent evaporation, solvent extraction and the co-solvent method are abbreviated as EVM, EXM and COM, respectively. Before preparation, SPG membrane 50.2 lm in size was installed in the equipment (Fig. 1a). For preparation of EVM
104 105 106 107 108 109 110 111 112 113
(Fig. 1b), 1 ml exenatide aqueous solution (3%, w/v, W1) was emulsified with 8 ml organic solvent (MC, O) containing PLGA (10%, w/ v) by homogenization (T18, IKA, Germany) at 18,000 rpm for 60 s to form W1/O. Next, the W1/O was stirred at 250 rpm for 1 min with external aqueous phase (W2) containing PVA (2%, w/v) and NaCl (0.5%, w/v) to form coarse W1/O/W2 emulsions. These were then poured into a premix reservoir and extruded through the SPG membrane by N2 pressure at 5 kPa to achieve uniform-sized droplets. After that, they were solidified at room temperature at 250 rpm for 5 h. Finally, the microspheres were collected and washed with distilled water five times by centrifugation for 3 min at 300 g. The washed microspheres were stored in 70 °C overnight, then lyophilized, and obtained after 48 h. The conditions for lyophilization were as follows: ice condenser 80 °C; vacuum 31 °C, 0.34 mbar. For preparation of EXM (Fig. 1b), EA was used as oil phase (O). The preparation process was the same as above. In addition, the uniform-sized droplets achieved by extrusion through SPG membrane were poured quickly into solidification solution with a large volume (1.6 l containing 0.9% (w/v) NaCl) under magnetic stirring at 250 rpm for 4 h to solidify the microspheres. The microspheres were obtained in the same way as above For preparation of COM (Fig. 1b), the exenatide powder was dissolved in mixed organic solvent (volume ratio of MC and methanol 6:2) containing PLGA (10%, w/v). The other steps were same as those for EVM.
121
2.3. Characterization of microspheres
147
2.3.1. Surface morphology observation and size distribution measurement The shape and surface morphology of PLGA microspheres were observed by scanning electron microscopy (SEM) with a JSM-6700F (JEOL, Japan). The particle size and size distribution was measured with a Mastersizer 2000 (Malvern, UK). The size distribution was referred as the Span value and was calculated as follows:
148
122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146
149 150 151 152 153 154 155
156
117 118 119 120
Span ¼
Dv;90% Dv;10% ; Dv;50%
where Dv,90%, Dv,50% and Dv,10% are volume size diameters at 90%, 50% and 10% of the cumulative volume, respectively. The smaller the Span value, the narrower the size distribution.
Fig. 1. (a) Schematic depiction of the SPG premix membrane emulsification equipment; (b) schemes of the EVM, EXM and COM preparation.
Please cite this article in press as: Qi F et al. Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2014.06.018
158
159 160 161
ACTBIO 3275
No. of Pages 10, Model 5G
3 July 2014 F. Qi et al. / Acta Biomaterialia xxx (2014) xxx–xxx 162 163 164 165 166 167 168 169 170 171 172 173 174
2.3.2. Drug-loading capacity and encapsulation efficiency measurement Microspheres (5 mg) were dissolved in a mixed solution composed of 150 ll acetonitrile and 850 ll 0.01 M HCl. The concentration of exenatide was determined by a reverse-phase HPLC (RP-HPLC) system at room temperature with a C18 (250 mm 4.6 mm 5 lm, Syncronis, Thermo) chromatographic column. The chromatography was performed with a linear gradient elution from 20% to 60% acetonitrile in ultrapure water containing 0.1% TFA for 16 min. The flow rate was 1 ml min1, and the UV absorbance was set at 214 nm. The loading capacity and encapsulation efficiency of the microspheres were calculated by the following equations:
175 177
Loadingcapacityðlg=mgÞ ¼
Mass of drug in microspheres Mass of microspheres
3
2.4.2. Pharmacodynamics study Male kk-ay mice (9–10 weeks old), T2DM model mice, were assigned into three groups (n = 6 per group): exenatide solution group (twice daily injection for 14 days, 50 lg kg1), EVM (single subcutaneous injection of EVM, 40 mg kg1 containing 1.44 mg kg1 exenatide, equivalent to the dose of twice-daily injection of exenatide solution for 2 weeks) and saline group (single subcutaneous injection, control). The blood samples were obtained after fasting 4 h. A one-touch blood glucose meter (ACCU-CHEKÒ Performa) was used to measure the blood glucose. The area under the concentration–time curve (AUC) was estimated by the linear trapezoidal method. The in vivo results were analyzed as mean ± SE. Statistics were performed using GraphPad Prism 5 (San Diego, CA). Comparisons between groups were analyzed by two-way ANOVA analysis followed by Tukey’s test, and P < 0.05 was considered statistically significant.
