Hyaluronic Acid Complex Microspheres ... - MDPI

1 downloads 0 Views 3MB Size Report
Nov 17, 2018 - growth factor-β1 (TGF-β1) can act in both capacities with mitosis and ... CH powder was procured from Aladdin Inc (Shanghai, China).
pharmaceutics Article

Chitosan-Polylactide/Hyaluronic Acid Complex Microspheres as Carriers for Controlled Release of Bioactive Transforming Growth Factor-β1 Qing Min 1,† , Jiaoyan Liu 2,† , Jing Li 3 , Ying Wan 2, * and Jiliang Wu 1, * 1 2 3

* †

School of pharmacy, Hubei University of Science and Technology, Xianning 437100, China; [email protected] College of Life Science and Technology, Huazhong University of Science and Technology, Wuhan 430074, China; [email protected] Hubei Province Key Laboratory on Cardiovascular, Cerebrovascular and Metabolic Disorders, Hubei University of Science and Technology, Xianning 437100, China; [email protected] Correspondence: [email protected] (Y.W.); [email protected] (J.W.); Tel.: +86-027-87792137 (Y.W.); +86-0715-8130584 (J.W.) These authors contributed equally to this work.

Received: 3 October 2018; Accepted: 14 November 2018; Published: 17 November 2018

 

Abstract: Chitosan(CH)-polylactide(PLA) copolymers containing varied PLA percentages were synthesized using a group-protection method and one of them with solubility in water-based solvents was used to prepare CH-PLA/hyaluronic acid (HA) complex microspheres for the delivery of transforming growth factor-β1 (TGF-β1). An emulsification processing method was developed for producing TGF-β1-loaded CH-PLA/HA microspheres using sodium tripolyphosphate (TPP) as ionic crosslinker and the size of the microspheres was devised to the micron level in order to achieve high encapsulating efficiency. The encapsulating efficiency, swelling property and release administration of the microspheres could be synergistically regulated by PLA component, the applied TPP dose and the incorporated HA amount. In comparison to CH/HA microspheres, the CH-PLA/HA microspheres had greatly reduced TGF-β1 release rates and were able to administrate the TGF-β1 release at controlled rates over a significant longer period of time. The released TGF-β1 was detected to be bioactive when compared to the free TGF-β1. These results suggest that the presently developed CH-PLA/HA complex microspheres have promising potential in delivering TGF-β1 for cartilage repair applications where the applied TGF-β1 amount in the early stage needs to be low whilst the sustained TGF-β1 release at an appropriate dose in the later stage has to be maintained Keywords: chitosan; polylactide; hyaluronic acid; microspheres; delivery of bioactive molecules; transforming growth factor-β1

1. Introduction Cartilaginous tissues play important roles in contributing certain crucial functions such as very low wear resistance, load bearing and energy dissipation in joints of the musculoskeletal system. These tissues have a limited capacity for self-repair after injury owing to its avascular and alymphatic nature as well as low cell density [1]. Currently used clinical therapies for cartilage repair mainly include microfracture, subchondral bone drilling and mosaicplasty and autologous or allologous implantation. However, these techniques have their respective shortcomings and in particular, many cases involving the mentioned surgical interventions have not proven to be successful from a long-term point of view [2,3].

Pharmaceutics 2018, 10, 239; doi:10.3390/pharmaceutics10040239

www.mdpi.com/journal/pharmaceutics

Pharmaceutics 2018, 10, 239

2 of 14

Tissue-engineering strategies have become attractive for their potential to repair cartilage lesions by using degradable scaffolds, exogenous bioactive molecules, and/or cells [2,4]. To date, a wide variety of scaffolds in the form of sponges, films, fibers and hydrogels have been developed [5]. In addition to the important role of the scaffold, many studies have suggested that the local presence of appropriate bioactive molecules at the requisite concentrations is of particular importance to the cases of cartilage repair since these molecules are able to induce the recruitment and proliferation of the cells that migrate from the synovial membrane and subsynovial space and to transform the recruited cells into chondrocytes [6–8]. Among the bioactive molecules applicable for cartilage repair, transforming growth factor-β1 (TGF-β1) can act in both capacities with mitosis and chemotaxis, being regulated by the factor concentration and the factor action time period [7,8]. In general, ectogenic TGF-β1 is not systematically administered in vivo because it is short-lived when exposed to a physiological environment. Proper carriers are usually required to protect TGF-β1 from proteolysis or antibody neutralization and to locally deliver it at effective concentrations and controlled rates [1,6,8,9]. Synthetic polyesters with biodegradable nature were commonly employed as vehicle materials to build carriers for delivery of protein growth factors [10]. Nevertheless, their high hydrophobicity, lack of functional groups and slow degradation rate limit their capability for the factor release administration [7,10,11]. In addition, polyester-based vehicles are generally constructed by means of organic solvents, which could possibly attenuate the bioactivity of the loaded factors [7,10]. Natural polymers, including collagen, gelatin, fibrin, silk fibroin, hyaluronic acid, alginate and chitosan (CH), have been investigated as alternative carrier materials for the delivery of growth factors [6,7,10]. Among these mentioned natural polymers, CH is a widely used one as it has many unique properties such as biodegradability, hydrophilicity, antimicrobial activity, bioadherence and cell affinity [12–14]. CH microspheres crosslinked by certain ionic crosslinkers (e.g., sulfate, citrate and sodium tripolyphosphate) have been used as carriers for delivery of TGF-β1 [10,13,15]. Since the mentioned anionic crosslinkers are nontoxic and the linkages among CH molecular chains are formed mainly basing on electrostatic interactions, the bioactivity of TGF-β1 released from CH microspheres can thus be well preserved [10,13]. Nevertheless, such constructed CH microspheres showed severe initial burst release features and had a limited ability to control the release of TGF-β1 at a proper level over a long enough period of time [10]. In recent years, polyelectrolyte composites have emerged as effective delivery systems and many new processing techniques have been developed for constructing polyelectrolyte composite carriers [16]. Polyelectrolyte complexes composed of positively charged CH and negatively charged natural polymers such as alginate, hyaluronic acid (HA) and pectin have aroused considerable interest in the field of drug delivery [17]. Such types of polyelectrolyte complexes are usually non-toxic and biocompatible because the interactions between polyanionic molecules and polycationic molecules majorly involve electrostatic association and/or hydrogen and hydrophobic bonds. Among the enumerated polyanionic polymers, HA is often used together with CH for constructing CH/HA polyelectrolyte complexes [17]. HA is a liner glycosaminoglycan composed of N-acetyl-D glucosamine and D-glucuronic acid units and it exists in many types of extracellular matrixes (ECMs) in human body as an important component for facilitating cell locomotion, proliferation and phenotype preservation [18,19]. HA shows excellent potential for application in drug delivery, wound healing and tissue engineering due to its good biocompatibility, biodegradability and mucoadhesion [15,17–19]. CH/HA complexes were also used as gene vectors and the optimal complex showed significantly higher transfection efficiency for different targeting cells as compared to the vectors comprised of CH only [17,20–22]. Several studies on CH/HA complex nanoparticles point to their utilization in delivery systems [17]. In this study, an attempt was made to develop a new type of CH-based complex microspheres for TGF-β1 delivery by using amphiphilic chitosan-polylactide (CH-PLA) copolymers and HA. CH-PLAs were synthesized by grafting PLA side chains onto the C-6 sites of CH backbone while leaving the amino groups at the C-2 sites free using a group-protection method. The optimal

Pharmaceutics 2018, 10, 239

3 of 14

CH-PLA with solubility in waster-based solvents was combined with HA to construct TGF-β1-loaded CH-PLA/HA complex microspheres in the presence of sodium tripolyphosphate (TPP) as crosslinker. Since CH-PLAs contain hydrophobic PLA side chains but their amino groups are free, CH-PLA/HA complexes could still be formed via the interactions between the amino groups in CH backbone and the carboxyl groups in HA. In addition, an emulsification preparation method was developed for producing TGF-β1-loaded CH-PLA/HA microspheres with their size at micron level in order to achieve high encapsulating efficiency. So produced CH-PLA/HA microspheres were found to show significantly improved capacity for reducing the initial fast release of TGF-β1 and for prolonging the TGF-β1 release at controlled rates in comparison to CH/HA microspheres, making them suitable for the use in cartilage repair [16–18]. Some results related to the preparation and characterization of the microspheres and to the TGF-β1 release patterns as well as its bioactivity preservation were reported. 2. Materials and Methods 2.1. Materials CH powder was procured from Aladdin Inc (Shanghai, China). Deacetylation degree and viscosity average molecular weight of CH were measured as 94.2 (±1.7)% and 1.12 (±0.14) × 105 , according to the reported methods [23]. Human recombinant TGF-β1 and TGF-β1 Quantikine ELISA Kit were supplied by PeproTech Inc (Rocky Hill, NJ, USA) and R&D Systems (Minneapolis, MN, USA), respectively. HA (sodium salt, Mw :90–110 kDa), lactide (LA) and polyvinyl alcohol (PVA, 87–89% hydrolyzed, Mw :13–23 kDa) were purchased from Sigma-Aldrich (Shanghai, China). XTT tetrazolium assay kit was purchased from Trevigen (Gaithersburg, MD, USA). All other reagents and chemicals were of analytical grade and purchased from Sinopharm, Shanghai, China. 2.2. Synthesis of CH-PLA Copolymers CH-PLA copolymers were synthesized using a group-protection method [24,25]. In brief, the sifted CH powder (106–150 µm; 1.5 g) and phthalic anhydride (4.1 g) were dispersed in 30 mL of dried DMF and the mixture was heated at 90 ◦ C for 8 h with stirring to produce phthaloyl-chitosan (PHCH). After reaction, the product was washed with anhydrous ethanol and dried to constant weight. In a typical procedure for synthesizing CH-PLA, LA (2.23 g) and PHCH (1.0 g) were added to a flask dried at reduced pressure for 1 h prior to the addition of toluene (10 mL). The reaction was carried out at 100 ◦ C in nitrogen atmosphere for varied durations up to 60 h. After that, the flask was cooled down to ambient temperature and the obtained mixture was washed with 20 mL of acetone to remove the unreacted LA monomer and the PLA homopolymer. The collected product was then extracted with acetone for at least 12 h to obtain phthaloyl-chitosan-polylactide (PHCH-PLA). The obtained PHCH-PLA was deprotected using hydrazine monohydrate to produce chitosan-polylactide (CH-PLA). By mainly changing the feed ratio of LA to PHCH, CH-PLAs with various PLA percentages were synthesized. To ensure potential solubility of CH-PLAs in aqueous media, CH-PLA with a PLA content of about 38 wt % was used to prepare CH-PLA/HA complex microspheres. 2.3. Preparation of Microspheres An emulsification method was used to prepare microspheres using TPP as a crosslinker. The preparation method is briefly described as follows. The selected CH-PLA was dissolved in 1% acetic acid aqueous solution to prepare 1 wt % CH-PLA solutions. To each solution, a given amount of TGF-β1 in 4 mM HCl was introduced, followed by adding PVA to reach a PVA concentration of 0.5 wt %. HA aqueous solutions with different concentrations (0.15 and 0.3 wt %) were also prepared. Each HA solution was added with PVA and TPP so that the concentration of PVA was 0.5 wt % and the ratio of TPP to the matrix of intended complex microspheres was 2 and 4, respectively. After that, a CH-PLA composite solution and one of HA composite solutions were isometrically introduced to a vessel containing mineral oil to reach a volume ratio of 7 (oil phase):1(aqueous phase). The mixture

