- Email: [email protected]
A novel strategy for the preparation of porous microspheres and its application in peptide drug loading
Accepted Manuscript A Novel Strategy for the Preparation of Porous Microspheres and Its Application in Peptide Drug Loading Yi Wei, Yuxia Wang, Huixia Zhang, Weiqing Zhou, Guanghui Ma PII: DOI: Reference:
S0021-9797(16)30330-7 http://dx.doi.org/10.1016/j.jcis.2016.05.045 YJCIS 21289
To appear in:
Journal of Colloid and Interface Science
Received Date: Revised Date: Accepted Date:
15 March 2016 23 May 2016 23 May 2016
Please cite this article as: Y. Wei, Y. Wang, H. Zhang, W. Zhou, G. Ma, A Novel Strategy for the Preparation of Porous Microspheres and Its Application in Peptide Drug Loading, Journal of Colloid and Interface Science (2016), doi: http://dx.doi.org/10.1016/j.jcis.2016.05.045
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
A Novel Strategy for the Preparation of SCT-loaded Porous mPEG-PLGA Microspheres with Narrow Size Distribution
Yi Wei1, Yuxia Wang1, Huixia Zhang2, Weiqing Zhou1, and Guanghui Ma1,* 1
National Key Laboratory of Biochemical Engineering, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, People’s Republic of China 2 Shenyang Agriculture University, Shenyang, Liaoning 110161, People’s Republic of China ∗ Corresponding author. Tel.: +86 10 82627072; fax: +86 10 82627072. E-mail addresses: [email protected] The first two authors, Yi Wei and Yuxia Wang, contributed equally.
1
A Novel Strategy for the Preparation of Porous Microspheres and Its Application in Peptide Drug Loading
Yi Wei1, Yuxia Wang1, Huixia Zhang2, Weiqing Zhou1, and Guanghui Ma1,* 1
National Key Laboratory of Biochemical Engineering, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, People’s Republic of China 2 Shenyang Agriculture University, Shenyang, Liaoning 110161, People’s Republic of China ∗ Corresponding author. Tel.: +86 10 82627072; fax: +86 10 82627072. E-mail addresses: [email protected] The first two authors, Yi Wei and Yuxia Wang, contributed equally.
Abstract: A new strategy is developed to prepare porous microspheres with narrow size distribution for peptides controlled release, involving a fabrication of porous microspheres without any porogens followed by a pore closing process. Amphiphilic polymers with different hydrophobic segments (poly(monomethoxypolyethylene glycol-co-D,L-lactide) (mPEG-PLA), poly(monomethoxypolyethylene glycol-coD,L-lactic-co-glycolic acid) (mPEG-PLGA)) are employed as microspheres matrix to prepare porous microspheres based on a double emulsion-premix membrane emulsification technique combined with a solvent evaporation method. Both microspheres possess narrow size distribution and porous surface, which are mainly caused by a) hydrophilic polyethylene glycol (PEG) segments absorbing water molecules followed by a water evaporation process and b) local explosion of microspheres due to fast evaporation of dichloromethane(MC). Importantly, mPEG-PLGA microspheres have a honeycomb like structure while mPEG-PLA microspheres have a solid structure internally, illustrating that the different
2
hydrophobic segments could modulate the affinity between solvent and matrix polymer and influence the phase separation rate of microspheres matrix. Long term release patterns are demonstrated with pore-closed microspheres, which are prepared from mPEG-PLGA microspheres loading salmon calcitonin (SCT). These results suggest that it is potential to construct porous microspheres for drug sustained release using permanent geometric templates as new porogens. Keywords: porous microspheres, narrow size distribution, mPEG-PLGA, sustained release, salmon calcitonin
1. Introduction Degradable porous polymer microspheres have attracted much interest for protein and peptide delivery due to their large porous surface, interconnecting pores internal and high specific surface1-3. Protein and peptide could be incorporated within porous microspheres by a solution dipping method, which could protect protein biological activity by avoiding the harsh preparation conditions, for example acute shearing strength produced by homogenizing and water/organic interface caused by a emulsifing.b4, 5. However, most of porous microspheres preparation methods suffered the problem that using pore-forming agents or porogens to produce the pores, including tetrahydrofuran, toluene, sodium oleate, and Pluronic6, 7. R. Jeyanthi reported that 0.4% (w/v) sodium oleate was used as continuous phase to prepare PLGA porous microspheres for peptide delivery8. Each of these must be removed in a second processing step and this requires a relatively tedious process, which was more difficulty to carry out in scale and has potential detrimental effects in clinical, such as immunogenicity. In our previous studies, it was surprising to find that the hydrophilic/hydrophobic characteristic of matrix polymer and solvent used as oil phase played important role on microparticles surface based on solvent evaporation method. When dichloromethane MC was used as oil phase, the surface morphology transferred from smooth to rough surface when using amphiphilic PELA as matrix polymer instead of hydrophobic PLA9. The results inspired us that the microspheres prepared by an appropriate polymer and solvent may be able to self-form the pores 3
without adding any porogens. In other words, traditional porogens can be replaced with amphiphilic polymer itself that are permanent geometric templates and thus do not require leaching in a more process. And this porous microspheres carrier could be applied as a much safer drug release system in further clinical treatment. Nevertheless, experimental and theoretical work is still lacking on using amphiphilic polymer to prepare biodegradable porous microspheres without porogens. And there were a few reports on the effect of polymer property on pore structure and drug loading of porous microspheres. In addition, although PLA and its amphiphilic copolymer were widely us as microspheres matrix, the drug loading methods were usually based on encapsulation technique
8, 10, 11
. There was little literature about the preparation
strategy and forming mechanism of porous microspheres used as adsorption carrier by amphiphilic polymer. In this respect, it is desirable to prepare porous microspheres used as adsorption carrier and explore its forming mechanisms. In addition, the satisfied sustained release systems not only own high loading drug capacity, but also have ideal release profile should include a low initial release followed by a continued release phase, while the protein or peptide release from porous microspheres usually exhibited a high initial burst release with a little further release, which easily leaded to the potential for both under- and over-dosing in vivo12. Thus, it is necessary to develop a strategy to improve the release behavior of porous microspheres after loading drug. Furthermore, the size distributions of microspheres prepared
by
conventional
methods
(e.g.
suspension
polymerization,
dispersion-polymerization, seed swelling polymerization, double emulsification) were very broad, which will result in poor repeatability of preparation, release behavior and drug
efficacy
in
vivo
among
batches13-17.
Therefore,
it
is
is challenging and demanding to develop porous microspheres with narrow size distribution and proper release profiles for drug sustained release without any porogens. In this article, we try to use various polymers as permanent geometric templates to tune the porous structure and elucidate the pores-forming mechanisms. Specifically, we designed the amphiphilic polymer with different hydrophobic segment: 4
mPEG-PLA and mPEG-PLGA. Compared with mPEG-PLA, mPEG-PLGA increased the ratio of glycolide (GA) to lactide (LA), which was used to adjust the property of matrix polymer. Here, a premix membrane emulsification technique combined with a solvent evaporation method was used to prepare porous microspheres with narrow size distribution. Salmon calcitonin (SCT), used in the treatment of postmenopausal osteoporosis, Paget’s disease, and hypercalcemia by injecting subcutaneously daily or three times a week for several months18, was employed as the model peptide drug in this study. SCT was entrapped into different porous microspheres using a solution immersing method and the critical parameters responsible for drug loading were further elucidated by using confocal laser scanning mcroscopy (LSCM). In order to overcome the release problem of porous microspheres, the pores of SCT pre-loaded porous microspheres were closed by treating with different water-miscible solvents that
partially dissolve
mPEG-PLGA copolymer.