220
2.4.3. Histological and immunohistochemistry studies Histological study was used to evaluate the safety of EVM in vivo. Subcutaneous tissues were removed from the injection sites. The tissue samples were fixed in 10% formalin, and sections were immersed in paraffin and cut using a microtome. The tissues were stained with hematoxylin and eosin. Insulin immunohistochemistry study was performed to confirm the insulinotropic bioactivity of the drug. Tissue samples were washed with PBS, dehydrated by graded alcohols and xylene, and embedded in paraffin wax. The embedded tissues were sectioned at 4 lm, and attached to polylysine-coated slides. Endogenous peroxidase blocking process was performed after 20 min pretreatment at room temperature. Sections were reacted with anti-insulin antibody diluted at a ratio of 1/800 as a primary antibody and incubated at 4 °C overnight. After washing the sections with PBS three times, secondary antibody was added and incubated for 4 h at room temperature before washing with PBS. Finally, sections were stained with DAB, counterstained in real hematoxylin and 0.3% ammonia water. Proliferating cell nuclear antigen (PCNA) immunohistochemistry was performed by the same steps. Photomicrographs of the histological and immunohistochemical slices were taken by an Olympus microscope (Tokyo, Japan). The images were analyzed by Image J version 1.48 (NIH Image, Maryland, MD).
236
3. Results and discussion
260
221 222 223 224 225 226 227 228 229 230 231 232 233 234 235
178
Encapsulationefficiencyð%; w=wÞ ¼ 180
181 182 183
184 185 186 187 188 189 190 191 192
Loading capacity 100% Theoretical loading capacity
2.3.3. Circular dichroism (CD) measurement A detailed description of method used for CD measurement is included in the Supporting Information. 2.3.4. In vitro drug release and polymer degradation studies PLGA microspheres (10 mg) were dispersed in 1 ml 10 mM phosphate buffer saline (PBS) medium (pH 7.4) and incubated under agitation at 37 °C. At each time interval, supernatants were collected by centrifugation for 3 min at 300 g and replaced with fresh buffer of equal volume. The concentration of exenatide in the supernatant was determined by Micro BCA Protein Assay Kit. The method for studying polymer degradation is shown in the Supporting Information.
237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253
200
2.3.5. Observation of drug distribution inside microspheres Super Fluor 488 SE (Fanbo Biochemicals, Beijing, PR China) was used to label exenatide to observe drug distribution within microspheres during incubation. The microsphere preparation and in vitro release experiment were the same as above. At predetermined times, a small amount of microspheres were removed from the samples and observed by confocal laser scanning microscopy (CLSM, TCS SP 5, Leica) at an excitation wavelength of 488 nm.
201
2.4. Animal experiments and statistics
3.1. Characteristics of microspheres
261
202
The animal experiments were consistent with the guidelines set by the National Institutes of Health (NIH Publication No. 85-23, revised 1985) and were approved by the Experimental Animal Ethics Committee, Beijing. The animal experiments were performed at Beijing Dingguo Changsheng Biotechnology Co. Ltd.
Uniform-sized exenatide-loaded PLGA microspheres were successfully prepared by SPG premix membrane emulsification combined with solvent evaporation, solvent extraction and co-solvent methods. As shown in Fig. 2a–c, all the microspheres had smooth external surfaces with a few pores, attributed to the drug diffusion from particle core to the surface during solidification [16]. The sizes of the three resultant microspheres were 20.63, 22.38 and 21.50 lm with small ranges (Table S1 in the Supporting Information), suggesting a narrow size distribution (Fig. 2d). EVM and EXM were prepared as double emulsions (W/O/W) with different organic solvents, but the encapsulation efficiency of the former (92.51 ± 3.41%, MC as oil phase) was higher than the later (84.42 ± 3.67%, EA as oil phase), as was the loading capacity (34.23 ± 1.31 vs 31.64 ± 1.43 lg mg1). The reason was that due to volumes of solidification solution added in emulsion droplets of EXM, the hydrophilic drug diffused into the external water phase more easily during solidification. The highest encapsulation efficiency and loading capacity (98.48 ± 4.94% and 35.21 ± 1.82 lg mg1) was achieved by co-solvent method (O/W), because
262
193 194 195 196 197 198 199
203 204 205 206
207 208 209 210 211 212 213 214 215 216 217 218 219
2.4.1. Pharmacokinetics study Male SD rats (250–300 g) were divided into four groups (n = 6 per group): exenatide solution, EVM, EXM and COM groups. The rats in the exenatide solution group were subcutaneously injected with exenatide at a dose of 36 lg per rat. The other groups were subcutaneously injected with microspheres containing exenatide at a dose of 1 mg per rat (equivalent to the dose of a twice-daily injection of exenatide solution for 2 weeks). Plasma was obtained at the same sampling time as in vitro release. The establishment of in vitro/in vivo correlation is included in the Supporting Information. The exenatide concentration in plasma was analyzed with an Exendin-4 EIA kit (Phoenix Pharmaceuticals, CA, USA) according to the manufacturer’s instructions.