Pharmaceutics 2018, 10, 239

4 of 14

was homogenized at 8000 rpm for 30 min. The resulting emulsion was centrifuged at 7000 rpm for 10 min to separate the aqueous phase and the oil phase and CH-PLA/HA microspheres were collected by further centrifugation. The microspheres were washed with petroleum ether, alcohol and water in order, followed by lyophilization. CH microspheres and CH/HA microspheres were also produced with the same method and they were fed with the same amount of TGF-β1 during their preparation. These CH and CH/HA microspheres were used as controls. 2.4. Characterization 1H

nuclear magnetic resonance (NMR) spectra were recorded on a spectrometer (Bruker AV 500, Rheinstetten, Germany) using a mixed solvent of D2 O and CF3 COOD (95:5, v/v). Elemental analyses were performed to determine the PLA content in the CH-PLAs using an elemental analyzer (Vario EL III, Elementar, Hanau, Germany). Microspheres were sputter-coated with gold and their morphology was viewed with a scanning electron microscope (SEM, Quanta 200, FEI, Eindhoven, the Netherlands). Fourier transform infrared (FTIR) spectra of CH-PLA samples were recorded on a spectrometer (VERTEX 70, Bruker, Ettlingen, Germany). For each kind of microspheres, their size-distribution was determined by measuring the diameters of 200 microspheres in their SEM images in a random manner and their average size was thus calculated. 2.5. Determination of TGF-β1 Encapsulating Efficiency Microspheres were extracted using a 4 mM HCl solution with shaking (60 rpm) at 37 ◦ C for 24 h. The collected supernatants were assayed to determine the TGF-β1 content in the microspheres by using a TGF-β1 Quantikine ELISA Kit and following the instructions provided by the manufacture. Encapsulating efficiency (EE) was calculated using following formula: EE (%) = (Mm /Mf ) × 100%

(1)

where Mm is the measured amount of TGF-β1 in the microspheres and Mf , the feed amount of TGF-β1 during microsphere preparation. 2.6. Swelling Index Weighed dry microspheres (Wd ) were immersed in a phosphate buffer saline (PBS) solution at 37 ◦ C for 5 h. After that, they were transferred into some glass tubes having a sintered glass bottom and excess water was removed by centrifugation at 2000 rpm for 1 min. Weight (Ws ) of swollen microspheres was measured and their swelling index (SI) was calculated as follows: SI (%) = [(Ws − Wd )/Wd ] × 100%

(2)

2.7. In Vitro Release of TGF-β1 Release studies were conducted in PBS (pH 7.4). In brief, 10 mg of microspheres was introduced into a microcentrifuge tube filled with 0.25 mL of PBS (pH 7.4) and the tube was maintained at 37 ◦ C. At predetermined time points, the tube was centrifuged to collect supernatant, replenished with fresh PBS and followed by vortex for 1 min. The released TGF-β1 amount in the supernatant was measured using the above mentioned TGF-β1 essay kit. All microsphere samples were tested with the same method and the average values reported were based on three specimens for each sample. 2.8. Activity Assessment of Released TGF-β1 Activity of TGF-β1 released from four kinds of microspheres (CH/HA-I(2), CH/HA-II(b) CH-PLA/HA-I(2) and CH-PLA/HA-II(b)) was assessed using the Mink lung cell growth inhibition assay [26]. Briefly, 10 mg of microspheres was placed in dulbecco’s modified eagle medium (DMEM), 250 µL) and kept at 37 ◦ C. At each time point (day 1, 3, 7, 10, 14, 17 and 21), the supernatant was

Pharmaceutics 2018, 10, 239

5 of 14

collected by centrifugation. The microsphere sample was refreshed with the same volume of DMEM and vortexed for 1 min. Supernatants matching with four kinds of microspheres were collected usingPharmaceutics the same 2018, method. triplicate 10, x Mink lung cells in DMEM were plated in 96-well culture dishes in 5 of 14 at a density of 5 × 103 cells/well, supplementing with 10% fetal calf serum. After attachment, volume of DMEM and vortexed for 1 min. Supernatants matching with four kinds of microspheres the supernatants collected from different microsphere groups were added to the wells, respectively, were collected using the same method. Mink lung cells in DMEM were plated in 96-well culture and cells were further cultured for 48 h.3 The control wells were added with the same amount of dishes in triplicate at a density of 5 × 10 cells/well, supplementing with 10% fetal calf serum. After free TGF-β1. Celltheviability was assessed withdifferent the colorimetric XTT tetrazolium assay following attachment, supernatants collected from microsphere groups were added to the wells, the manufacturer’s and usingcultured a standard TGF-β1 the range of respectively,instructions and cells were further for 48curve h. Thewith control wells concentrations were added withinthe same 0.01–1amount ng/mL. of cell was expressed as percentages related to the cells of The free level TGF-β1. Cell growth viability inhibition was assessed with the colorimetric XTT tetrazolium assay following thetreatment. manufacturer’s instructions and using a standard curve with was TGF-β1 concentrations without TGF-β1 In addition, a certain amount of free TGF-β1 stored in DMEM and in37 the◦ Crange of 0.01–1 ng/mL.upThe level of cell growth as percentages kept at for various periods to 21 days. It was usedinhibition as controlwas for expressed making activity comparisons related to the cells without TGF-β1 treatment. In addition, a certain amount of free TGF-β1 was between such treated free TGF-β1 and the released TGF-β1. stored in DMEM and kept at 37 °C for various periods up to 21 days. It was used as control for making activity comparisons between such treated free TGF-β1 and the released TGF-β1. 2.9. Statistical Analysis

Data were presented 2.9. Statistical Analysis as mean ± standard deviation. One-way analysis of variance was conducted and the level of statistical significance is defined as p < 0.05. Data were presented as mean ± standard deviation. One-way analysis of variance was conducted and the level of statistical significance is defined as p < 0.05.

3. Results and Discussions

3. Results and Discussions

3.1. CH-PLA Characterization

3.1.this CH-PLA Characterization In study, PLA side chains were selectively grafted onto the C-6 sites of the CH backbone thisC-2 study, side chains were selectively onto the C-6 sites ofinthe CH to backbone rather thanInthe sitesPLA using phthalic anhydride as a grafted group protective reagent order leave amino rather than C-2 sites phthalic anhydride as microsphere a group protective reagent and in order leave groups at the C-2the site free forusing the subsequent complex preparation TPP to crosslinking. amino groups at the C-2 site free for the subsequent complex microsphere preparation and TPP A schematic illustration for synthesis of CH-PLA is presented in Figure 1A. crosslinking. A schematic illustration for synthesis of CH-PLA is presented in Figure 1A. Figure 1B presents two FTIR spectra for CH and CH-PLA. A shoulder-like band at round Figure 1B presents two FTIR spectra−for CH and CH-PLA. A shoulder-like band at round 1655 1 are 1655 cm−1−1 and another band at 1598 cm characteristic bands for CH with a relatively high cm and another band at 1598 cm−1 are characteristic bands for CH with a relatively high deacetylation degree and they correspond to primary amine(N–H (N–H bending) secondary deacetylation degree and they correspond to primary amine bending) andand secondary amineamine (C=O(C=O stretching) of CH, respectively [27]. In the case of the spectrum for CH-PLA, two new stretching) of CH, respectively [27]. In the case of the spectrum for CH-PLA, two new bandsbands −1−1and they could be assigned to the carboxylic ester in PLA side are recorded at 1756 andand 1249 cmcm are recorded at 1756 1249 and they could be assigned to the carboxylic ester in PLA side 1 ) –1 [25,28]. bands ) that originally belongtotothe theamide amideI and I andamide chainschains [25,28]. Two Two otherother bands (1653(1653 and and 15961596 cm−cm that originally belong amide of CHregistered are clearlyfor registered forsuggesting CH-PLA, suggesting that chains PLA side chains been II of CH are IIclearly CH-PLA, that PLA side have beenhave conjugated to conjugated to the C-6 sites of CH chain units instead of their C-2 sites. the C-6 sites of CH chain units instead of their C-2 sites.