The
prepared non-porous
microspheres were determined for sustained SCT delivery. 2. Experimental Section Material: mPEG-PLA and mPEG-PLGA were provided by the Dai Gang Company (Shandong, China), in which the mPEG block has a molecular weight of 2000 Da, PLA and PLGA have molecular weights of 18000 Da. The mole ratio of lactide (LA) and glycolide (GA) in PLGA is 85:15. Poly (vinyl alcohol) (PVA-217, degree of polymerization 1700, degree of hydrolysis 88.5%) was purchased from Kuraray (Japan). SCT (Mw 3431 Da) was kindly supplied by Han Yu Company (Shenzhen, China).All other reagents were of analysis grade. Fast Membrane Emulsifier (FM0210/500M) and micro-porous membrane were provided by Senhui Microsphere Tech(Suzhou)Co., Ltd. Preparation of Porous Microspheres: Porous microspheres were prepared by double emulsion and membrane emulsification method. Distilled water was used as internal
aqueous
phase,
mixing
with
methylene
chloride
dissolving
mPEG-PLA/PLGA (500 mg) by homogenizing. The primary W/O emulsion was further added into 1% w/v polyvinyl alcohol (PVA) solution using as external aqueous phase and emulsified by magnetic stirring. The coarse double emulsions were then 5
pressed into micro-porous membrane to form relative uniformed emulsions. The obtained emulsion was stirring to evaporate organic solvent and collected by centrifugation. Finally, the prepared microspheres were washed with distilled water and then lyophilized for 2 days. Characterization of Porous Microspheres: The surface and internal structure of porous and pore-closed microspheres were observed by scanning electron microscopy(JSM-6700F, JEOL, Japan). The diameter of the microspheres was examined by laser diffraction (Mastersizer 2000, Malvern, UK)19. The span value was used to the express microspheres uniformity. The size distribution of the microspheres became narrower when the value of the span was lower. The pore size, specific surface area and pore size distribution were determined by N 2 adsorption-desorption method on a specific surface area and porosity analyzer (Micromeritics, ASAP2020, USA). 0.15 g dried samples for N2 adsorption were degassed under vacuum at 30 °C for
70
hrs
and
adsorption/desorption
the
characteristics
parameters
method at -195.1
°C.
The
were
results
measured were
by
N2
analyzed
by
Barrett–Joyner–Halenda (BJH) method. SCT Loading: The porous mPEG-PLA/PLGA microspheres loading SCT by immersing porous microspheres (50 mg) in 5 ml of SCT solution and shaking at 25 °C. The SCT-loaded microspheres were recollected by centrifugation. The rest SCT was determined by UV-visible spectrophotometer at absorbance of 275 nm (UV-2550, Shimadzu, Japan). The loading amount of SCT adsorbed into pores was calculated using following equation:
q
ci cf V m
(1)
where ci is the initial SCT solution concentration and c f is the SCT concentration of supernatant after adsorption. V is the total volume of SCT solution. And m is the weight of porous microspheres. All experiments were run in triplicate (n=3) and the results were presented as means±SD. Confocal Laser Scanning Microscopy: For further analyzing the SCT adsorption 6
within microspheres, FITC was employed to label SCT to observe the SCT distribution within microspheres by CLSM. The process for FITC labeling SCT is as follows: FITC solution (5 mg mL-1) was slowly added into SCT solution and mixed. After incubating for 4 h at 4 °C, unbound FITC was separated by ultrafiltration (3,000 MWCO, Millipore). The preparation procedures of FITC-SCT loading microspheres were the same as described in Section 2.2. The FITC-SCT loaded microspheres were added into PBS solution and observed by CLSM (TCS SP2, Leica). In order to avoid the potential source of artifacts, before scanning sample, the site of each glass dish must be determined, for example, x for sample A, y for sample B; then added the same height value “z” to x or y. When we scanned sample, the height were “x+z” for sample A and “y+z” for sample B. Moreover, the samples were prepared with a single layer. Pore Closed of SCT-loaded Porous Microspheres in Aqueous Condition: NaHCO3 and ethyl acetate solution were used to investigate the effect of pore-closed agents on release behavior of porous microspheres. The SCT incorporated porous microspheres (100 mg) were dispersed in 5 mL ethyl acetate (EA) solution with various concentrations (4%, 5%, 7.5%, and 10%, (v/v)) and mixed by gently shaking for 6 hrs. The pore closing process was terminated and the obtained microspheres were centrifuged, washed by distill water and freeze-dried. The surface and internal structure of porous microspheres were observed by scanning electron microscopy (SEM). In the case of or NaHCO3, the solution concentrations were 0.4%, 0.8%, 2%, 5%, and 10% (w/v) , the other procedures were the same as detailed above. In vitro SCT Release Measurement: To detect peptide release profiles, mPEG-PLGA microspheres loading SCT were suspended in PBS buffer and shook at 37 °C. The SCT released in the medium from microspheres was separated by centrifugation and replaced by the same amount of fresh buffer each day at pre-determined time intervals, the release medium. The release medium was used to determine SCT concentration by UV-visible spectrophotometer at absorbance of 275 nm. All analyses were carried out in triplicate (n = 3) and were presented as means ± SD. 7
3. Results and Discussion 3.1. Preparation of Porous Microspheres In our previous study, it was merely to use one type of amphiphilic polymer, mPEG-PLA, to prepare microspheres with pores on surface, and the effect of polymer property on microspheres structure was not certain. In this research, amphiphilic polymer mPEG-PLA and mPEG-PLGA were chosen as microsphere matrix to investigate the influence of polymer characteristic on microspheres forming. At the meanwhile, methylene chloride, the low boil point solvent, was used as oil phase to prepare porous microspheres without adding any porogens. As seen in Figure 1, uniformed porous mPEG-PLA and mPEG-PLGA microspheres were successfully prepared by a double emulsion-premix membrane emulsification technique combined with a solvent evaporation method. mPEG-PLA and mPEG-PLGA microspheres had mean diameters of 9.82 and 11.61μm, respectively, and span values of 0.672 and 0.724, respectively. (Figure S1 in Supporting Information). The similarity of particles size and distribution could ensure the credibility of comparison of different polymer since particle size maybe affect the microspheres morphology20-22. Figure 1 shows that both microspheres displayed regular sphericity, but different component microspheres had various porous surface morphology and internal structure. mPEG-PLGA microspheres showed obvious macro-pores morphology (Figure 1b2), while mPEG-PLA showed relatively small and shallow pores (Figure 1a2). As seen in Table 1, the mean pore sizes of mPEG-PLA and mPEG-PLGA microspheres were 4.10 and 7.27 nm respectively, specific surface were 22.10 and 105.96 m2/g respectively. Importantly, close examination of cross-section view suggested obviously different structure. Figure 1b3 shows that mPEG-PLGA microspheres have a honeycomb like structure, whereas mPEG-PLA microspheres have a solid structure internally (Figure 1a3).
8
Figure 1. Scanning electron micrographs of surface morphology (upper row a1,b1), magnified surface morphology (middle row a2,b2), and cross-section (bottom row a3,b3) for mPEG- PLA (left column) and mPEG- PLGA (right column) microspheres.