Please cite this article in press as: Qi F et al. Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2014.06.018
254 255 256 257 258 259
263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280
ACTBIO 3275
No. of Pages 10, Model 5G
3 July 2014 4
F. Qi et al. / Acta Biomaterialia xxx (2014) xxx–xxx
Fig. 2. SEM images of (a) EVM, (b) EXM and (c) COM; (d) size distribution. Scale bar: 10 lm.
283
methanol, miscible with MC, could dissolve the drug, making it distribute uniformly in the oil phase, and leading to a fast polymer precipitation favorable for high encapsulation efficiency.
284
3.2. Drug stability
285
Using organic solvent may lead to denaturation of encapsulated peptide/protein during preparation, owing to formation of an oil– water interface. For example, MC had a denaturing effect on the stability of protein C, and the effect would be worse if EA were used [17]. EVM and EXM were prepared using MC and EA as the organic solvent, respectively. In COM preparation, MC and methanol were both used. To study the stability of exenatide during preparation, CD spectroscopy was used to monitor its secondary structure changes. The peptide was extracted from the microspheres after lyophilization and compared with the native one. As shown in Fig. S1 (Supporting Information), all the spectra show identical curves with two negative minima around 208 and 222 nm, indicating the predominant a-helix and b-structure. Thus, the stability of exenatide was not influenced by the preparation method used.
281 282
286 287 288 289 290 291 292 293 294 295 296 297 298
299
3.3. In vitro release
300
It is well known that drug-release behavior is significantly affected by preparation method [18,19]. Similarly, in our study, all the release profiles were completely different (Fig. 3). EVM exhibited a low burst initial release within 24 h (25%), followed by constant release until the end. EXM showed a plateau with the highest burst (45%), and after 5 weeks the release rate increased. On the other hand, COM exhibited a similar release behavior as EXM but with the slowest burst (13%). To further investigate the mechanisms, studies of the morphological evolution and polymer degradation of microspheres were undertaken as follows.
301 302 303 304 305 306 307 308 309 310
Fig. 3. In vitro release profiles (n = 3, mean ± SD).
3.4. Structural evolution and polymer degradation
311
PLGA microspheres release drug via gradual erosion of their matrix after hydration and cleavage of hydrolytically sensitive bonds in the polymer chains [20]. Therefore, the morphological and inner structural evolution can reflect the microsphere degradation states, and the molecular weight decrease can quantitatively characterize the polymer degradation.
312
3.4.1. Morphological evolution As illustrated in Fig. S2 (Supporting Information), irrespective of the preparation method used, we observed the general trend that the external surfaces of the microspheres became coarse with large pores, and more pores were formed by day 14. Moreover, the pores disappeared gradually, but all of them still maintained their integrity up to 60 days. The reason was that the reduced glass transition temperature (Tg) caused by decreased Mw, promoted the mobility of degraded oligomers and accelerated their rearrangement by plasticization [21]. For the same reason, the microspheres did not collapse even up to 100 days. However, they all became smaller,
318
Please cite this article in press as: Qi F et al. Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2014.06.018
313 314 315 316 317
319 320 321 322 323 324 325 326 327 328
ACTBIO 3275
No. of Pages 10, Model 5G
3 July 2014 F. Qi et al. / Acta Biomaterialia xxx (2014) xxx–xxx 329 330 331 332 333 334 335 336 337 338 339 340
341 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393
and some were aggregated and had coarse surfaces. The maintenance of microsphere integrity might be one of the reasons for the incomplete release. In conclusion, no significant differences of the external surfaces were observed. 3.4.2. Polymer degradation Polymer degradation is the result of hydrolysis of ester bonds through the whole polymer matrix [22]. As shown in Fig. S3 (Supporting Information), surprisingly, all of the polymers followed the similar Mw degradation profile. The polymer Mw decreased almost linearly with time, suggesting pseudo-first-order kinetics for the polymer degradation during incubation. The Mw decrease was calculated as follows:
Mw ðtÞ ¼ Mw0 expðkd tÞ
ð1Þ
where Mw0 is the initial Mw of polymer and kd is the pseudo-firstorder degradation rate constant of the polymer [23]. The values of kd calculated by Eq. (1) from the experimental results were almost identical (Fig. S3). This indicated that the degradations proceeded via the same mechanism, and were not strongly affected by the preparation method. 