1. Schematic illustration (A)for forsynthesis synthesis of spectra (B) (B) of CH and CH-PLA; FigureFigure 1. Schematic illustration (A) of CH-PLA; CH-PLA;FTIR FTIR spectra of CH and CH-PLA; 1H NMR spectra of CH (C) and CH-PLA (D) (PLA content in CH-PLA: 40.7 wt %). and and 1 H NMR spectra of CH (C) and CH-PLA (D) (PLA content in CH-PLA: 40.7 wt %).

Pharmaceutics 2018, 10, 239

6 of 14

Figure 1C, D show 1 H NMR spectra of CH and CH-PLA, respectively. Shifts recorded at about 2.59 (H-2), 3.2–3.5 (H-3, -4, -5, -6) and 4.8 (H-1) ppm are typically attributed to CH [25]. With respect to CH-PLA, a new signal appeared at ca. 1.2 ppm can be ascribed to the methyl protons located at the terminal groups and the backbone of the polylactide moiety. Another new singlet signal at about 4.1 ppm belongs to the protons in the repeat units of PLA chains [24,25]. CH-PLA has the characteristic signals for the CH component with small shifts in its spectrum and the signal area ratio of H-2 to H-1 is detected to be nearly the same as that for CH, corroborating that the PLA side chains are grafted onto the C-6 sites of CH backbone. On the basis of presented FTIR and 1 H NMR spectra, it can be reached that so synthesized CH-PLA possesses the designed structures. Several kinds of CH-PLAs with various PLA percentages were synthesized by mainly changing the ratio of LA to PHCH, reaction time and the volume of reaction media while keeping other reaction conditions constant and their relevant parameters are listed in Table 1. It is known that PLA is hydrophobic polyester with notably slower degradation rate as compared to CH [24,29]. When certain CH-PLAs with soluble nature in aqueous media are used to produce CH-PLA/HA complex microspheres for delivery of drugs or active molecules, the PLA side chains in the CH-PLAs would facilitate to drag the loaded drugs or molecules and thereby, to reduce their release rates, implying that the higher PLA content in the CH-PLAs will be potentially advantageous for improving the release administration of CH-PLA/HA microspheres. Based on a number of preparatory trials, it was found that CH-PLAs with a PLA percentage higher than 80 wt % could be achieved under the present synthesis conditions. Nevertheless, data in Table 1 exhibit that the CH-PLA with a PLA content of around 46 wt % is insoluble in a 1% acetic acid aqueous solution, which means that this kind of CH-PLA is unsuitable for the subsequent preparation of CH-PLA/HA complex microspheres. In order to ensure the full solubility of CH-PLAs in aqueous media while taking advantage of PLA component in CH-PLAs, CH-PLA (2) was selected for the following microsphere preparation. Table 1. Parameters of CH-PLA copolymers (n = 3). Copolymer Name

Feed Ratio of LA to PHCH (Molar Ratio) (a)

PLA percentage in CH-PLA (wt %) (b)

CH-PLA(1) CH-PLA(2) CH-PLA(3) CH-PLA(4)

2/1 4/1 6/1 8/1

23.6 (±1.51) 38.7 (±1.46) 46.1 (±1.64) 52.4 (±1.71)

Solubility (c) DMSO Acetic Acid (1.0%)

− ± ±± ±±

+ + ±± ±±

(a)

Ratio of LA to glucosamine units in PHCH (phthaloyl-chitosan); (b) Contents of C, H and N in the CH-PCLs were measured using an elemental analyzer and the PLA content in CH-PLAs was estimated using the C/N ratio; (c) “−”, “±”, “±±” and “+” denote that CH-PLAs are insoluble, swelled, partially soluble or highly swelled and soluble, respectively, and DMSO refers to dimethyl sulfoxide.

3.2. Parameters of Microspheres CH is a cationic polymer and it is able to form ionic complexes with certain anionic substances [16]. In the light of anionic features of HA, CH/HA nanoparticles can be produced in aqueous media via blending method while altering the component proportions and the concentrations of component solutions [15–17,30]. Despite the various applications of CH-based complex nanoparticles, they generally have low EE when used for delivering hydrophilic drugs or biomolecules [14,15,30]. Considering that the presently devised microspheres will be used for cartilage repair by injection and their EE needs to be high for saving expensive TGF-β1, their size can thus be designed to the micron level instead of the nano level. In view of structure and property of CH-PLA copolymers, the technique used for the preparation of CH/HA nanoparticles seems to be applicable for the preparation of CH-PLA/HA complex microspheres since CH-PLA has the similarity to CH in the presence of free amino groups at the C-2 sites of CH backbone. Disappointedly, it was found that simply blending a CH-PLA solution with a HA solution was unworkable for preparing CH-PLA/HA microspheres that have controlled morphologies and sizes even with the aid of different crosslinkers because CH-PLA