Table 1. Properties of mPEG-PLA and mPEG-PLGA Porous Microspheres Particle size Polymer
Span
Surface area
Drug loading
(m /g)
2
(mg/g)
Pore size (nm)
(μm) mPEG-PLA
9.82
0.672
4.10
22.10
26.7
mPEG-PLGA
10.61
0.724
7.27
105.96
41.2
3.2. Pore-forming Mechanisms of the Porous Microspheres Both kinds of prepared microspheres had porous surface morphology although showing different pore size and shape. There are mainly two reasons to explain the pore-forming mechanisms of microspheres surface (Figure 2). We proposed that the main reason for pores forming was the hydrophilic PEG segments extending on the surface of microspheres. During solidification process, these hydrophilic PEG 9
segments absorbed a large amount of water molecules and fully swelled. After microspheres lyophilizing, the water molecules adsorbed around PEG segments would evaporate, then forming the pores on the microspheres surface. The other reason was the evaporate rate of solvent used as oil phase. Shi et al. investigated the effect of solvent types on porous structure of microspheres．It was concluded that the removal rate of solvent plays the decisive role in determining the porous structure of microspheres23. In this research, MC owned low boiling point (39.8°C) and then evaporated fast during solidification process. The fast rate of solvent evaporation easily led to local explosion on un-solidified microspheres surface and the solvent could easily penetrate through the soft layer to form pores. Honeycomb like internal structure of mPEG-PLGA microspheres in this study was mainly formed by the affinity effect between MC and the polymer. It was reported that the affinity between solvent and PLGA decreased as the LA/GA ratio decreased24. This implied that compared with mPEG-PLA, the affinity between solvent (methylene dichloride) and mPEG-PLGA was relative weak due to introduction of GA segment. The phase separation of matrix polymer became easily and the quick diffusion of methylene dichloride led to rapid solidification of mPEG-PLGA microspheres. This means there is no time for polymer precipitation to fill the water droplets pores, and then the internal and external porous structure could be retained25, 26. Therefore the pore size and specific surface of mPEG-PLGA microspheres were higher (Figure 2c). While for mPEG-PLA, the affinity between methylene dichloride and mPEG-PLA was stronger. The phase separation of matrix polymer became difficultly. Then it has enough time for polymer precipitation to fill the pores and the porous surface would gradually transform into relative smooth surface, resulting in the small and shallow pores. During the microspheres forming process, there would be an influx of water from inner water phase When the osmotic pressure of the outer water phase was greater than that of the inner water, resulting in the connect channels forming. Meanwhile, the internal pores and channels were also filled in by polymer precipitation, forming the low specific surface (Figure 2b). The mechanism was consistent with the study of Ruan et al. They investigated the influences of material 10
hydrophilic/hydrophobic characteristic on surface morphology of microspheres prepared by double emulsion method27. The literature reported that the pores of relative hydrophilic microspheres were more compact than those of hydrophobic microspheres. So, the property of hydrophobic segment played important role on structure characteristic of pores. We can prepare different porous microspheres by using polymers with different hydrophobic segments. Amphiphilic polymer can be employed as permanent geometric templates to prepare porous microspheres without adding any porogens.
Figure 2. A schematic diagram of the mechanism resulting in porous microspheres with different structure. (a) Porous surface morphology formed by local explosion and PEG adsorbing water; (b) Low specific surface area and solid internal structure of microspheres due to the strong affinity between MC and mPEG-PLA; (c) High specific surface area and honeycomb like internal structure due to weak affinity between MC and mPEG-PLGA.
3.3. Adsorption of SCT on/into Porous Microspheres The ideal adsorption carrier usually obtained high specific surface. In order to confirm mPEG-PLGA microspheres own the better drug adsorption capacity, SCT was loaded into different porous microspheres by a simple solution dipping method. The drug loading amounts were 26.7 mg/g and 41.2 mg/g for mPEG-PLA and mPEG-PLGA
11
microspheres, respectively. A probable explanation for the remarkable difference in drug loading was that mPEG-PLGA microspheres owned larger pore size and higher specific surface. mPEG-PLGA microspheres have larger pore size, which was beneficial to adsorb SCT and the adsorption rate was fast. Meanwhile, mPEG-PLGA microspheres have more sites to adsorb SCT due to the internal pore structure and high specific surface, which led to the more loaded SCT during the adsorbing process. It is proposed that the porous on surface and reticulated structure internally was a crucial pathway for SCT adsorption. Moreover, in order to certify the above conclusion, the distribution of SCT within the microspheres was observed by CLSM. Figure 3 shows the CLSM images of mPEG-PLA and mPEG-PLGA microspheres adsorbing FITC-salcatonin. The green fluorescence was only existed near the surface of mPEG-PLA microspheres but no internal, forming a thin fluorescence circle outside (Figure 3b). Compared to mPEG-PLA, the much thicker fluorescence circle can be seen in mPEG-PLGA microspheres. More FITC-salcatonin were observed in mPEG-PLGA microspheres(Figure 3c) and some FITC-salcatonin even entered into the interior or core of mPEG-PLGA microspheres (white arrow indicated). The results suggested that the different LA/GA ratios could modulate the pore sizes and specific surface of porous microspheres, and then influence the drug loading. Thus, it is critical to select suitable polymers with appropriate hydrophilicity to prepare porous microspheres as drug sustained release systems. In this respect, mPEG-PLGA porous microspheres was more suitable to load peptides and chosen as the drug delivery carrier to investigate the release behavior of SCT.
Figure 3. Drug loading of different porous microspheres (a) and corresponding 12
CLSM images of microspheres absorbing FITC labeled salcatonin: (b) mPEG-PLA; (c) mPEG-PLGA.
Furthermore, we investigated the loading dynamics curves of mPEG-PLGA porous microspheres since the adsorption process of peptides into microspheres is a dynamic equilibrium process. The adsorption capacity is no longer increasing when it reaches adsorption equilibrium. As shown in Figure S2, when the adsorption time is 8 h, the adsorption capacity of the microspheres is saturated. This is because the concentration of the SCT in the liquid phase at beginning is higher, and the differential concentration between liquid phase and the solid phase was high. It can be seen from the slope of curve that there was a quick adsorption rate during 0-8 h, and the adsorption amount of the microspheres reaches the highest at 8 h. After 8 h, the concentration of SCT in solution decreases gradually, and the concentration difference between the liquid and the solid phase decreases correspondingly, which leads to the reducing of the adsorption capacity. Therefore, 8 h was chosen as the optimal adsorption time.
3.4. In vitro Release Profiles of SCT and Pore Closing The ideal release profile should include a low initial burst release with a sustained release phase later on. The microspheres morphology surface and other factors may affect the burst release profile28, 29. As shown in Figure 4, mPEG-PLGA microspheres showed a high burst release of 38.7% and complete release only within 8 hrs because SCT adsorbed on the surface or loaded in internal diffused out quickly in first few hours. Similarly, mPEG-PLA microspheres as control showed a burst release of 29.6% and a total of complete release within 10 hrs. mPEG-PLA microspheres showed a lower burst release since the pores were smaller and then the diffusion rate of SCT was slower. However, the complete release time of mPEG-PLA microspheres merely extended to 10 hrs compared to that of mPEG-PLGA microspheres due to the low loading amounts. A large initial release was not beneficial to patients who needed to be injected for long term period. In order to inhibit SCT fast diffusion and improve the release behavior, the pores would be closed by water-miscible solvents that
13
partially dissolve mPEG-PLGA. In this work, NaHCO3 and ethyl acetate were used respectively to compare the pore closing effect. As shown in Figure 5, the pores became small with increasing the concentration of NaHCO 3, however there are still pores could be found on microspheres surface. When the concentration increased to 10%, the microspheres showed rough surface rather than porous morphology. The release profile of pore-closed microspheres by 10% NaHCO3 was shown in Figure 6, the burst release decreased to about 35% and the later release lasted more than 4 days, suggesting that the release profile was improved than before. However, the whole release period was nearly four days, which can’t attend the treatment requirement for at least one week.
Figure 4. Cumulative release of SCT from porous mPEG-PLGA and mPEG-PLA microspheres
14
Figure 5. The surface morphology of microspheres was closed by different concentration of NaHCO3: (a) 0.4%; (b) 0.8%; (c) 2%; (d) 5%; (e) 10%.
Figure 6. Cumulative release of SCT from pore-closed microspheres by NaHCO3 (blue circle) and EA (red triangle) solution, respectively.
15
Figure 7. The surface morphology of microspheres was closed by different concentration of EA: (a) 4%; (b) 5%; (c) 7.5%; (d) 10%.