3.4.3. Inner structural evolution The inner structural changes that could affect in vitro release behavior [8,24] were reflected in the drug distribution variations visualized by CLSM. In order to explore the mechanism of internal microsphere degradation, labeled exenatide was encapsulated in PLGA microspheres. As illustrated in Fig. 4, EVM exhibited non-uniform drug distribution with many large non-homogeneous aggregations, as the peptide was dissolved in the inner aqueous droplets within the matrix of particle. EXM also presented the same distribution, but it was obvious that there were more drugs on or near the surfaces. This was because that when the uniform-sized emulsion droplets were added to the solidification solution, they were exposed to volumes of water, resulting in the peptide diffusing near or on the particle surface. This also explained the reason for its highest initial burst release. As for COM, exenatide was uniformly distributed inside these microsphere, since the drug was dissolved in the oil phase. During incubation, the microspheres degraded gradually. Simultaneously, the drug domains changed into aqueous pores showing no green color [11,25]. As incubation proceeded, more pores were generated inside EVM, and contacted the external phase. Thus, water and degraded oligomers diffused into and out of the matrix easily, leading to a constant release rate. For EXM, the drug near the surface quickly diffused into medium, resulting in a high burst initial release. In addition, the internal structural changes were almost the same as those in EVM. In COM, however, at an early phase (day 14), the drug distribution was still relatively uniform except for generation of small pores near the surfaces, indicating the drug nearby was released first. At day 30, some drug remained in the core of microspheres; hence, the release rate was very slow in this phase. Afterwards, the drug in the core began to diffuse out of the microspheres showing a green margin, resulting in a high release rate. Compared with the corresponding morphological evolution, the internal structure clearly evolved, because the pH inside the particle decreased due to local polymer degradation. Whereas pH at the surface remained unchanged, the morphology could only be changed by Tg caused by degradation of the whole polymer [26]. The internal structural evolution reflected by drug distribution clearly demonstrated and explained the differences in release behaviors. Above all, it was found that although the drug release was controlled by polymer degradation, the internal structural changes of microspheres rather than the decrease of polymer Mw played the most important role.
5
3.5. Pharmacokinetics
394
Pharmacokinetics using SD rats was evaluated to further explore the differences between the microsphere preparations in vivo (Fig. 5, Table S2). EVM achieved maximal plasma drug concentration (136.16 ± 19.11 ng ml1) within 2 h after injection, which was significantly lower than the exenatide solution group (239.83 ± 15.48 ng ml1). Furthermore, it showed constant release within 14 days, and then leveled off. The EXM group presented a quick release with high concentration in 1 h, and achieved the highest value (190.09 ± 16.04 ng ml1) within 2 h. Moreover, its plasma drug level was lower than that of EVM almost throughout the entire investigations period except the initial hours, in accordance with the in vitro release behavior. On the other hand, COM showed its highest value (60.12 ± 12.62 ng ml1) at 1 h. Despite this, its maximal value was the lowest among the three. In addition, its drug concentration presented the lowest level within the first 14 days. Afterwards, however, it became very high until the end. The AUC01d value of the exenatide solution group (25.01 ± 1.66 ng day ml1) was significantly higher than the group Q3 COM (9.46 ± 0.59 ng day ml1), showed no difference compared with EVM (20.14 ± 2.29 ng day ml1), and was significantly lower than EXM (33.65 ± 3.12 ng day ml1). The AUC030d values of EVM (188.89 ± 19.45 ng day ml1) and COM (255.01 ± 72.60 ng day ml1) were significant higher than EXM (149.74 ± 13.85 ng day ml1). According to the above results, EVM was the best formulation for a long-term effective system, because it possessed both the high AUC030d value and constant release rate both in vitro or in vivo. In the following experiments, EVM was employed to further investigate the relevant pharmacology. The in vitro/in vivo correlations of EVM, EXM and COM were established by the profiles of cumulative release in vitro and in vivo (Fig. 5b), exhibiting a good linear regression correlation. Based on these results, it was feasible to predict the release in vivo by the established correlation according to the in vitro release.
395
3.6. The pharmacological study of EVM
430
3.6.1. Hypoglycemic efficacy Kk-ay mice were used as a T2DM model to evaluate the pharmacodynamics. After injection, the blood glucose level in exenatide solution and in the EVM group were significantly lower than that in control (saline group), and remained lower for first 14 days (Fig. 6a). However, when the mice in the solution group stopped being injected from day 15, the blood glucose returned to the control level gradually. In contrast, the EVM group still presented significantly low blood glucose throughout the entire experimental period. The trend of hypoglycemic efficacy was well explained by the pharmacokinetics of EVM, i.e. the drug concentration in blood was high for the first 14 days, leading to a significantly low blood glucose level. Afterwards, the release leveled off, and the efficacy gradually became weak. The AUCblood glucose 014d values of exenatide solution (186.73 ± 40.35 mmol l1 day1) and EVM (154.47 ± 75.90 mmol l1 day1) group were significantly smaller than that of control (296.84 ± 41.40 mmol l1 day1), indicating the former two groups both presented hypoglycemic efficacy. Furthermore, there was no significant difference between them, suggesting they offered equivalent efficacy.