Pharmaceutics 2018, 10, 239

7 of 14

and HA would fast form into quite irregular agglomerates when being blended in aqueous media. An emulsification approach was therefore developed for the preparation of CH-PLA/HA microsphere by using PVA as an emulsifier. In general, multiple factors can influence the size, morphology, structure and property of the resulting microspheres when an emulsification technique is employed [16,31]. In the present instance, a number of preliminary experiments were conducted to optimize several major processing parameters, including concentration of solutions, stirring speed, the amount of the applied emulsifier and the volume ratio of 10, dispersed phase and continuous phase. Therefrom, two sets of TGF-β1-loaded Pharmaceutics 2018, x 9 of 19 Pharmaceutics 2018, 10, prepared x 9 of 19 microspheres were and relevant parameters for them are summarized in Table 2. () Table 2. Parameters of TGF-β1-loaded microspheres (n = 4) . Table2.2.Parameters ParametersofofTGF-β1-loaded TGF-β1-loadedmicrospheres microspheres Table (n(n = =4)4)(†)() .. Feed Ratio of Feed Ratio of Microsphere Average Feed of FeedRatio Feed of of Feed Ratio of HA to Ratio CH-PLA TPP toRatio Matrix EE(%) SI(%) Microsphere Average Average Size Name Name HA Size (µm) EEEE(%) (%) SI SI(%) (%) TPP toto Matrix Microsphere HA to to CH-PLA CH-PLA TPP Matrix (mg/mg) (mg/mg) Pharmaceutics 2018, 10, x Pharmaceutics 2018, 10, x 9 of 19 (µm)(µm) Name Size (mg/mg) (mg/mg) (mg/mg) (mg/mg) CH-I  − 2/1 3.73 (±0.41) 29.6 (±3.4) 69.7 (±5.1)  CH-I − 2/1 3.73 ((n ±(±0.35) 0.41) 29.6 (±(±2.7) 3.4) 69.7 (±(±4.6) 5.1) − 2/1microspheres 3.73 (±0.41) 29.6 (±3.4) 69.7 (±5.1) CH-I Table 2. Parameters Table 2. Parameters = 4) (). of TGFβ1-loaded microspheres (n = 4) (). CH/HA-I(1) 0.15/1.0of TGF-β1-loaded 2/1 3.72 35.5 61.5 CH/HA-I(1) 0.15/1.0 2/1 3.72 ( ± 0.35) 35.5 ( ± 2.7) 61.5 ( ± 4.6) CH/HA-I(1) 2/1 3.72 (±0.32) (±0.35) 40.7 35.5 (±2.9) (±2.7) 61.5 (±3.9) (±4.6) CH/HA-I(2) 0.3/1.0 2/1 3.74 59.1 of Feed Ratio2/1 of Feed Ratio of Feed Ratio of CH/HA-I(2) Feed Ratio0.3/1.0 0.3/1.0 3.74 (±(±0.32) 0.32) 40.7 (±(±2.9) 2.9) 59.1 (±3.9) CH/HA-I(2) 2/1 3.74 40.7 59.1 (±3.9) Microsphere Microsphere Average Average CH-PLA/HA-I(1) 0.15/1.0 2/1 55.2 46.4 CH-PLA/HA-I(1) 0.15/1.0 TPP to Matrix 2/1 (±(±0.29) 0.29) EE(%) 55.2 (±(±2.8) 2.8)Matrix 46.4 (±(±3.5) 3.5) HA to CH-PLA HA3.86 to3.86 CH-PLA TPP to SI(%) EE(%) CH-PLA/HA-I(1) 0.15/1.0 2/1 3.86 (±0.29) 55.2 (±2.8) 46.4 (±3.5) Name Name Size (µm) Size (µm) Pharmaceutics 2018, 10, x Pharmaceutics 2018, 10, x 9 of 19 9 CH-PLA/HA-I(2) 0.3/1.0 2/1 4.08 (±0.25) 61.4 (±2.5) 44.6 (±3.1) CH-PLA/HA-I(2) (mg/mg)0.3/1.0 2/1 4.08 ( ± 0.25) 61.4 ( ± 2.5) 44.6 ( ± 3.1) (mg/mg) 2/1 (mg/mg) (mg/mg) CH-PLA/HA-I(2) 0.3/1.0 4.08 (±0.25) 61.4 (±2.5) 44.6 (±3.1)  − 4/1 3.16 (±(±0.27) 0.27) 37.1 (±(±3.3) 3.3) 61.2 (±(±4.8) 4.8) CH-II − 4/1 3.16 37.1 61.2 CH-II CH-I  CH-II CH-I − 2/1  4/1 3.73 (±0.41) − (±0.27) 29.6 (±3.4) 2/1 69.7 (±5.1) 3.73 (±0.41) 29.6 (±3.4  () − of TGF-β 3.16 37.1 (±3.3) 61.2 (±4.8) Table 1-loaded Table 2.microspheres Parameters of(n 4) β 1-loaded . (n 4) (). CH/HA-II(a) 0.15/1.0 4/1 3.46 (TGF±=(±0.31) 0.31) 46.8 (microspheres ±(±3.1) 3.1) 53.4 (±=(±4.3) 4.3) CH/HA-II(a) CH/ 2. Parameters 0.15/1.0 4/1 3.46 46.8 53.4 CH/HA-I(1) CH/HA-I(1) 2/1 4/1 0.15/1.0 35.5 (±2.7) 2/1 61.5 (±4.6) (±0.35) 35.5 (±2.7 CH/HA-II(b) 0.3/1.0 3.63 (±(±0.28) 0.28) 51.6 (±(±3.6) 3.6) 48.5 (±3.72 3.6) CH/HA-II(a) CH/ 0.15/1.00.15/1.0 4/1 3.72 (±0.35) 3.46 (±0.31) 46.8 (±3.1) 53.4 (±4.3) HA-II(b) CH0.3/1.0 4/1 3.63 51.6 48.5 (±3.6) Feed0.3/1.0 Ratio 0.15/1.0 of Feed Ratio Feed of Ratio of (±0.32) Feed Ratio of CH/HA-I(2) CH/HA-I(2) 2/1 3.74 0.3/1.0 40.7 (±2.9) 2/1 59.1 (±3.9) 3.74 (±0.32) 40.7 (±2.9 CH-PLA/HA-II(a) 4/1 3.83 ( ± 0.24) 75.3 ( ± 3.2) 32.3 ( ± 3.4) HA-II(b) CH0.3/1.0 4/1 Average 3.63 (±0.24) (±0.28) 75.3 51.6 (±3.6) 48.5 (±3.4) (±3.6) Microsphere Microsphere Average PLA/HA-II(a) 0.15/1.0 4/1 3.83 32.3 HA to0.15/1.0 CH-PLA TPP to2/1 Matrix HA4/1 to CH-PLA to EE(%) Matrix SI(%) CH-PLA/HA-II(b) 0.3/1.0 CH-PLA/HA-I(1) 4.16 (TPP ±0.28) 81.9 (±(±3.2) 3.8) 30.2 (±3.86 3.2) EE(%) CH-PLA/HA-I(1) 3.86 (±0.29) 0.15/1.0 55.2 (±2.8) 2/1 46.4 (±3.5) (±0.29) 55.2SI(%) (±2.8 PLA/HA-II(a) 0.15/1.0 4/1 Size (µm) 3.83 (±0.28) (±0.24) 81.9 75.3 (±3.8) (±3.2) 32.3 (±3.2) (±3.4) Name Name Size (µm) CH-PLA/HA-II(b) 0.3/1.0 4/1 4.16 30.2 (†) CH-PLA with a (mg/mg) (mg/mg) (mg/mg) (mg/mg) PLA percentage ofCH-PLA/HA-I(2) 38.7 wt 2/1 % (see Table 1)4.08 used for preparing all kinds2/1 of44.6 microspheres; CH-PLA/HA-I(2) 0.3/1.0 (±0.25) 0.3/1.0 61.4 (±2.5) (±3.1) (±0.25) 61.4 (±2.5 CH-PLA/HA-II(b) 0.3/1.0 4/1 was 4.16 (±0.28) 81.9 (±3.8) 30.24.08 (±3.2) () with a PLA CH-I percentage 38.7 (wt Table 1) was 29.6 used for preparing kinds of29.6 (±3.4) ( ,  ) CH-PLA (  ) and of CH-I 2/1 3.73 (±0.41) 2/1(±3.4) 69.7 3.73 (±5.1) (±0.41) 69.7(±3.3 (±5 CH-II CH-II − set-one 4/1  ) . % −(see 3.16 (±0.27) − 37.1 (±3.3) 4/1 61.2all (±4.8) 3.16 (±0.27) 37.1 Different sample − sets: set-two () CH-PLA with a PLA percentage 38.7 wt % () (seeand Table 1) was used for preparing all kinds of microspheres; (,) Different sample of sets: set-one set-two (). CH/HA-I(1) CH/HA-I(1) 0.15/1.0 2/1 0.15/1.0 3.72 (±0.35) 35.5 2/1(±2.7) 61.5 3.72 (±4.6) (±0.35) 35.5 (±2.7) 46.8 61.5(±3.1 (±4 CH/HA-II(a) CH/ CH/HA-II(a) CH/ 0.15/1.0 4/1 3.46 (±0.31) 0.15/1.0 46.8 (±3.1) 4/1 53.4 (±4.3) 3.46 (±0.31) microspheres; (,) Different sample sets: set-one () and set-two (). CH/HA-I(2) CH/HA-I(2) 0.3/1.0 2/1 0.3/1.0 3.74 (±0.32) 40.7 2/1 (±2.9) 59.1 3.74 (±3.9) (±0.32) 40.7 (±2.9) 59.1 (±3 HA-II(b) CH-now, varied HA-II(b) 0.3/1.0 4/1 CH3.63 (±0.28) 51.6 (±3.6)diisocyanate, 4/1 48.5 (±3.6)genipin, 3.63 (±0.28) 51.6 (±3.6 Up kinds of covalent crosslinkers, such as0.3/1.0 glutaraldehyde, Uptoto now, varied kinds of covalent crosslinkers, such as glutaraldehyde, diisocyanate, CH-PLA/HA-I(1) CH-PLA/HA-I(1) 0.15/1.0 2/1 0.15/1.0 3.86 55.2 2/1 (±2.8) 46.4 3.86 (±0.29) 55.2 (±2.8) 75.3 46.4(±3.2 (±3 PLA/HA-II(a) PLA/HA-II(a) 0.15/1.0 4/1 crosslinkers, 3.83(±0.29) (±0.24) 0.15/1.0 75.3 (±3.2) 4/1 32.3(±3.5) (±3.4) 3.83 (±0.24) ethylene diglycidyl ether andofglycidoxypropyltrimethoxy silane, have been utilized to crosslink Upglycol to now, varied kinds covalent such as glutaraldehyde, diisocyanate, genipin, ethylene glycol diglycidyl ether and glycidoxypropyltrimethoxy silane, have been utilized CH-PLA/HA-I(2) CH-PLA/HA-I(2) 0.3/1.0 2/1 0.3/1.0 4.08 (±0.25) 61.4 2/1 (±2.5) 44.6 4.08 (±3.1) (±0.25) 61.4 (±2.5) 44.6 (±3 CH-PLA/HA-II(b) CH-PLA/HA-II(b) 0.3/1.0 4.16for (±0.28) 0.3/1.0 81.9 preservation (±3.8) 4/1 30.2of (±3.2) 4.16 (±0.28) 81.9 (±3.8 CH [32–34]. Taking into consideration the4/1 requirements the activity the loaded genipin, ethylene glycol diglycidyl ether and glycidoxypropyltrimethoxy silane, have been utilized Taking into consideration the requirements for the activity preservation of to crosslink CH [32–34].   CH-II CH-IIofinto − microspheres, − the 3.16 (±0.27) 4/1 (±3.3) 3.16 (±4.8) (±0.27) (±3.3)for preparing 61.2 (±4 () aresulting PLA percentage 38.7() wt4/1 CH-PLA %this (see study, Table with 1) a requirements PLA was used percentage for37.1 preparing of 38.7activity wt all 61.2 % kinds (see of Table 1) 37.1 was TGF-β1 in with the in ionic crosslinker, selected forofused Taking for the preservation toCH-PLA crosslink CH [32–34]. the loaded TGF-β1 in the resulting consideration microspheres, inTPP, this an study, TPP, an ionicwas crosslinker, was CH/HA-II(a) CH/ CH/HA-II(a) CH/ 0.15/1.0 4/1 0.15/1.0 3.46 (±0.31) 46.8 4/1 (±3.1) 53.4 3.46 (±4.3) (±0.31) 46.8 (±3.1) 53.4 (±4 microspheres; (,) Different sample sets: set-one microspheres; () and (,) set-two Different (). sample sets: set-one () and set-two (). crosslinking microspheres because it can in crosslink CH incrosslinker, acidic aqueous the loaded TGF-β1 in the resulting microspheres, this study, TPP, an ionic selected forCH-PLA/HA crosslinking CH-PLA/HA microspheres because itthe canprotonated crosslink the protonated CHwas in HA-II(b) CHHA-II(b) CH0.3/1.0 4/1 0.3/1.0 3.63 (±0.28) 51.6 4/1 (±3.6) 48.5 3.63 (±3.6) (±0.28) 51.6 (±3.6) 48.5 (±3 media via ionic interactions and also,interactions is a safe and and biocompatible for bioactive molecules [31,32]. selected for crosslinking CH-PLA/HA microspheres because itsafe can and crosslink the protonated CHfor in acidic aqueous media via ionic also, is areagent biocompatible reagent Up to varied kinds covalent Upcrosslinkers, toHA now, varied such asaCH-PLA/HA glutaraldehyde, covalent crosslinkers, diisocyanate, as(±3.2) glutaraldehyde PLA/HA-II(a) PLA/HA-II(a) 0.15/1.0 4/1 0.15/1.0 3.83 75.3 4/1 (±3.2) 32.3 3.83 (±3.4) (±0.24) 75.3 32.3 (±3 HA is anow, water-soluble polysaccharide the content in(±0.24) the microspheres was such found acidic aqueous media viaofionic and also, iskinds safeof and biocompatible reagent for bioactive molecules [31,32]. HA interactions isand a water-soluble polysaccharide and the HA content in the genipin, ethylene glycol diglycidyl ether genipin, and glycidoxypropyltrimethoxy ethylene glycol diglycidyl silane, ether and have glycidoxypropyltrimethoxy been utilized silane, hav CH-PLA/HA-II(b) CH-PLA/HA-II(b) 0.3/1.0 4/1 0.3/1.0 4.16 (±0.28) 81.9 4/1 (±3.8) 30.2 4.16 (±3.2) (±0.28) 81.9 (±3.8) 30.2 (±3 toCH-PLA/HA significantly affect the[31,32]. properties ofisthe high HA in CH-PLA/HA complexes bioactive molecules HAfound a microspheres. water-soluble and the HAmicrospheres. content in the microspheres was to significantlyApolysaccharide affect thecontent properties of the A Taking into consideration the requirements Taking for into the consideration activity preservation the requirements of for the activity to crosslink CH [32–34]. to crosslink CH [32–34]. ()could CH-PLA withinamicrospheres PLA percentage () was CH-PLA of 38.7 with wt to % (see PLATable percentage 1) was of used 38.7for wt preparing % (seethan Table all the kinds 1) was of used foreven preparing all kinds of result preponderant formation ofasignificantly irregular rather microspheres CH-PLA/HA found affect properties of microspheres. A high HA content in CH-PLA/HA complexes could agglomerates result in the preponderant formation of irregular the microspheres; loaded TGF-β1 in the resulting microspheres, the loaded TGF-β1 in this in study, the resulting TPP, an microspheres, ionic crosslinker, in this was study, TPP, an ionic c (,) Different sample microspheres; sets: set-one (,) () Different and set-two sample (). sets: set-one () and set-two (). though a liberal amount of microspheres TPP was complexes applied. the other hand, a high HA content CH-PLA/HA high HA content in than CH-PLA/HA could result in preponderant formation of irregular agglomerates rather evenOn though a liberal amount of TPP wasinapplied. On