Therefore, we tried to use another pore closing reagent, ethyl acetate, to investigate its effect of inhibiting burst release. As shown in Figure 7, the surface morphology changed from more pores to less pores with increasing the concentration of ethyl acetate. When the concentration of ethyl acetate was 7.5%, few pores could be found on microspheres surface. While, when the concentration increased to 10%, the microspheres were completely dissolved from sphericity to film. The results revealed that the porous microspheres were closed best after treated by 7.5% ethyl acetate. Meanwhile, the release of pore-closed microspheres was also investigated. Shown in Figure 6, the burst release decreased to about 20% and the later release lasted more than 10 days, suggesting that the release profile of pore-closed microspheres was greatly improved than those of unsealed microspheres. The improved release patterns were mainly due to a diffusion barrier on the microspheres surface and reducing the loss of peptides. Therefore, compared with NaHCO3, ethyl acetate was more effective 16
for mPEG-PLGA microspheres to treat pores and decrease burst release (from 38.7% to 20%) and prolongs the later period release (from 8 h to 10 days). It was certified that using ethyl acetate solution to close pores could improve the release behavior of porous microspheres.
4. Conclusion A novel drug delivery system with porous microspheres followed a pore closing process was designed in this manuscript. mPEG-PLGA porous microspheres were successfully prepared without porogens by using suitable amphiphilic copolymer and solvent with boiling point based on double emulsion-solvent evaporation method. The particle sizes could be controlled by premix membrane emulsification technique. The porous surface morphology was mainly caused by hydrophilic PEG segments absorbing water molecules and local explosion of microspheres. Moreover, the pore sizes and internal structure could be control by adjusting the component of matrix polymer. In addition, porous microspheres containing SCT were treated with ethyl acetate solution can overcome the release problems of porous microspheres and improve release profile. These results suggest that the mPEG-PLGA can be used as a permanent geometric template to tune the porous structure, thus offering an alternative to prepare porous microspheres in the drug delivery formulations. Compared with conventional methods[6,7], porogens were needn’t in this study. Most importantly, the study investigated the key factors influencing preparation and illustrated the forming mechanisms in detail, which will be significant for the development of amphiphilic polymer formulation. The drug delivery system with porous microspheres followed a pore closing process could be potential use in small molecular drug loading, such as peptide, gene, and chemical medicine. We will investigate the effect of pore size and porosity on the loading efficiency of different drug in future work.
Acknowledgements We thank the financial support of National Natural Science Foundation of China 17
(Nos.21306208, 51173187, and 21336010).
References 1
T. Ehtezazi, C. Washington, Journal of Controlled Release 2000, 68, 361-372.
2
X. Niu, Q. Feng, M. Wang, X. Guo, Q. Zheng, Polymer Degradation and Stability 2009, 94,
176-182 http://dx.doi.org/10.1016/j.polymdegradstab.2008.11.008. 3
C. Dai, Y. Wang, X. Hou, Carbohydrate Polymers 2012, 87, 2338-2343
http://dx.doi.org/10.1016/j.carbpol.2011.11.004. 4
L. Sun, S. Zhou, W. Wang, X. Li, J. Wang, J. Weng, Colloids and Surfaces A-physicochemical
and Engineering Aspects 2009, 345, 173-181 10.1016/j.colsurfa.2009.04.053. 5
G. Ma, Journal of Controlled Release 2014, 193, 324-340 10.1016/j.jconrel.2014.09.003.
6
D.-x. Hao, F.-l. Gong, W. Wei, G.-h. Hu, G.-h. Ma, Z.-g. Su, Journal of Colloid and Interface
Science 2008, 323, 52-59 10.1016/j.jcis.2008.03.039. 7
H. K. Kim, H. J. Chung, T. G. Park, Journal of Controlled Release 2006, 112, 167-174
10.1016/j.jconrel.2006.02.004. 8
R. Jeyanthi, B. C. Thanoo, R. C. Metha, P. P. DeLuca, Journal of Controlled Release 1996, 38,
235-244 10.1016/0168-3659(95)00125-5. 9
Y. Wei, Y. Wang, L. Wang, D. Hao, G. Ma, Colloids and Surfaces B-Biointerfaces 2011, 87,
399-408 10.1016/j.colsurfb.2011.05.051. 10
F. Ito, H. Fujimora, H. Honnami, H. Kawakami, K. Kanamura, K. Makino, Colloids and
Surfaces B-biointerfaces 2008, 66, 65-70 10.1016/j.colsurfb.2008.05.011. 11
J. Buske, C. Konig, S. Bassarab, A. Lamprecht, S. Muhlau, K. G. Wagner, European Journal
Of Pharmaceutics and Biopharmaceutics 2012, 81, 57-63 10.1016/j.ejpb.2012.01.009. 12
D. A. Edwards, J. Hanes, G. Caponetti, J. Hrkach, A. BenJebria, M. L. Eskew, J. Mintzes, D.
Deaver, N. Lotan, R. Langer, Science 1997, 276, 1868-1871 10.1126/science.276.5320.1868. 13
P. Tiainen, P.-E. Gustavsson, A. Ljunglof, P.-O. Larsson, Journal Of Chromatography A 2007,
1138, 84-94 10.1016/j.chroma.2006.10.029. 14
M. Nakai, M. Niinomi, D. Ishii, Journal of the Mechanical Behavior of Biomedical Materials
2011, 4, 1206-1218 10.1016/j.jmbbm.2011.04.003. 15
C. T. L. Luz, F. M. B. Coutinho, Polymer 2001, 42, 4931-4938.
16
S. Zhang, X. Huang, N. Yao, C. Horváth, Journal Of Chromatography A 2002, 948, 193-201
http://dx.doi.org/10.1016/S0021-9673(01)01425-X. 17
G. De Rosa, F. Quaglia, M. I. La Rotonda, M. Besnard, E. Fattal, International Journal of
Pharmaceutics 2002, 242, 225-228 10.1016/s0378-5173(02)00162-x. 18
J. Wei, J. Wang, Y. Gong, R. Zeng, Molecular Medicine Reports 2015, 12, 1717-1726
10.3892/mmr.2015.3637. 19
Y. Wei, Y.-X. Wang, W. Wang, S. V. Ho, W. Wei, G.-H. Ma, Journal of Materials Chemistry
2011, 21, 12691-12699 10.1039/c1jm12643a. 20
G. J. S. Dawes, L. E. Fratila-Apachitei, K. Mulia, I. Apachitei, G. J. Witkamp, J. Duszczyk,
Journal of Materials Science-Materials in Medicine 2009, 20, 1089-1094 10.1007/s10856-008-3666-0. 21
Y. Y. Yang, T. S. Chung, X. L. Bai, W. K. Chan, Chemical Engineering Science 2000, 55,
2223-2236 10.1016/s0009-2509(99)00503-5.
18
22
Y. Liu, X. Deng, Journal of Controlled Release 2002, 83, 147-155.
23
X. Shi, L. Sun, Z. Gan, Acta Polymerica Sinica 2011, 866-873 10.3724/sp.j.1105.2011.11113.
24
H. Murakami, M. Kobayashi, H. Takeuchi, Y. Kawashima, Powder Technology 2000, 107,
137-143 10.1016/s0032-5910(99)00182-5. 25
M. Cho, H. Sah, Journal of Microencapsulation 2005, 22, 1-12 10.1080/02652040400026269.
26
A. Rawat, D. J. Burgess, International Journal of Pharmaceutics 2010, 394, 99-105
10.1016/j.ijpharm.2010.05.013. 27
G. Ruan, S. S. Feng, Q. T. Li, Journal of Controlled Release 2002, 84, 151-160.
28
M. van de Weert, W. Hennink, W. Jiskoot, Pharmaceutical Research 2000, 17, 1159-1167.
29
M. A. Ibrahim, A. Ismail, M. I. Fetouh, A. Gopferich, Journal of Controlled Release 2005,
106, 241-252 10.1016/j.jconrel.2005.02.025.