431
3.6.2. Insulin immunohistochemistry analysis After 14 days of treatment, the mice pancreatic tissues were removed to analyze the insulinotropic bioactivity of exenatide by insulin immunohistochemistry, which reflected the insulin secretion in pancreas islets visually [27,28].
451
Please cite this article in press as: Qi F et al. Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2014.06.018
396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429
432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450
452 453 454 455
ACTBIO 3275
No. of Pages 10, Model 5G
3 July 2014 6
F. Qi et al. / Acta Biomaterialia xxx (2014) xxx–xxx
EVM
EXM
COM
Day 0
Day 14
Day 30
Day 60
Fig. 4. CLSM images for inner structural (drug distribution) evolutions of EVM, EXM and COM at different incubation times (day 0, 14, 30 and 60). Scale bar: 10 lm.
Fig. 5. (a) Plasma exenatide concentration vs. time profiles of EVM, EXM and COM (n = 6); the insert presents the plasma exenatide concentrations vs. time within 24 h. (b) In vitro/in vivo correlations of EVM, EXM and COM.
456 457 458 459 460 461 462
As shown in Fig. 7a–c, in the cytoplasm of the endocrine islet, positive signals (dark yellow) with larger area were obvious in the exenatide solution and EVM groups, indicative of higher insulin secretion and more b-cells. Semiquantitative analysis of the ratios (b-cell area to islet area) (Fig. 7d) also demonstrated that insulin secretions in these two groups were significantly higher than in control, in line with the result of hypoglycemic efficacy. This result
also suggested that EVM achieved the same insulinotropic activity as twice-daily injection of exenatide solution for 2 weeks.
463
3.6.3. PCNA immunohistochemistry analysis In addition to hypoglycemic efficacy and insulinotropic activity, exenatide is also capable of increasing b-cell mass through stimulation of b-cell neogenesis, proliferation and suppression of b-cell
465
Please cite this article in press as: Qi F et al. Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2014.06.018
464
466 467 468
ACTBIO 3275
No. of Pages 10, Model 5G
3 July 2014 F. Qi et al. / Acta Biomaterialia xxx (2014) xxx–xxx
7
Fig. 6. (a) Blood glucose vs. time profiles of exenatide solution, EVM and saline group. (b) AUC values of blood glucose concentration in three groups (0–14 days) (AUCblood Statistical differences were tested using two-way ANOVA followed by Tukey’s test. Differences were termed statistically significant at ⁄P < 0.05, ⁄⁄P < 0.01 vs. saline group. Data are mean ± SE. n = 6.
glucose 014d).
Fig. 7. Insulin immunohistochemistry of pancreas tissues from kk-ay mice in (a) exenatide solution, (b) EVM and (c) saline group after 14 days of treatment, the dark yellow areas indicate the insulin-positive signals. (d) Percentages of insulin-positive b-cell area to the islet area (n = 4). Scale bar: 200 lm. Statistical differences were tested using two-way ANOVA followed by Tukey’s test. Differences were termed statistically significant at ⁄P < 0.05 vs. saline group. Data are mean ± SE.
469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488
apoptosis, which is the biggest difference compared with chemical drugs such as metformin and sulfonylurea treatment for T2DM [29]. Because PCNA is distinguished by its apparent association with cell division [30], its immunohistochemistry can be used to indicate cell neogenesis including b-cells in pancreas islets. As illustrated in Fig. 8a,b, proliferative cells (dark brown points) were distributed sporadically within pancreas islets (in frame) in the exenatide solution and EVM groups. However, the control group shows few dark brown points (Fig. 8c). Semiquantitative analysis of the ratios (PCNA-positive cells to all cells) (Fig. 8d) indicated that the numbers of proliferative cells in the exenatide solution and EVM groups were significant more than that in the saline group, and that some of them were b-cells. Furthermore, the area of islet was smaller than with the other two groups. Although PCNA immunohistochemistry indicates no direct b-cell neogenesis, it was suggested that the pre-existing b-cell in the islets might proliferate slowly [31], and there were more in the former two groups than in the control. As we discussed above, in Fig. 7, the positive areas indicated not only the insulin secretion, but also the number of b-cells. Combined with the results of insulin and PCNA immuno-
histochemistry, it was concluded that the exenatide and EVM group both improved pancreatic function.