the selected for crosslinking CH-PLA/HA microspheres selected for crosslinking because it can CH-PLA/HA crosslink the microspheres protonated because CH in it can crosslink the pr microspheres also cause the microspheres to markedly swell inalso aqueous media even ifOn such agglomerates ratherHA than microspheres even though a liberalwould amount of cause TPP was applied. the other hand,varied awould high content innow, CH-PLA/HA microspheres the microspheres to Up to now, kinds Up of covalent to crosslinkers, varied kinds such of covalent as glutaraldehyde, crosslinkers, diisocyanate, such as glutaraldehyde, diisocyan acidicmicrospheres aqueous media via ionic interactions acidic and disadvantageous also, media is a safe via ionic and biocompatible interactions and reagent also, for is a safe be achieved, which is aqueous quite for the microspheres tomicrospheres administer theto and biocompati other hand, acould high HA content in CH-PLA/HA microspheres would also cause the markedly swell in aqueous media even if such microspheres could behave achieved, which is silane, quite genipin, ethylene glycol diglycidyl genipin, ether ethylene and glycidoxypropyltrimethoxy glycol diglycidyl ether and glycidoxypropyltrimethoxy silane, been utilized have been utili bioactive molecules [31,32]. HA is a bioactive water-soluble molecules polysaccharide [31,32]. HA and is the a water-soluble HA content in polysaccharide the release of theswell loaded Therefore, the compositional of be CH-PLA/HA microspheres markedly in aqueous media even such microspheres could achieved, which is quite and the HA disadvantageous forTGF-β1. the microspheres to ifadminister theproportions release of activity the loaded TGF-β1. Therefore, Taking into consideration the Taking requirements into consideration for the the requirements preservation for of the activity preservation toCH-PLA/HA crosslink CH [32–34]. to crosslink CH [32–34]. microspheres was found CH-PLA/HA to significantly microspheres affect the properties was found of to the significantly microspheres. affect A the properties of the m and the applied amount of TPP were optimized using thethe orthogonal testing method. The weight ratio disadvantageous for the microspheres to administer release of the loaded TGF-β1. Therefore, the compositional proportions ofTGF-β1 CH-PLA/HA microspheres and an the applied amount of TPP an were the loaded TGF-β1 in the resulting the loaded microspheres, in the in this resulting study, microspheres, TPP, ionic in crosslinker, this study, was TPP, ionic crosslinker, high HA content in CH-PLA/HA complexes HA could content result in in CH-PLA/HA preponderant complexes formation could ofofirregular result in preponderant format ofoptimized HA to CH-PLA/HA was thereby controlled at 0.3/1.0 or lower whereas the ratio TPP matrix the compositional proportions ofhigh CH-PLA/HA microspheres and the applied amount of to TPP were using CH-PLA/HA the orthogonal testing method. The weight ratio of because HA to CH-PLA/HA was selected for crosslinking selected for microspheres crosslinking because CH-PLA/HA it can microspheres crosslink the protonated it can CH crosslink in the protonated CH agglomerates rather than microspheres agglomerates even though rather a liberal than amount microspheres of TPP was even applied. though On a liberal the amount of TPP was was formulated as 2/1 and 4/1, respectively, as illustrated inTPP Table 2. optimized using the orthogonal testing method. The ofweight ratio of was HA formulated to CH-PLA/HA was thereby controlled at 0.3/1.0 or lower whereas the ratio to matrix as 2/1 and acidic media via ionic acidic interactions aqueous media and also, via ionic is a interactions safe and biocompatible and also, is reagent a safe and for biocompatible reagent other aqueous hand, a high HA content in CH-PLA/HA other hand, microspheres a high HA would content also in CH-PLA/HA cause the microspheres microspheres to would also cause the Representative micrographs for presented Figure 2. The image thereby controlled at 0.3/1.0 orin lower whereas themicrospheres ratio of TPP are to matrix wasinformulated as 2/1 and 4/1, respectively, asSEM illustrated Table 2.different bioactive molecules [31,32]. bioactive HA is a molecules water-soluble [31,32]. polysaccharide HA is a water-soluble and the HA polysaccharide content in the and the HAbecontent in markedly swell in aqueous media even markedly if such swell microspheres in aqueous could media be achieved, even if such which microspheres is quite could achieved, in4/1, Figure 2A shows some CH-I microspheres with regular sphericity could be produced but respectively, as that illustrated in Table 2. Representative SEM micrographs for different microspheres are presented in Figure 2. The CH-PLA/HA microspheres was CH-PLA/HA found to significantly affect was found the to the significantly ofasthe microspheres. affect by the properties ofofthe disadvantageous forclumps the microspheres disadvantageous tomicrospheres administer the for release the properties microspheres of loaded to TGF-β1. administer Therefore, the Arelease themicrospheres loaded TGF some cracked and irregularly shaped beads were also seen, denoted arrows. Representative SEM micrographs for microspheres different microspheres are presented in white Figure 2. The image in Figure 2A shows that some CH-I with regular sphericity could be produced high HA content in CH-PLA/HA high HA complexes content in could CH-PLA/HA result in preponderant complexes could formation result in of preponderant irregular formation of amou irregu the compositional proportions of CH-PLA/HA the compositional microspheres proportions and the applied of CH-PLA/HA amount microspheres of TPP were and the applied Figure 2B that the fewer beads CH-I withshaped irregular shapes were viewed asdenoted compared that in image indisplays Figure 2A shows thatirregularly some microspheres with regular sphericity could be shown produced but some cracked clumps and beads were also seen, as by white arrows. agglomerates rather than microspheres agglomerates even rather though than amicrospheres liberal amount even of though TPP was aCH-PLA/HA applied. liberal amount On the of TPP wasofapplied. optimized using the orthogonal testing optimized method. using The weight the spherical orthogonal ratio of seen, testing HA to denoted method. The was weight ratio HA to On CH Figure 2A many CH/HA-II(b) microspheres were with smooth surface. The image for but some cracked clumps and irregularly shaped beads were also as by white arrows. Figure 2Band displays that the fewer beads with irregular shapes were viewed as compared that shown other hand, a high HA content other in CH-PLA/HA hand, a high HA microspheres content in would CH-PLA/HA also cause microspheres the microspheres would also to cause the microsphere thereby controlled at 0.3/1.0 or lower whereas thereby the controlled ratio of at TPP 0.3/1.0 to matrix or lower was whereas formulated the as ratio 2/1 of and TPP to matrix was formu CH-PLA/HA-II(b) exhibits that of them possessed regular sphericity. These images Figure 2B 2A displays that the fewer beads withmost irregular shapes were with viewed as compared that shown in Figure and microspheres many CH/HA-II(b) microspheres were spherical smooth surface. The image markedly swell in aqueous markedly media even swell if such in aqueous microspheres media could even if be such achieved, microspheres which is could quite be achieved, which is q 4/1, respectively, as illustrated in Table 4/1, 2. respectively, as illustrated in Table 2. reveal that CH-PLA/HA microspheres with well-controlled morphology can be obtained under the in Figure 2A and many CH/HA-II(b) microspheres were spherical with smooth surface. The image for CH-PLA/HA-II(b) microspheres exhibits that most of them possessed regular sphericity. These disadvantageous for the disadvantageous tofor administer for the the microspheres the release of to the loaded TGF-β1. the release Therefore, of the loaded are TGF-β1. Theref Representative SEMmicrospheres micrographs different Representative microspheres SEM micrographs areadminister presented for in different Figure 2. The2. presented in present processing while following formulated compositions provided inmicrospheres Table for CH-PLA/HA-II(b) microspheres exhibits that most of them possessed regular sphericity. These images reveal thatconditions CH-PLA/HA microspheres with well-controlled morphology can be applied obtained the compositional proportions theof compositional CH-PLA/HA proportions microspheres of and CH-PLA/HA the applied microspheres amount ofbeTPP and were the amount of TPPcou w image in Figure 2A shows that some CH-I image microspheres in Figure 2A with shows regular that sphericity some CH-I could microspheres produced with regular sphericity imagesthe reveal thatprocessing CH-PLA/HA microspheres with well-controlled morphology can provided be obtained under present conditions the formulated compositions optimized using the orthogonal optimized testing using method. thewhile orthogonal Thefollowing weight testing ratio of method. HA toThe CH-PLA/HA weight ratio was of also HA in to CH-PLA/HA but some cracked clumps and irregularly but shaped some cracked beads were clumps also and seen, irregularly as denoted shaped by white beads arrows. were seen, as denoted b under the present processing conditions while following the formulated compositions provided in Table 2. thereby controlled atthat 0.3/1.0 thereby or lower controlled whereas the at 0.3/1.0 ratio oforTPP lower to matrix whereas was the formulated ratio ofirregular TPP as to 2/1matrix and was formulated 2/1 Figure Table 2B displays the fewer beads Figure with irregular 2B displays shapes that were the fewer viewed beads as compared with that shown shapes were viewed as as compa 2. 4/1, respectively, asmany illustrated 4/1, inrespectively, Table microspheres 2.in Figure as illustrated in Table 2.with smooth in Figure 2A and CH/HA-II(b) 2Awere and many spherical CH/HA-II(b) microspheres surface. The were image spherical with smooth sur SEM micrographs Representative for SEM microspheres micrographs are for presented different microspheres insphericity. Figure 2. These are The presented in Figure for Representative CH-PLA/HA-II(b) microspheres exhibits for different CH-PLA/HA-II(b) that most of them microspheres possessed regular exhibits that most of them possessed regular2.sp image in reveal Figure that 2A shows that image some in CH-I Figure microspheres 2A shows that with some regular CH-Isphericity microspheres could with bewith regular sphericity could be produc images CH-PLA/HA microspheres images reveal with well-controlled that CH-PLA/HA morphology microspheres can beproduced obtained well-controlled morphology