19
Graphical Abstract
S0021-9797(16)30330-7 http://dx.doi.org/10.1016/j.jcis.2016.05.045 YJCIS 21289
To appear in:
Journal of Colloid and Interface Science
Received Date: Revised Date: Accepted Date:
15 March 2016 23 May 2016 23 May 2016
Please cite this article as: Y. Wei, Y. Wang, H. Zhang, W. Zhou, G. Ma, A Novel Strategy for the Preparation of Porous Microspheres and Its Application in Peptide Drug Loading, Journal of Colloid and Interface Science (2016), doi: http://dx.doi.org/10.1016/j.jcis.2016.05.045
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
A Novel Strategy for the Preparation of SCT-loaded Porous mPEG-PLGA Microspheres with Narrow Size Distribution
Yi Wei1, Yuxia Wang1, Huixia Zhang2, Weiqing Zhou1, and Guanghui Ma1,* 1
National Key Laboratory of Biochemical Engineering, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, People’s Republic of China 2 Shenyang Agriculture University, Shenyang, Liaoning 110161, People’s Republic of China ∗ Corresponding author. Tel.: +86 10 82627072; fax: +86 10 82627072. E-mail addresses: [email protected] The first two authors, Yi Wei and Yuxia Wang, contributed equally.
1
A Novel Strategy for the Preparation of Porous Microspheres and Its Application in Peptide Drug Loading
Yi Wei1, Yuxia Wang1, Huixia Zhang2, Weiqing Zhou1, and Guanghui Ma1,* 1
National Key Laboratory of Biochemical Engineering, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, People’s Republic of China 2 Shenyang Agriculture University, Shenyang, Liaoning 110161, People’s Republic of China ∗ Corresponding author. Tel.: +86 10 82627072; fax: +86 10 82627072. E-mail addresses: [email protected] The first two authors, Yi Wei and Yuxia Wang, contributed equally.
Abstract: A new strategy is developed to prepare porous microspheres with narrow size distribution for peptides controlled release, involving a fabrication of porous microspheres without any porogens followed by a pore closing process. Amphiphilic polymers with different hydrophobic segments (poly(monomethoxypolyethylene glycol-co-D,L-lactide) (mPEG-PLA), poly(monomethoxypolyethylene glycol-coD,L-lactic-co-glycolic acid) (mPEG-PLGA)) are employed as microspheres matrix to prepare porous microspheres based on a double emulsion-premix membrane emulsification technique combined with a solvent evaporation method. Both microspheres possess narrow size distribution and porous surface, which are mainly caused by a) hydrophilic polyethylene glycol (PEG) segments absorbing water molecules followed by a water evaporation process and b) local explosion of microspheres due to fast evaporation of dichloromethane(MC). Importantly, mPEG-PLGA microspheres have a honeycomb like structure while mPEG-PLA microspheres have a solid structure internally, illustrating that the different
2
hydrophobic segments could modulate the affinity between solvent and matrix polymer and influence the phase separation rate of microspheres matrix. Long term release patterns are demonstrated with pore-closed microspheres, which are prepared from mPEG-PLGA microspheres loading salmon calcitonin (SCT). These results suggest that it is potential to construct porous microspheres for drug sustained release using permanent geometric templates as new porogens. Keywords: porous microspheres, narrow size distribution, mPEG-PLGA, sustained release, salmon calcitonin
1. Introduction Degradable porous polymer microspheres have attracted much interest for protein and peptide delivery due to their large porous surface, interconnecting pores internal and high specific surface1-3. Protein and peptide could be incorporated within porous microspheres by a solution dipping method, which could protect protein biological activity by avoiding the harsh preparation conditions, for example acute shearing strength produced by homogenizing and water/organic interface caused by a emulsifing.b4, 5. However, most of porous microspheres preparation methods suffered the problem that using pore-forming agents or porogens to produce the pores, including tetrahydrofuran, toluene, sodium oleate, and Pluronic6, 7. R. Jeyanthi reported that 0.4% (w/v) sodium oleate was used as continuous phase to prepare PLGA porous microspheres for peptide delivery8. Each of these must be removed in a second processing step and this requires a relatively tedious process, which was more difficulty to carry out in scale and has potential detrimental effects in clinical, such as immunogenicity. In our previous studies, it was surprising to find that the hydrophilic/hydrophobic characteristic of matrix polymer and solvent used as oil phase played important role on microparticles surface based on solvent evaporation method. When dichloromethane MC was used as oil phase, the surface morphology transferred from smooth to rough surface when using amphiphilic PELA as matrix polymer instead of hydrophobic PLA9. The results inspired us that the microspheres prepared by an appropriate polymer and solvent may be able to self-form the pores 3
without adding any porogens. In other words, traditional porogens can be replaced with amphiphilic polymer itself that are permanent geometric templates and thus do not require leaching in a more process. And this porous microspheres carrier could be applied as a much safer drug release system in further clinical treatment. Nevertheless, experimental and theoretical work is still lacking on using amphiphilic polymer to prepare biodegradable porous microspheres without porogens. And there were a few reports on the effect of polymer property on pore structure and drug loading of porous microspheres. In addition, although PLA and its amphiphilic copolymer were widely us as microspheres matrix, the drug loading methods were usually based on encapsulation technique
8, 10, 11
. There was little literature about the preparation
strategy and forming mechanism of porous microspheres used as adsorption carrier by amphiphilic polymer. In this respect, it is desirable to prepare porous microspheres used as adsorption carrier and explore its forming mechanisms. In addition, the satisfied sustained release systems not only own high loading drug capacity, but also have ideal release profile should include a low initial release followed by a continued release phase, while the protein or peptide release from porous microspheres usually exhibited a high initial burst release with a little further release, which easily leaded to the potential for both under- and over-dosing in vivo12. Thus, it is necessary to develop a strategy to improve the release behavior of porous microspheres after loading drug. Furthermore, the size distributions of microspheres prepared
by
conventional
methods
(e.g.
suspension
polymerization,
dispersion-polymerization, seed swelling polymerization, double emulsification) were very broad, which will result in poor repeatability of preparation, release behavior and drug
efficacy
in
vivo
among
batches13-17.
Therefore,
it
is
is challenging and demanding to develop porous microspheres with narrow size distribution and proper release profiles for drug sustained release without any porogens. In this article, we try to use various polymers as permanent geometric templates to tune the porous structure and elucidate the pores-forming mechanisms. Specifically, we designed the amphiphilic polymer with different hydrophobic segment: 4
mPEG-PLA and mPEG-PLGA. Compared with mPEG-PLA, mPEG-PLGA increased the ratio of glycolide (GA) to lactide (LA), which was used to adjust the property of matrix polymer. Here, a premix membrane emulsification technique combined with a solvent evaporation method was used to prepare porous microspheres with narrow size distribution. Salmon calcitonin (SCT), used in the treatment of postmenopausal osteoporosis, Paget’s disease, and hypercalcemia by injecting subcutaneously daily or three times a week for several months18, was employed as the model peptide drug in this study. SCT was entrapped into different porous microspheres using a solution immersing method and the critical parameters responsible for drug loading were further elucidated by using confocal laser scanning mcroscopy (LSCM). In order to overcome the release problem of porous microspheres, the pores of SCT pre-loaded porous microspheres were closed by treating with different water-miscible solvents that
partially dissolve
mPEG-PLGA copolymer.