489
3.7. Histological study
491
After administration to SD rats, the injection sites were removed periodically to perform the histological study. As shown in Fig. 9, compared with control, no inflammatory response was observed in the tissue-injected exenatide solution. Similarly, the EVM group also presented no significant inflammatory reaction in any phase (day 7–30). This is an exciting result compared with other works that also used PLGA as material to encapsulate drugs, because accumulation of acidic oligomers degraded by PLGA might lead to an inflammatory response [32,33]. One of the possible reasons was that the high LA/GA (75/25) ratio caused slow degradation PLGA, generating lower levels of acidic oligomers. Meanwhile, the oligomers would be quickly transported to further adjust the local acidity. Another reason was that a only a small number of microspheres needed to be injected to achieve an efficacious dose. Thus, the inflammatory response was very weak or
492
Please cite this article in press as: Qi F et al. Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2014.06.018
490
493 494 495 496 497 498 499 500 501 502 503 504 505 506
ACTBIO 3275
No. of Pages 10, Model 5G
3 July 2014 8
F. Qi et al. / Acta Biomaterialia xxx (2014) xxx–xxx
(b)
(a)
(d)
(c)
Fig. 8. PCNA immunohistochemistry of pancreas tissues from kk-ay mice in (a) exenatide solution, (b) EVM and (c) saline group after 14 days of treatment; (d) percentages of PCNA-positive cells to all cells in islet (n = 4). The dark brown points which the arrow points to in the frame are PCNA-positive signals. Statistical differences were tested using two-way ANOVA followed by Tukey’s test. Differences were termed statistically significant at ⁄P < 0.05 vs. saline group. Scale bar: 100 lm.
Control
Exenatide solution
Exenatide 7d
14 d
30 d
Fig. 9. Histology of injection sites after s.c. administration with exenatide solution and EVM. Top panel shows normal tissue (control) and tissue injected with exenatide solution. Bottom panel shows tissues injected with EVM after 7, 14 and 30 days. Scale bar: 200 lm.
507 508
even negligible. In conclusion, the EVM was acceptable and safe for administration.
509
3.8. Morphological evolution of microspheres in vivo
510
It is well known that after injection, large PLGA microspheres will remain in place under the skin and degrade slowly affected by biological action [35]. During this degradation, exenatide is released continuously into circulation as a result of polymer erosion [36–38], a process that can be mirrored by the biodegradation state of the microspheres. Thus, the microspheres injected were removed periodically to observe any morphological changes by SEM. After injection of EVM, a nodule was formed under the skin at the injection site. At day 7, the EVM remained as individuals, although they were no longer spherical (Fig. 10). After 14 and
511 512 513 514 515 516 517 518 519 520
22 days, the microspheres agglomerated and interspersed among surrounding connective tissues. At the end of 1 month, from the exterior, no subcutaneous nodule could be detected in any rat (figure not shown). Internal observation, however, showed that the microspheres kept interspersing, invaded deeper into tissue, and could not be distinguished clearly. This was because in the physiological environment, the area of microspheres injected into the body decreased owing to degradation, and the surrounding tissues integrated with the microspheres and reorganized [34]. As the injection time increased, the phenomenon was more obvious. At the later stage, the microspheres became submerged by the connective tissue and degraded further. This study showed the biodegradation states of EVM with a size of 20 lm in vivo. According to our observations, it might not be a significant burden for the body in the late phase of injection.
Please cite this article in press as: Qi F et al. Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2014.06.018
521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536
ACTBIO 3275
No. of Pages 10, Model 5G
3 July 2014 F. Qi et al. / Acta Biomaterialia xxx (2014) xxx–xxx
(a)
(b)
(c)
(d)
9
Fig. 10. Morphological changes of EVM in vivo: (a) day 7, (b) day 14, (c) day 22 and (d) day 30. Scale bar: 20 lm.
537
4. Conclusion
Appendix B. Supplementary data
566
538
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.actbio.2014. 06.018.
567
References
570
554
Exenatide-loaded PLGA microspheres were prepared by SPG premix membrane emulsification combined with solvent evaporation, solvent extraction and co-solvent methods. All of the microspheres had a narrow size distribution and an average size of 20 lm. EVM presented a constant release, and EXM showed the highest burst release. As for COM, it exhibited a slow release, followed by a fast release in the later phases. The preparation methods did not influence the polymer Mw degradation, but affected the internal structural evolution that played an important role in the in vitro release behaviors. These relevant studies provide a new guidance for microsphere preparation in pharmaceutics. EVM was the best formulation due to its constant release rate in vivo over 14 days. The pharmacodynamics showed that the EVM could effectively reduce the blood glucose with no inflammatory response, and also possessed the same function in the pancreas as exenatide solution. Above all, it is feasible for a single injection of EVM to replace a twice-daily injection of exenatide for 2 weeks.