in Figure 2A and many CH/HA-II(b) microspheres were spherical with smooth surface. The image for CH-PLA/HA-II(b) microspheres exhibits that most of them possessed regular sphericity. These images reveal that CH-PLA/HA microspheres with well-controlled morphology can be obtained under the 2018, present processing conditions while following the formulated compositions provided Pharmaceutics 10, 239 8 ofin14 Table 2.

Figure2.2.Representative Representative scanning electron microscopy (SEM) micrographs for (B) (A)CH/HA-II(b) CH-I; (B) Figure scanning electron microscopy (SEM) micrographs for (A) CH-I; CH/HA-II(b) and (C) CH-PLA/HA-II(b) (arrows cracked clumps and and (C) CH-PLA/HA-II(b) microspheresmicrospheres (arrows denote the denote crackedthe clumps and irregularly irregularly shaped beads). shaped beads).

Figure3 3depicts depicts size-distributions different kinds of microspheres. The microspheres in Figure size-distributions forfor different kinds of microspheres. The microspheres in set-one set-one had broad and scraggly size-distributions whilst the size-distributions for the microspheres had broad and scraggly size-distributions whilst the size-distributions for the microspheres in set-two in set-two became significantly narrowed peaked characteristics. shown in microspheres Table 2, the became significantly narrowed with peakedwith characteristics. As shown inAs Table 2, the set-one were crosslinked by TPP at a TPP/matrix ratio of TPP 2/1, amount the applied TPP inmicrospheres set-one werein crosslinked by TPP at a TPP/matrix ratio of 2/1, the applied might not amount might not be enough to sufficiently crosslink these microspheres, which is indirectly be enough to sufficiently crosslink these microspheres, which is indirectly confirmed by presence of confirmed by presence of cracked clumps and(see irregularly shaped beads to cracked clumps and irregularly shaped beads Figure 2A), leading to (see theirFigure broad 2A), and leading fluctuating their broad and fluctuating size-distributions. On the other hand,were the crosslinked microspheres in set-two were size-distributions. On the other hand, the microspheres in set-two with the redoubled crosslinked with the redoubled of and TPPthese as compared to could that inthus set-one and theseto amount of TPP as compared to thatamount in set-one microspheres be crosslinked could thusAccordingly, be crosslinkedthe to microspheres a relatively high degree. Accordingly, thedenser microspheres in amicrospheres relatively high degree. in set-two would become than those set-two would become denser than those crosslinked by a lower amount of TPP, resulting in their crosslinked by a lower amount of TPP, resulting in their relatively narrowed size-distributions. relatively narrowed size-distributions. The effects of HA and PLA components on the size-distributions of microspheres can also be The effects and3A PLA components the size-distributions of microspheres can also be seen from Figureof 3. HA Figure exhibits that in on contrast to CH-I microspheres, the size-distributions seen from Figure 3. Figure 3A exhibits that in contrast to CH-I microspheres, the size-distributions of of CH/HA-I(1) and CH/HA-I(2) microspheres remained to be scraggly and asymmetric but the CH/HA-I(1) and CH/HA-I(2) microspheres remained to be scraggly and asymmetric but the size-distributions for CH-PLA/HA-I(1) and CH-PLA/HA-I(2) microspheres showed somewhat size-distributions for CH-PLA/HA-I(1) and CH-PLA/HA-I(2) microspheres showed somewhat symmetry. In Figure 3B, improved smoothness and better symmetry of size-distributions for symmetry. In Figure 3B, improved smoothness and better symmetry of size-distributions for CH/HA-II(a), CH/HA-II(b), CH-PLA/HA-II(a) and CH-PLA/HA-II(b) were clearly observed as CH/HA-II(a), CH/HA-II(b), andresults CH-PLA/HA-II(b) as compared to that for CH-II CH-PLA/HA-II(a) microspheres. These indicate thatwere (1) atclearly a lowobserved TPP dosage compared to that for CH-II microspheres. These results indicate that (1) at a low TPP dosage (Figure (Figure 3A), the effect of HA component on the smoothness and shape symmetry of size-distributions 3A), the effect of HA component on the smoothness and shape symmetry of size-distributions is not is not significant but HA component together with PLA component could possibly improve the significant but HA component together with PLA component could possibly improve the size-distribution symmetry of microspheres; and (2) at a high TPP dosage (Figure 3B), HA component size-distribution symmetry of microspheres; and (2) at a high TPP dosage (Figure 3B), HA alone or in conjunction with PLA component can help to smooth the size-distributions of microspheres component alone or in conjunction with PLA component can help to smooth the size-distributions of and to reform their symmetry. It is known that symmetrical size-distribution and narrow distribution microspheres and to reform their symmetry. It is known that symmetrical size-distribution and interval are two extremely needed characteristics for the desired polymer microsphere carriers since narrow distribution interval are two extremely needed characteristics for the desired polymer these two parameters are strongly correlated to the repeatable controlled release behavior of the microsphere carriers since these two parameters are strongly correlated to the repeatable controlled microspheres [10,15,16]. Results in Figure 3 suggest that CH-PLA/HA-II(a) and CH-PLA/HA-II(b) release behavior of the microspheres [10,15,16]. Results in Figure 3 suggest that CH-PLA/HA-II(a) microspheres should be optimal ones in terms of their size-distribution. and CH-PLA/HA-II(b) microspheres should be optimal ones in terms of their size-distribution.

Pharmaceutics 2018, 10, 239 Pharmaceutics 2018, 10, x

9 of 14 9 of 14

Figure Figure 3. 3. Variations Variations in in size-distribution size-distribution of of microspheres microspheres in in set-one set-one (A) (A) and and in in set-two set-two (B) (B) (see (see Table Table 22 for their compositions). for their compositions).

3.3. Encapsulating Efficiency and Swelling Property of Microspheres 3.3. Encapsulating Efficiency and Swelling Property of Microspheres Data numerated in Table 2 show that CH/HA microspheres had higher EE in comparison to their Data numerated in Table 2 show that CH/HA microspheres had higher EE in comparison to respectively matched CH microspheres, meaning that presence of HA component in microspheres their respectively matched CH microspheres, meaning that presence of HA component in or utilization of an increasing amount of TPP could help to increase EE of microspheres. Table 2 also microspheres or utilization of an increasing amount of TPP could help to increase EE of elucidates that further rising EE can be achieved, given that the HA amount in the CH-PLA/HA microspheres. Table 2 also elucidates that further rising EE can be achieved, given that the HA microspheres is proper while the applied TPP is sufficient for crosslinking them, which is supported amount in the CH-PLA/HA microspheres is proper while the applied TPP is sufficient for by EE of CH-PLA/HA-II(a) and CH-PLA/HA-II(b) microspheres. As described in the experimental crosslinking them, which is supported by EE of CH-PLA/HA-II(a) and CH-PLA/HA-II(b) section, TGF-β1 was loaded into CH or CH/HA microspheres via physical blending and the chain microspheres. As described in the experimental section, TGF-β1 was loaded into CH or CH/HA network formed inside microspheres was loose in the wet sate because these chains were held by microspheres via physical blending and the chain network formed inside microspheres was loose in TPP-associated ionic linkages and HA-involved electrostatic force. As a result, some amounts of the wet sate because these chains were held by TPP-associated ionic linkages and HA-involved TGF-β1 on the superficial layer of the CH or CH/HA microspheres would be easily washed away electrostatic force. As a result, some amounts of TGF-β1 on the superficial layer of the CH or CH/HA during the microsphere preparation, leading to their relatively low EE. microspheres would be easily washed away during the microsphere preparation, leading to their TGF-β1 is a type of protein factor with isoelectric point of around 8.6 [35] and it will be in relatively low EE. positively charged state under neutral or acidic pH conditions. TGF-β1 molecular chains can thus TGF-β1 is a type of protein factor with isoelectric point of around 8.6 [35] and it will be in easily interact with negatively charged HA chains during the microsphere preparation. On the other positively charged state under neutral or acidic pH conditions. TGF-β1 molecular chains can thus hand, negatively charged HA chains will also interact or entangle with CH-PLA chains because CH easily interact with negatively charged HA chains during the microsphere preparation. On the other main chains contain a large amount of free amino groups. The short PLA side chains in CH-PLA are hand, negatively charged HA chains will also interact or entangle with CH-PLA chains because CH hydrophobic and they could possibly shape into heliciform hooks while being protruded from CH main chains contain a large amount of free amino groups. The short PLA side chains in CH-PLA are main chains when CH-PLA is exposed to aqueous media. As a result, PLA side chains will snag the hydrophobic and they could possibly shape into heliciform hooks while being protruded from CH TGF-β1, HA and CH chains to form entanglemants, which will certainly prevent the loss of TGF-β1 main chains when CH-PLA is exposed to aqueous media. As a result, PLA side chains will snag the during the microsphere preparation and result in high EE for the resulting microspheres. TGF-β1, HA and CH chains to form entanglemants, which will certainly prevent the loss of TGF-β1 SI of hydrophilic polymer carriers is closely correlated to their capability for administrating the during the microsphere preparation and result in high EE for the resulting microspheres. release of loaded drugs [36]. Hence, microspheres were measured to determine their SI and relevant SI of hydrophilic polymer carriers is closely correlated to their capability for administrating the data are listed in Table 2. It can be seen that (1) CH microspheres in both sets showed significantly release of loaded drugs [36]. Hence, microspheres were measured to determine their SI and relevant larger SI than other kinds of microspheres and CH/HA microspheres in each set had slightly reduced data are listed in Table 2. It can be seen that (1) CH microspheres in both sets showed significantly SI when compared to CH microspheres; and (2) SI of CH-PLA/HA microspheres in each set was much larger SI than other kinds of microspheres and CH/HA microspheres in each set had slightly smaller than that for CH or CH/HA microspheres. The large SI for CH microspheres is attributed reduced SI when compared to CH microspheres; and (2) SI of CH-PLA/HA microspheres in each to their high hydrophilicity and the loose chain network constructed by ionic bonding. CH contains set was much smaller than that for CH or CH/HA microspheres. The large SI for CH microspheres many polar groups such as amino and hydroxyl groups and thus, intermolecular and intramolecular is attributed to their high hydrophilicity and the loose chain network constructed by ionic bonding. hydrogen bonds can easily form among CH chains in the dry state. However, these hydrogen bonds CH contains many polar groups such as amino and hydroxyl groups and thus, intermolecular and could be greatly weakened or even broken in the wet state [32] and consequently, hydrated CH intramolecular hydrogen bonds can easily form among CH chains in the dry state. However, these microspheres would be swollen significantly. With respect to CH/HA microspheres, HA chains can hydrogen bonds could be greatly weakened or even broken in the wet state [32] and consequently, interact with CH chains via electrostatic force and physical entanglement. Taking into account the hydrated CH microspheres would be swollen significantly. With respect to CH/HA microspheres, low HA/CH ratio and the applied amount of TPP, the resulting CH/HA microspheres could become HA chains can interact with CH chains via electrostatic force and physical entanglement. Taking more compact than CH microspheres and show their SI similar to or slightly smaller than that for CH into account the low HA/CH ratio and the applied amount of TPP, the resulting CH/HA microspheres. In cases of CH-PLA/HA microspheres, their SI would be mediated by PLA component, microspheres could become more compact than CH microspheres and show their SI similar to or slightly smaller than that for CH microspheres. In cases of CH-PLA/HA microspheres, their SI