The
prepared non-porous
microspheres were determined for sustained SCT delivery. 2. Experimental Section Material: mPEG-PLA and mPEG-PLGA were provided by the Dai Gang Company (Shandong, China), in which the mPEG block has a molecular weight of 2000 Da, PLA and PLGA have molecular weights of 18000 Da. The mole ratio of lactide (LA) and glycolide (GA) in PLGA is 85:15. Poly (vinyl alcohol) (PVA-217, degree of polymerization 1700, degree of hydrolysis 88.5%) was purchased from Kuraray (Japan). SCT (Mw 3431 Da) was kindly supplied by Han Yu Company (Shenzhen, China).All other reagents were of analysis grade. Fast Membrane Emulsifier (FM0210/500M) and micro-porous membrane were provided by Senhui Microsphere Tech(Suzhou)Co., Ltd. Preparation of Porous Microspheres: Porous microspheres were prepared by double emulsion and membrane emulsification method. Distilled water was used as internal
aqueous
phase,
mixing
with
methylene
chloride
dissolving
mPEG-PLA/PLGA (500 mg) by homogenizing. The primary W/O emulsion was further added into 1% w/v polyvinyl alcohol (PVA) solution using as external aqueous phase and emulsified by magnetic stirring. The coarse double emulsions were then 5
pressed into micro-porous membrane to form relative uniformed emulsions. The obtained emulsion was stirring to evaporate organic solvent and collected by centrifugation. Finally, the prepared microspheres were washed with distilled water and then lyophilized for 2 days. Characterization of Porous Microspheres: The surface and internal structure of porous and pore-closed microspheres were observed by scanning electron microscopy(JSM-6700F, JEOL, Japan). The diameter of the microspheres was examined by laser diffraction (Mastersizer 2000, Malvern, UK)19. The span value was used to the express microspheres uniformity. The size distribution of the microspheres became narrower when the value of the span was lower. The pore size, specific surface area and pore size distribution were determined by N 2 adsorption-desorption method on a specific surface area and porosity analyzer (Micromeritics, ASAP2020, USA). 0.15 g dried samples for N2 adsorption were degassed under vacuum at 30 °C for
70
hrs
and
adsorption/desorption
the
characteristics
parameters
method at -195.1
°C.
The
were
results
measured were
by
N2
analyzed
by
Barrett–Joyner–Halenda (BJH) method. SCT Loading: The porous mPEG-PLA/PLGA microspheres loading SCT by immersing porous microspheres (50 mg) in 5 ml of SCT solution and shaking at 25 °C. The SCT-loaded microspheres were recollected by centrifugation. The rest SCT was determined by UV-visible spectrophotometer at absorbance of 275 nm (UV-2550, Shimadzu, Japan). The loading amount of SCT adsorbed into pores was calculated using following equation:
q
ci cf V m
(1)
where ci is the initial SCT solution concentration and c f is the SCT concentration of supernatant after adsorption. V is the total volume of SCT solution. And m is the weight of porous microspheres. All experiments were run in triplicate (n=3) and the results were presented as means±SD. Confocal Laser Scanning Microscopy: For further analyzing the SCT adsorption 6
within microspheres, FITC was employed to label SCT to observe the SCT distribution within microspheres by CLSM. The process for FITC labeling SCT is as follows: FITC solution (5 mg mL-1) was slowly added into SCT solution and mixed. After incubating for 4 h at 4 °C, unbound FITC was separated by ultrafiltration (3,000 MWCO, Millipore). The preparation procedures of FITC-SCT loading microspheres were the same as described in Section 2.2. The FITC-SCT loaded microspheres were added into PBS solution and observed by CLSM (TCS SP2, Leica). In order to avoid the potential source of artifacts, before scanning sample, the site of each glass dish must be determined, for example, x for sample A, y for sample B; then added the same height value “z” to x or y. When we scanned sample, the height were “x+z” for sample A and “y+z” for sample B. Moreover, the samples were prepared with a single layer. Pore Closed of SCT-loaded Porous Microspheres in Aqueous Condition: NaHCO3 and ethyl acetate solution were used to investigate the effect of pore-closed agents on release behavior of porous microspheres. The SCT incorporated porous microspheres (100 mg) were dispersed in 5 mL ethyl acetate (EA) solution with various concentrations (4%, 5%, 7.5%, and 10%, (v/v)) and mixed by gently shaking for 6 hrs. The pore closing process was terminated and the obtained microspheres were centrifuged, washed by distill water and freeze-dried. The surface and internal structure of porous microspheres were observed by scanning electron microscopy (SEM). In the case of or NaHCO3, the solution concentrations were 0.4%, 0.8%, 2%, 5%, and 10% (w/v) , the other procedures were the same as detailed above. In vitro SCT Release Measurement: To detect peptide release profiles, mPEG-PLGA microspheres loading SCT were suspended in PBS buffer and shook at 37 °C. The SCT released in the medium from microspheres was separated by centrifugation and replaced by the same amount of fresh buffer each day at pre-determined time intervals, the release medium. The release medium was used to determine SCT concentration by UV-visible spectrophotometer at absorbance of 275 nm. All analyses were carried out in triplicate (n = 3) and were presented as means ± SD. 7
3. Results and Discussion 3.1. Preparation of Porous Microspheres In our previous study, it was merely to use one type of amphiphilic polymer, mPEG-PLA, to prepare microspheres with pores on surface, and the effect of polymer property on microspheres structure was not certain. In this research, amphiphilic polymer mPEG-PLA and mPEG-PLGA were chosen as microsphere matrix to investigate the influence of polymer characteristic on microspheres forming. At the meanwhile, methylene chloride, the low boil point solvent, was used as oil phase to prepare porous microspheres without adding any porogens. As seen in Figure 1, uniformed porous mPEG-PLA and mPEG-PLGA microspheres were successfully prepared by a double emulsion-premix membrane emulsification technique combined with a solvent evaporation method. mPEG-PLA and mPEG-PLGA microspheres had mean diameters of 9.82 and 11.61μm, respectively, and span values of 0.672 and 0.724, respectively. (Figure S1 in Supporting Information). The similarity of particles size and distribution could ensure the credibility of comparison of different polymer since particle size maybe affect the microspheres morphology20-22. Figure 1 shows that both microspheres displayed regular sphericity, but different component microspheres had various porous surface morphology and internal structure. mPEG-PLGA microspheres showed obvious macro-pores morphology (Figure 1b2), while mPEG-PLA showed relatively small and shallow pores (Figure 1a2). As seen in Table 1, the mean pore sizes of mPEG-PLA and mPEG-PLGA microspheres were 4.10 and 7.27 nm respectively, specific surface were 22.10 and 105.96 m2/g respectively. Importantly, close examination of cross-section view suggested obviously different structure. Figure 1b3 shows that mPEG-PLGA microspheres have a honeycomb like structure, whereas mPEG-PLA microspheres have a solid structure internally (Figure 1a3).
8
Figure 1. Scanning electron micrographs of surface morphology (upper row a1,b1), magnified surface morphology (middle row a2,b2), and cross-section (bottom row a3,b3) for mPEG- PLA (left column) and mPEG- PLGA (right column) microspheres.