555
Acknowledgements
[1] Aubert-Pouessel A, Venier-Julienne MC, Clavreul A, Sergent M, Jollivet C, Montero-Menei CN, et al. In vitro study of GDNF release from biodegradable PLGA microspheres. J Control Release 2004;95:463–75. [2] Jiang WL, Gupta RK, Deshpande MC, Schwendeman SP. Biodegradable poly(lactic-co-glycolic acid) microparticles for injectable delivery of vaccine antigens. Adv Drug Deliv Rev 2005;57:391–410. [3] Eswaramoorthy R, Chang CC, Wu SC, Wang GJ, Chang JK, Ho ML. Sustained release of PTH(1–34) from PLGA microspheres suppresses osteoarthritis progression in rats. Acta Biomater 2012;8:2254–62. [4] Ahn JH, Park EJ, Lee HS, Lee KC, Na DH. Reversible blocking of amino groups of octreotide for the inhibition of formation of acylated peptide impurities in poly(lactide-co-glycolide) delivery systems. AAPS PharmSciTech 2011;12: 1220–6. [5] Ye ML, Kim S, Park K. Issues in long-term protein delivery using biodegradable microparticles. J Control Release 2010;146:241–60. [6] Elkharraz K, Ahmed AR, Dashevsky A, Bodmeier R. Encapsulation of watersoluble drugs by an o/o/o-solvent extraction microencapsulation method. Int J Pharm 2011;409:89–95. [7] Lassalle V, Ferreira ML. PLA nano- and microparticles for drug delivery: an overview of the methods of preparation. Macromol Biosci 2007;7:767–83. [8] Yang YY, Chung TS, Bai XL, Chan WK. Effect of preparation conditions on morphology and release profiles of biodegradable polymeric microspheres containing protein fabricated by double-emulsion method. Chem Eng Sci 2000;55:2223–36. [9] Mao SR, Xu J, Cai CF, Germershaus O, Schaper A, Kissel T. Effect of WOW process parameters on morphology and burst release of FITC-dextran loaded PLGA microspheres. Int J Pharm 2007;334:137–48. [10] Sah HK, Toddywala R, Chien YW. Biodegradable microcapsules prepared by a W/O/W technique—effects of shear force to make a primary W/O Emulsion on their morphology and protein release. J Microencapsul 1995;12:59–69. [11] Yan CH, Resau JH, Hewetson J, West M, Rill WL, Kende M. Characterization and morphological analysis of protein-loaded poly(lactide-co-glycolide) microparticles prepared by water-in oil-in-water emulsion technique. J Control Release 1994;32:231–41. [12] Wang LY, Gu YH, Zhou QZ, Ma GH, Wan YH, Su ZG. Preparation and characterization of uniform-sized chitosan microspheres containing insulin
571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606
539 540 541 542 543 544 545 546 547 548 549 550 551 552 553
556 557 558 559 560
We thank the NSFC/RGC Joint Research Scheme sponsored by the Research Grants Council of Hong Kong and the National Natural Q4 Science Foundation of China (No. 21161160555) and National NatQ5 ural Science Foundation of China (No. 21336010) for the financial support provided.
561
Appendix A. Figures with essential color discrimination
562
Certain figures in this article, particularly Figs. 1–9 are difficult to interpret in black and white. The full color images can be found in the on-line version, at http://dx.doi.org/10.1016/ j.actbio.2014.06.018.
563 564 565
Please cite this article in press as: Qi F et al. Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2014.06.018
568 569
ACTBIO 3275
No. of Pages 10, Model 5G
3 July 2014 10 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646
[13]
[14]
[15]
[16]
[17]
[18] [19]
[20]
[21]
[22] [23]
[24] [25]
F. Qi et al. / Acta Biomaterialia xxx (2014) xxx–xxx by membrane emulsification and a two-step solidification process. Colloids Surf B 2006;50:126–35. Qi F, Wu J, Fan QZ, He F, Tian GF, Yang TY, et al. Preparation of uniform-sized exenatide-loaded PLGA microspheres as long-effective release system with high encapsulation efficiency and bio-stability. Colloids Surf B 2013;112:492–8. Drucker DJ, Buse JB, Taylor K, Kendall DM, Trautmann M, Zhuang DL, et al. Exenatide once weekly versus twice daily for the treatment of type 2 diabetes: a randomised, open-label, non-inferiority study. Lancet 2008;372:1240–50. Zhou J, Cai Z-H, Li L, Kou C, Gao Y-F. Preparation and PEGylation of exendin-4 peptide secreted from yeast Pichia pastoris. Eur J Pharm Biopharm 2009;72:412–7. Ibrahim MA, Ismail A, Fetouh MI, Gopferich A. Stability of insulin during the erosion of