Pharmaceutics 2018, 10, x Pharmaceutics 2018, 10, 239

10 of 14 10 of 14

would be mediated by PLA component, CH component as well as TPP. The hydrophobic PCL side chains in CH-PLA chains could turn inward, snag molecular chains chains and aggregate inside the CH component as well as TPP. The hydrophobic PCLother side chains in CH-PLA could turn inward, microspheres whereas the CH main chains could stretch outward and entangle with HA chains snag other molecular chains and aggregate inside the microspheres whereas the CH main chains could duringoutward the preparation of CH-PLA a water-based solvent was applied. The stretch and entangle with HAmicrospheres chains duringbecause the preparation of CH-PLA microspheres because entanglements formed among these components would reside inside the microsphere and resist the a water-based solvent was applied. The entanglements formed among these components would reside microsphere swelling, giving rise to their reduced SI. inside the microsphere and resist the microsphere swelling, giving rise to their reduced SI. 3.4. Release Release Profiles Profiles of of Microspheres Microspheres 3.4. In principle, principle, multiple multiple factors factors involving involving in in the the structure, structure, morphology morphology and and physicochemical physicochemical In property of the drug-loaded microspheres can exert complicated impacts on the release patterns of property of the drug-loaded microspheres can exert complicated impacts on the release patterns the loaded drugs [35–37]. In this study, the major processing parameters and the compositions for of the loaded drugs [35–37]. In this study, the major processing parameters and the compositions for CH-PLA/HA microspheres microspheres were werealready alreadyoptimized optimized to achieve optional microspheres that basic meet CH-PLA/HA to achieve optional microspheres that meet basic requirements for the delivery of TGF-β1. On this basis, the release profiles of the microspheres requirements for the delivery of TGF-β1. On this basis, the release profiles of the microspheres were were examined to screen out the optimal examined to screen out the optimal one. one. Figure 44 represents represents the the time-dependent time-dependent TGF-β1 TGF-β1 release release profiles profiles for for different different microspheres. microspheres. CH CH Figure microspheres in both sets had fast release rates and the cumulative TGF-β1 amount released from microspheres in both sets had fast release rates and the cumulative TGF-β1 amount released from these microspheres microspheres reached reached around around 70% 70% or or even evenhigher higherin inone oneweek. week.CH/HA CH/HA microspheres microspheres in in each each these set showed reduced release rate within the same sampling time interval when compared to CH set showed reduced release rate within the same sampling time interval when compared to CH microspheres. In Figure 4A, CH/HA-I(2) microspheres showed their release pattern very similar to microspheres. In Figure 4A, CH/HA-I(2) microspheres showed their release pattern very similar to that for forCH/HA-I(1) CH/HA-I(1)microspheres microsphereswithout without significant difference > 0.05); in Figure the that significant difference (p > (p 0.05); but inbut Figure 4B, the 4B, release release profile for CH/HA-II(b) microspheres notably differed from that for CH/HA-II(a) profile for CH/HA-II(b) microspheres notably differed from that for CH/HA-II(a) microspheres with microspheres with measurable (p data < 0.05). Based in onTable the data provided in Table can be measurable difference (p < 0.05). difference Based on the provided 2, it can be drawn that at2,a it low TPP drawn that at a low TPP dosage, the of HAbehavior component on the microspheres release behavior of CH/HA dosage, the effect of HA component oneffect the release of CH/HA is insignificant microspheres is insignificant but at a high TPP dosage, the HA component could combine TPP but at a high TPP dosage, the HA component could combine with TPP to significantly with regulate to significantly regulate release pattern of the CH/HA tomicrospheres. In contrast to these the release pattern of the the CH/HA microspheres. In contrast these observations, CH-PLA/HA observations, CH-PLA/HA microspheres behaved in different manners. CH-PLA/HA microspheres microspheres behaved in different manners. CH-PLA/HA microspheres in both sets showed greatly in both sets showed greatly reducedwith initial release as compared with CH CH/HA microspheres in reduced initial release as compared CH or CH/HA microspheres inor the corresponding set and the corresponding their TGF-β1 release rateby could be content conjointly by the HA their TGF-β1 release set rateand could be conjointly modulated the HA and modulated the applied TPP dosage. content and the applied TPP dosage. Among them, CH-PLA/HA-II(b) microspheres could serve as Among them, CH-PLA/HA-II(b) microspheres could serve as a suitable carrier for the delivery ofa suitable since carrier for the the first delivery of TGF-β1 since after day, these microspheres able in to TGF-β1 after day, these microspheres arethe ablefirst to control the release rate of are TGF-β1 control the releaselinear rate offashion TGF-β1 an approximately an approximately forinaround 2 weeks. linear fashion for around 2 weeks.

Figure Figure 4. 4. Release Release profiles profiles of of TGF-β1 TGF-β1 from from microspheres microspheres in in set-one set-one (A) (A) and and set-two set-two (B) (B) (n (n == 3). 3).

These results are reasonable if more details for the microspheres are figured out. In principle, These results are reasonable if more details for the microspheres are figured out. In principle, the release of drugs or bioactive reagents from polymeric matrices usually involves in swelling, the release of drugs or bioactive reagents from polymeric matrices usually involves in swelling, diffusion, swelling followed by diffusion and erosion [10,38]. CH microspheres had high SI because diffusion, swelling followed by diffusion and erosion [10,38]. CH microspheres had high SI because of the loose ionic linkages among CH chains and the chain network is unable to effectively hinder of the loose ionic linkages among CH chains and the chain network is unable to effectively hinder TGF-β1 molecules from rapidly defusing into the surrounding media, leading to their high initial burst TGF-β1 molecules from rapidly defusing into the surrounding media, leading to their high initial and subsequent fast release. TGF-β1 molecules inside CH/HA microspheres face somewhat different burst and subsequent fast release. TGF-β1 molecules inside CH/HA microspheres face somewhat