Table 1. Properties of mPEG-PLA and mPEG-PLGA Porous Microspheres Particle size Polymer
Span
Surface area
Drug loading
(m /g)
2
(mg/g)
Pore size (nm)
(μm) mPEG-PLA
9.82
0.672
4.10
22.10
26.7
mPEG-PLGA
10.61
0.724
7.27
105.96
41.2
3.2. Pore-forming Mechanisms of the Porous Microspheres Both kinds of prepared microspheres had porous surface morphology although showing different pore size and shape. There are mainly two reasons to explain the pore-forming mechanisms of microspheres surface (Figure 2). We proposed that the main reason for pores forming was the hydrophilic PEG segments extending on the surface of microspheres. During solidification process, these hydrophilic PEG 9
segments absorbed a large amount of water molecules and fully swelled. After microspheres lyophilizing, the water molecules adsorbed around PEG segments would evaporate, then forming the pores on the microspheres surface. The other reason was the evaporate rate of solvent used as oil phase. Shi et al. investigated the effect of solvent types on porous structure of microspheres．It was concluded that the removal rate of solvent plays the decisive role in determining the porous structure of microspheres23. In this research, MC owned low boiling point (39.8°C) and then evaporated fast during solidification process. The fast rate of solvent evaporation easily led to local explosion on un-solidified microspheres surface and the solvent could easily penetrate through the soft layer to form pores. Honeycomb like internal structure of mPEG-PLGA microspheres in this study was mainly formed by the affinity effect between MC and the polymer. It was reported that the affinity between solvent and PLGA decreased as the LA/GA ratio decreased24. This implied that compared with mPEG-PLA, the affinity between solvent (methylene dichloride) and mPEG-PLGA was relative weak due to introduction of GA segment. The phase separation of matrix polymer became easily and the quick diffusion of methylene dichloride led to rapid solidification of mPEG-PLGA microspheres. This means there is no time for polymer precipitation to fill the water droplets pores, and then the internal and external porous structure could be retained25, 26. Therefore the pore size and specific surface of mPEG-PLGA microspheres were higher (Figure 2c). While for mPEG-PLA, the affinity between methylene dichloride and mPEG-PLA was stronger. The phase separation of matrix polymer became difficultly. Then it has enough time for polymer precipitation to fill the pores and the porous surface would gradually transform into relative smooth surface, resulting in the small and shallow pores. During the microspheres forming process, there would be an influx of water from inner water phase When the osmotic pressure of the outer water phase was greater than that of the inner water, resulting in the connect channels forming. Meanwhile, the internal pores and channels were also filled in by polymer precipitation, forming the low specific surface (Figure 2b). The mechanism was consistent with the study of Ruan et al. They investigated the influences of material 10
hydrophilic/hydrophobic characteristic on surface morphology of microspheres prepared by double emulsion method27. The literature reported that the pores of relative hydrophilic microspheres were more compact than those of hydrophobic microspheres. So, the property of hydrophobic segment played important role on structure characteristic of pores. We can prepare different porous microspheres by using polymers with different hydrophobic segments. Amphiphilic polymer can be employed as permanent geometric templates to prepare porous microspheres without adding any porogens.
Figure 2. A schematic diagram of the mechanism resulting in porous microspheres with different structure. (a) Porous surface morphology formed by local explosion and PEG adsorbing water; (b) Low specific surface area and solid internal structure of microspheres due to the strong affinity between MC and mPEG-PLA; (c) High specific surface area and honeycomb like internal structure due to weak affinity between MC and mPEG-PLGA.
3.3. Adsorption of SCT on/into Porous Microspheres The ideal adsorption carrier usually obtained high specific surface. In order to confirm mPEG-PLGA microspheres own the better drug adsorption capacity, SCT was loaded into different porous microspheres by a simple solution dipping method. The drug loading amounts were 26.7 mg/g and 41.2 mg/g for mPEG-PLA and mPEG-PLGA
11
microspheres, respectively. A probable explanation for the remarkable difference in drug loading was that mPEG-PLGA microspheres owned larger pore size and higher specific surface. mPEG-PLGA microspheres have larger pore size, which was beneficial to adsorb SCT and the adsorption rate was fast. Meanwhile, mPEG-PLGA microspheres have more sites to adsorb SCT due to the internal pore structure and high specific surface, which led to the more loaded SCT during the adsorbing process. It is proposed that the porous on surface and reticulated structure internally was a crucial pathway for SCT adsorption. Moreover, in order to certify the above conclusion, the distribution of SCT within the microspheres was observed by CLSM. Figure 3 shows the CLSM images of mPEG-PLA and mPEG-PLGA microspheres adsorbing FITC-salcatonin. The green fluorescence was only existed near the surface of mPEG-PLA microspheres but no internal, forming a thin fluorescence circle outside (Figure 3b). Compared to mPEG-PLA, the much thicker fluorescence circle can be seen in mPEG-PLGA microspheres. More FITC-salcatonin were observed in mPEG-PLGA microspheres(Figure 3c) and some FITC-salcatonin even entered into the interior or core of mPEG-PLGA microspheres (white arrow indicated). The results suggested that the different LA/GA ratios could modulate the pore sizes and specific surface of porous microspheres, and then influence the drug loading. Thus, it is critical to select suitable polymers with appropriate hydrophilicity to prepare porous microspheres as drug sustained release systems. In this respect, mPEG-PLGA porous microspheres was more suitable to load peptides and chosen as the drug delivery carrier to investigate the release behavior of SCT.
Figure 3. Drug loading of different porous microspheres (a) and corresponding 12
CLSM images of microspheres absorbing FITC labeled salcatonin: (b) mPEG-PLA; (c) mPEG-PLGA.
Furthermore, we investigated the loading dynamics curves of mPEG-PLGA porous microspheres since the adsorption process of peptides into microspheres is a dynamic equilibrium process. The adsorption capacity is no longer increasing when it reaches adsorption equilibrium. As shown in Figure S2, when the adsorption time is 8 h, the adsorption capacity of the microspheres is saturated. This is because the concentration of the SCT in the liquid phase at beginning is higher, and the differential concentration between liquid phase and the solid phase was high. It can be seen from the slope of curve that there was a quick adsorption rate during 0-8 h, and the adsorption amount of the microspheres reaches the highest at 8 h. After 8 h, the concentration of SCT in solution decreases gradually, and the concentration difference between the liquid and the solid phase decreases correspondingly, which leads to the reducing of the adsorption capacity. Therefore, 8 h was chosen as the optimal adsorption time.
3.4. In vitro Release Profiles of SCT and Pore Closing The ideal release profile should include a low initial burst release with a sustained release phase later on. The microspheres morphology surface and other factors may affect the burst release profile28, 29. As shown in Figure 4, mPEG-PLGA microspheres showed a high burst release of 38.7% and complete release only within 8 hrs because SCT adsorbed on the surface or loaded in internal diffused out quickly in first few hours. Similarly, mPEG-PLA microspheres as control showed a burst release of 29.6% and a total of complete release within 10 hrs. mPEG-PLA microspheres showed a lower burst release since the pores were smaller and then the diffusion rate of SCT was slower. However, the complete release time of mPEG-PLA microspheres merely extended to 10 hrs compared to that of mPEG-PLGA microspheres due to the low loading amounts. A large initial release was not beneficial to patients who needed to be injected for long term period. In order to inhibit SCT fast diffusion and improve the release behavior, the pores would be closed by water-miscible solvents that
13
partially dissolve mPEG-PLGA. In this work, NaHCO3 and ethyl acetate were used respectively to compare the pore closing effect. As shown in Figure 5, the pores became small with increasing the concentration of NaHCO 3, however there are still pores could be found on microspheres surface. When the concentration increased to 10%, the microspheres showed rough surface rather than porous morphology. The release profile of pore-closed microspheres by 10% NaHCO3 was shown in Figure 6, the burst release decreased to about 35% and the later release lasted more than 4 days, suggesting that the release profile was improved than before. However, the whole release period was nearly four days, which can’t attend the treatment requirement for at least one week.
Figure 4. Cumulative release of SCT from porous mPEG-PLGA and mPEG-PLA microspheres
14
Figure 5. The surface morphology of microspheres was closed by different concentration of NaHCO3: (a) 0.4%; (b) 0.8%; (c) 2%; (d) 5%; (e) 10%.
Figure 6. Cumulative release of SCT from pore-closed microspheres by NaHCO3 (blue circle) and EA (red triangle) solution, respectively.
15
Figure 7. The surface morphology of microspheres was closed by different concentration of EA: (a) 4%; (b) 5%; (c) 7.5%; (d) 10%.