poly(lactic acid) and poly(lactic-co-glycolic acid) microspheres. J Control Release 2005;106:241–52. Zambaux MF, Bonneaux F, Gref R, Dellacherie E, Vigneron C. Preparation and characterization of protein C-loaded PLA nanoparticles. J Control Release 1999;60:179–88. Sinha VR, Trehan A. Biodegradable microspheres for protein delivery. J Control Release 2003;90:261–80. Miyajima M, Koshika A, Okada J, Kusai A, Ikeda M. The effects of drug physicochemical properties on release from copoly (lactic/glycolic acid) matrix. Int J Pharm 1998;169:255–63. Samadi N, Abbadessa A, Di Stefano A, van Nostrum CF, Vermonden T, Rahimian S, et al. The effect of lauryl capping group on protein release and degradation of poly(D, L-lactic-co-glycolic acid) particles. J Control Release 2013;172:436–43. Wei GB, Pettway GJ, McCauley LK, Ma PX. The release profiles and bioactivity of parathyroid hormone from poly(lactic-co-glycolic acid) microspheres. Biomaterials 2004;25:345–52. Gopferich A. Mechanisms of polymer degradation and erosion. Biomaterials 1996;17:103–14. Siepmann J, Faisant N, Akiki J, Richard J, Benoit JP. Effect of the size of biodegradable microparticles on drug release: experiment and theory. J Control Release 2004;96:123–34. Freiberg S, Zhu X. Polymer microspheres for controlled drug release. Int J Pharm 2004;282:1–18. Yushu H, Venkatraman S. The effect of process variables on the morphology and release characteristics of protein-loaded PLGA particles. J Appl Polym Sci 2006;101:3053–61.
[26] Schliecker G, Schmidt C, Fuchs S, Kissel T. Characterization of a homologous series of D, L-lactic acid oligomers; a mechanistic study on the degradation kinetics in vitro. Biomaterials 2003;24:3835–44. [27] Hahn SK, Kong JH, Oh EJ, Chae SY, Lee KC. Long acting hyaluronate—exendin 4 conjugate for the treatment of type 2 diabetes. Biomaterials 2010;31:4121–8. [28] Chae SY, Jin CH, Shin JH, Son S, Kim TH, Lee S, et al. Biochemical, pharmaceutical and therapeutic properties of long-acting lithocholic acid derivatized exendin-4 analogs. J Control Release 2010;142:206–13. [29] Xu G, Kaneto H, Lopez-Avalos MD, Weir GC, Bonner-Weir S. GLP-1/exendin-4 facilitates beta-cell neogenesis in rat and human pancreatic ducts. Diabetes Res Clin Pract 2006;73:107–10. [30] Mathews MB, Bernstein RM, Franza BR, Garrels JI. Identity of the proliferating cell nuclear antigen and cyclin. Nature 1984;309:374–6. [31] Fujisawa H, Zhang Z, Sun W, Huang MS, Kobayashi J, Yasuda H, et al. Histopathological changes in the pancreas from a spontaneous hyperglycemic cynomolgus monkey. J Toxicol Pathol 2012;25:215–9. [32] Yang JA, Kim H, Park K, Hahn SK. Molecular design of hyaluronic acid hydrogel networks for long-term controlled delivery of human growth hormone. Soft Matter 2011;7:868–70. [33] Zolnik BS, Burgess DJ. Evaluation of in vivo-in vitro release of dexamethasone from PLGA microspheres. J Control Release 2008;127:137–45. [34] Kim BS, Oh JM, Hyun H, Kim KS, Lee SH, Kim YH, et al. Insulin-loaded microcapsules for in vivo delivery. Mol Pharm 2009;6:353–65. [35] Blanco MD, Sastre RL, Olmo R, Teijon C, Muniz E, Teijon JM. 5-Fluorouracil plasma levels and biodegradation of subcutaneously injected drug-loaded microspheres prepared by spray-drying poly(D, L-lactide) and poly(D, L-lactideco-glycolide) polymers. Int J Pharm 2007;338:180–90. [36] DeYoung MB, MacConell L, Sarin V, Trautmann M, Herbert P. Encapsulation of exenatide in poly-(D, L-Lactide-Co-Glycolide) microspheres produced an investigational long-acting once-weekly formulation for Type 2 diabetes. Diabetes Technol Ther 2011;13:1145–54. [37] Wysham C, Grimm M, Chen S. Once weekly exenatide: efficacy, tolerability and place in therapy. Diabetes Obes Metab 2013;15:871–81. [38] Chang LC, Yang CY, Chua ACN, Lin YJ, Lai SM. Sustained release of transgenic human factor IX: preparation, characterization, and in vivo efficacy. Mol Pharm 2011;8:1767–74.
Please cite this article in press as: Qi F et al. Mechanistic studies for monodisperse exenatide-loaded PLGA microspheres prepared by different methods based on SPG membrane emulsification. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2014.06.018
647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683