Therefore, we tried to use another pore closing reagent, ethyl acetate, to investigate its effect of inhibiting burst release. As shown in Figure 7, the surface morphology changed from more pores to less pores with increasing the concentration of ethyl acetate. When the concentration of ethyl acetate was 7.5%, few pores could be found on microspheres surface. While, when the concentration increased to 10%, the microspheres were completely dissolved from sphericity to film. The results revealed that the porous microspheres were closed best after treated by 7.5% ethyl acetate. Meanwhile, the release of pore-closed microspheres was also investigated. Shown in Figure 6, the burst release decreased to about 20% and the later release lasted more than 10 days, suggesting that the release profile of pore-closed microspheres was greatly improved than those of unsealed microspheres. The improved release patterns were mainly due to a diffusion barrier on the microspheres surface and reducing the loss of peptides. Therefore, compared with NaHCO3, ethyl acetate was more effective 16
for mPEG-PLGA microspheres to treat pores and decrease burst release (from 38.7% to 20%) and prolongs the later period release (from 8 h to 10 days). It was certified that using ethyl acetate solution to close pores could improve the release behavior of porous microspheres.
4. Conclusion A novel drug delivery system with porous microspheres followed a pore closing process was designed in this manuscript. mPEG-PLGA porous microspheres were successfully prepared without porogens by using suitable amphiphilic copolymer and solvent with boiling point based on double emulsion-solvent evaporation method. The particle sizes could be controlled by premix membrane emulsification technique. The porous surface morphology was mainly caused by hydrophilic PEG segments absorbing water molecules and local explosion of microspheres. Moreover, the pore sizes and internal structure could be control by adjusting the component of matrix polymer. In addition, porous microspheres containing SCT were treated with ethyl acetate solution can overcome the release problems of porous microspheres and improve release profile. These results suggest that the mPEG-PLGA can be used as a permanent geometric template to tune the porous structure, thus offering an alternative to prepare porous microspheres in the drug delivery formulations. Compared with conventional methods[6,7], porogens were needn’t in this study. Most importantly, the study investigated the key factors influencing preparation and illustrated the forming mechanisms in detail, which will be significant for the development of amphiphilic polymer formulation. The drug delivery system with porous microspheres followed a pore closing process could be potential use in small molecular drug loading, such as peptide, gene, and chemical medicine. We will investigate the effect of pore size and porosity on the loading efficiency of different drug in future work.
Acknowledgements We thank the financial support of National Natural Science Foundation of China 17
(Nos.21306208, 51173187, and 21336010).
References 1
T. Ehtezazi, C. Washington, Journal of Controlled Release 2000, 68, 361-372.
2
X. Niu, Q. Feng, M. Wang, X. Guo, Q. Zheng, Polymer Degradation and Stability 2009, 94,
176-182 http://dx.doi.org/10.1016/j.polymdegradstab.2008.11.008. 3
C. Dai, Y. Wang, X. Hou, Carbohydrate Polymers 2012, 87, 2338-2343
http://dx.doi.org/10.1016/j.carbpol.2011.11.004. 4
L. Sun, S. Zhou, W. Wang, X. Li, J. Wang, J. Weng, Colloids and Surfaces A-physicochemical
and Engineering Aspects 2009, 345, 173-181 10.1016/j.colsurfa.2009.04.053. 5
G. Ma, Journal of Controlled Release 2014, 193, 324-340 10.1016/j.jconrel.2014.09.003.
6
D.-x. Hao, F.-l. Gong, W. Wei, G.-h. Hu, G.-h. Ma, Z.-g. Su, Journal of Colloid and Interface
Science 2008, 323, 52-59 10.1016/j.jcis.2008.03.039. 7
H. K. Kim, H. J. Chung, T. G. Park, Journal of Controlled Release 2006, 112, 167-174
10.1016/j.jconrel.2006.02.004. 8
R. Jeyanthi, B. C. Thanoo, R. C. Metha, P. P. DeLuca, Journal of Controlled Release 1996, 38,
235-244 10.1016/0168-3659(95)00125-5. 9
Y. Wei, Y. Wang, L. Wang, D. Hao, G. Ma, Colloids and Surfaces B-Biointerfaces 2011, 87,
399-408 10.1016/j.colsurfb.2011.05.051. 10
F. Ito, H. Fujimora, H. Honnami, H. Kawakami, K. Kanamura, K. Makino, Colloids and
Surfaces B-biointerfaces 2008, 66, 65-70 10.1016/j.colsurfb.2008.05.011. 11
J. Buske, C. Konig, S. Bassarab, A. Lamprecht, S. Muhlau, K. G. Wagner, European Journal
Of Pharmaceutics and Biopharmaceutics 2012, 81, 57-63 10.1016/j.ejpb.2012.01.009. 12
D. A. Edwards, J. Hanes, G. Caponetti, J. Hrkach, A. BenJebria, M. L. Eskew, J. Mintzes, D.
Deaver, N. Lotan, R. Langer, Science 1997, 276, 1868-1871 10.1126/science.276.5320.1868. 13
P. Tiainen, P.-E. Gustavsson, A. Ljunglof, P.-O. Larsson, Journal Of Chromatography A 2007,
1138, 84-94 10.1016/j.chroma.2006.10.029. 14
M. Nakai, M. Niinomi, D. Ishii, Journal of the Mechanical Behavior of Biomedical Materials
2011, 4, 1206-1218 10.1016/j.jmbbm.2011.04.003. 15
C. T. L. Luz, F. M. B. Coutinho, Polymer 2001, 42, 4931-4938.
16
S. Zhang, X. Huang, N. Yao, C. Horváth, Journal Of Chromatography A 2002, 948, 193-201
http://dx.doi.org/10.1016/S0021-9673(01)01425-X. 17
G. De Rosa, F. Quaglia, M. I. La Rotonda, M. Besnard, E. Fattal, International Journal of
Pharmaceutics 2002, 242, 225-228 10.1016/s0378-5173(02)00162-x. 18
J. Wei, J. Wang, Y. Gong, R. Zeng, Molecular Medicine Reports 2015, 12, 1717-1726
10.3892/mmr.2015.3637. 19
Y. Wei, Y.-X. Wang, W. Wang, S. V. Ho, W. Wei, G.-H. Ma, Journal of Materials Chemistry
2011, 21, 12691-12699 10.1039/c1jm12643a. 20
G. J. S. Dawes, L. E. Fratila-Apachitei, K. Mulia, I. Apachitei, G. J. Witkamp, J. Duszczyk,
Journal of Materials Science-Materials in Medicine 2009, 20, 1089-1094 10.1007/s10856-008-3666-0. 21
Y. Y. Yang, T. S. Chung, X. L. Bai, W. K. Chan, Chemical Engineering Science 2000, 55,
2223-2236 10.1016/s0009-2509(99)00503-5.
18
22
Y. Liu, X. Deng, Journal of Controlled Release 2002, 83, 147-155.
23
X. Shi, L. Sun, Z. Gan, Acta Polymerica Sinica 2011, 866-873 10.3724/sp.j.1105.2011.11113.
24
H. Murakami, M. Kobayashi, H. Takeuchi, Y. Kawashima, Powder Technology 2000, 107,
137-143 10.1016/s0032-5910(99)00182-5. 25
M. Cho, H. Sah, Journal of Microencapsulation 2005, 22, 1-12 10.1080/02652040400026269.
26
A. Rawat, D. J. Burgess, International Journal of Pharmaceutics 2010, 394, 99-105
10.1016/j.ijpharm.2010.05.013. 27
G. Ruan, S. S. Feng, Q. T. Li, Journal of Controlled Release 2002, 84, 151-160.
28
M. van de Weert, W. Hennink, W. Jiskoot, Pharmaceutical Research 2000, 17, 1159-1167.
29
M. A. Ibrahim, A. Ismail, M. I. Fetouh, A. Gopferich, Journal of Controlled Release 2005,
106, 241-252 10.1016/j.jconrel.2005.02.025.
19
Graphical Abstract