Pharmaceutics 2018, 10, x Pharmaceutics 2018, 10, 239

11 of 14 11 of 14

different situations. At physiological pH, TGF-β1 will be positively charged and can interact electrostatically with negatively charged molecules.charged In addition, molecules will also situations. At physiological pH, TGF-β1 willHA be positively and canHA interact electrostatically entangle with CH chainsHA due to physical or electrostatic interactions. TGF-β1 molecules with negatively charged molecules. In addition, HA molecules will Therefore, also entangle with CH chains would be held by both CH and HA and are relatively difficult to be released by diffusion, giving due to physical or electrostatic interactions. Therefore, TGF-β1 molecules would be held by both CH rise to their reduced release rates. In the case of CH-PLA/HA microspheres, the CH-PLA component and HA and are relatively difficult to be released by diffusion, giving rise to their reduced release rates. contains side chains that have highthe hydrophobicity and low degradation rate and that hence, in In the casePLA of CH-PLA/HA microspheres, CH-PLA component contains PLA side chains have addition to the mentioned interactions occurred in CH or CH/HA microspheres, TGF-β1 molecules high hydrophobicity and low degradation rate and hence, in addition to the mentioned interactions loaded ininCH-PLA/HA microspheres willTGF-β1 be dragged by PCL side in chains and trapped inside the occurred CH or CH/HA microspheres, molecules loaded CH-PLA/HA microspheres microspheres during the microsphere fabrication, which will cause increased resistance to TGF-β1 will be dragged by PCL side chains and trapped inside the microspheres during the microsphere molecules and slow them release rate. fabrication, which will cause increased resistance to TGF-β1 molecules and slow them release rate. In the thetherapeutic therapeuticmodel modelinvolving involving cartilage defects, applied TGF-β1 dosage needs be In cartilage defects, thethe applied TGF-β1 dosage needs to betolow low for functioning as a chemotaxis agent in the early stage and thereafter, to be sufficient for for functioning as a chemotaxis agent in the early stage and thereafter, to be sufficient for serving as serving as an effective chondrogenic agent until repair is[13,39]. completed [13,39]. Results in Figure an effective chondrogenic agent until the repair is the completed Results in Figure 4 indicate that4 indicate that TGF-β1-loaded CH or CH/HA microspheres in cartilage TGF-β1-loaded CH or CH/HA microspheres are unsuitableare for unsuitable applicationsfor inapplications cartilage repair because repair because of their severe initial burst release features and fast release rates. On the other hand, of their severe initial burst release features and fast release rates. On the other hand, CH-PLA/HA-II(b) CH-PLA/HA-II(b) show promising potential for controlling TGF-β1 release in the microspheres showmicrospheres promising potential for controlling the TGF-β1 releasethe in the desired manner to desired manner to meet the requirements in cartilage repair. meet the requirements in cartilage repair. 3.5. TGF-β1 Bioactivity 3.5. Four kinds kindsof of microspheres selected fromsets two sets of microsphere samples for the Four microspheres werewere selected from two of microsphere samples for the bioactivity bioactivity of assessment of TGF-β1 the released TGF-β1that considering that they had higher HA content and assessment the released considering they had higher HA content and were crosslinked wereacrosslinked with a higher in corresponding comparison toones the corresponding ones in each with higher TPP/matrix ratio inTPP/matrix comparisonratio to the in each set. Relevant results set.cell Relevant for are cellgraphed growth in inhibition graphedininFigure Figure Bar-graphs in Figure 5A for growthresults inhibition Figure 5. are Bar-graphs 5A5.indicate that cell inhibition indicate that higher than no 35% before day 7 and in there were no significant was higher thancell 35%inhibition before daywas 7 and there were significant differences the cell growth inhibition differences in thesample cell growth inhibition amongitdifferent sample it candetected be seen among different groups. Nevertheless, can be seen thatgroups. the cell Nevertheless, growth inhibition that the cell growth inhibition detected from microsphere groups at day 10 was shown to be from microsphere groups at day 10 was shown to be significantly lower than that of free TGF-β1 significantly lower than that thatthe of free TGF-β1 sample meaningatthat released that sample group, meaning released TGF-β1 thatgroup, was collected daythe 10 had lowerTGF-β1 bioactivity was collected at with day 10 had lower bioactivity when compared with free TGF-β1. when compared free TGF-β1.

Figure Figure 5. 5. Percent Percent mink mink lung lung cell cell growth growth inhibition inhibition (A, (A, free free TGF-β1 TGF-β1 without without storing storing treatment; treatment; and and B, B, ◦ C for different time intervals free TGF-β1 stored in DMEM (dulbecco’s modified eagle medium) at 37 free TGF-β1 stored in DMEM (dulbecco’s modified eagle medium) at 37 °C for different time changing from 10 tofrom 21 days) different microspheres during varied intervals changing 10 to of 21TGF-β1 days) ofreleased TGF-β1 from released from different microspheres duringperiods varied (n = 3; N.S. no significance; * p < 0.05). periods (n = 3; N.S. no significance; * p < 0.05).

Significant activity loss of the released TGF-β1 has been detected from some polylactide and Significant activity loss of the released TGF-β1 has been detected from some polylactide and poly(DL-lactide-co-glycolide) microspheres [40,41]. These polyester microspheres are usually prepared poly(DL-lactide-co-glycolide) microspheres [40,41]. These polyester microspheres are usually by using organic solvents and their high hydrophobicity and the involved solvents could denature prepared by using organic solvents and their high hydrophobicity and the involved solvents could the bioactivity of protein factors [7,10]. In the present study, the used polysaccharides are highly denature the bioactivity of protein factors [7,10]. In the present study, the used polysaccharides are hydrophilic and the microspheres were prepared using an aqueous solvent. Therefore, the employed highly hydrophilic and the microspheres were prepared using an aqueous solvent. Therefore, the materials and the microsphere preparation method would not significantly impair the activity of the employed materials and the microsphere preparation method would not significantly impair the

Pharmaceutics 2018, 10, 239

12 of 14

loaded TGF-β1. In the present case, one possible reason for the activity loss of the released TGF-β1 is that the growth factor lost its functionality by autolysis, as mentioned in some studies [40,42]. In this study, the supernatant collected at day 10 contained a cumulative amount of TGF-β1 that was released from the microspheres during 10 days. TGF-β1 is a type of protein factor and the released TGF-β1 in the supernatant might be autolyzed by way of proteolysis to some extent during this period of time, resulting in its lower activity than free TGF-β1. To clarify whether this explanation is reasonable, the free TGF-β1 was stored in DMEM at 37 ◦ C for the matched periods of time up to 21 days so that it was already exposed to the same environment for the same time interval in comparison to the released TGF-β1 before they both were used for the cell tests. The bioactivity of the released TGF-β1 was further examined for an extended period of time to make additional comparisons and relevant results are presented in Figure 5B. It can be seen that the cell growth inhibition matching with the store-treated free TGF-β1 progressively decreased as time advanced, verifying that the bioactivity of free TGF-β1 would gradually lose if it is exposed to the hydrolysis environment for a period of time longer than a threshold value. It is worth noting from Figure 5B that the released TGF-β1 from different microsphere samples were able to inhibit the cell growth at the same level as compared with the store-treated free TGF-β1 without showing significant differences, indirectly conforming that the activity loss of the released TGF-β1 is not arisen from the employed materials as well as the used microsphere preparation method but possibly from proteolysis. On the basis of results presented in Figure 5, it can be concluded that the presently developed CH-PLA/HA microspheres are reliable for effectively preserving the bioactivity of released TGF-β1. 4. Conclusions Chitosan(CH)-polylactide(PLA) copolymers with PLA side chains at the C-6 sites of CH backbone and good solubility in certain water-based solvents could be used to combine with polyanionic hyaluronic acid (HA) for preparing CH-PLA/HA complexes. Simply blending CH-PLA component with HA component in aqueous phase at formulated ratios would generate irregularly shaped CH-PLA/HA particles or CH-PLA/HA agglomerates and this simple blending method was found to be unpractical for preparing CH-PLA/HA complex microspheres with regular sphericity and controlled sizes. The proposed emulsification preparation method was demonstrated to be successful for constructing TGF-β1-loaded CH-PLA/HA complex microspheres and the resulting microspheres could be endowed with controllable sizes and high encapsulating efficiency. Under optimized preparation conditions, the structures and properties of the CH-PLA/HA microspheres would be synergistically regulated by the PLA component, the applied dose of crosslinker and the incorporated HA amount. The optimal TGF-β1-loaded CH-PLA/HA microspheres had symmetrical size-distribution as well as narrow distribution interval and were able to effectively administrate TGF-β1 release in approximately linear manners for around 2 weeks without showing severe initial burst release features. The released TGF-β1 was detected to be bioactive as compared with the free TGF-β1. Results suggest that these TGF-β1-loaded CH-PLA/HA microspheres had potential in cartilage repair applications due to their controlled TGF-β1 release features and good TGF-β1 bioactivity preservation. Author Contributions: Y.W. and J.W. conceived and designed the experiments; Q.M., J.L. (Jing Li) and J.L. (Jiaoyan Liu) performed the experiments; Y.W. and J.W. wrote the paper. Funding: This work was funded by the National Key R&D Program of China (Grant No. 2016YFC1100100) and the National Natural Science Foundation of China (Grant No. 81572144). Conflicts of Interest: The authors declare no conflict of interest.

Pharmaceutics 2018, 10, 239

13 of 14

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.

13.

14. 15. 16.

17. 18.

19. 20. 21. 22. 23. 24.

Hunziker, E.B. Articular cartilage repair: Basic science and clinical progress. Osteoarthr. Cartil. 2002, 10, 432–463. [CrossRef] [PubMed] Camarero-Espinosa, S.; Rothen-Rutishauser, B.; Foster, E.J.; Weder, C. Articular cartilage: From formation to tissue engineering. Biomater. Sci. 2016, 4, 734–767. [CrossRef] [PubMed] Chiang, H.; Jiang, C.C. Repair of articular cartilage defects: Review and perspectives. J. Formos. Med. Assoc. 2009, 108, 87–101. [CrossRef] Berthiaume, F.; Maguire, T.J.; Yarmush, M.L. Tissue engineering and regenerative medicine: History, progress and challenges. Annu. Rev. Chem. Biomol. Eng. 2011, 2, 403–430. [CrossRef] [PubMed] Edgar, L.; McNamara, K.; Wong, T.; Tamburrini, R.; Katari, R.; Orlando, G. Heterogeneity of scaffold biomaterials in tissue engineering. Materials 2016, 9, 332. [CrossRef] [PubMed] Lam, J.; Lu, S.; Kasper, F.K.; Mikos, A.G. Strategies for controlled delivery of biologics for cartilage repair. Adv. Drug Deliv. Rev. 2015, 84, 123–134. [CrossRef] [PubMed] Wang, Z.; Wang, Z.; Lu, W.W.; Zhen, W.; Yang, D.; Peng, S. Novel biomaterial strategies for controlled growth factor delivery for biomedical applications. NPG Asia Mater. 2017, 9, e435. [CrossRef] Hunziker, E.B. Growth-factor-induced healing of partial-thickness defects in adult articular cartilage. Osteoarthr. Cartil. 2001, 9, 22–32. [CrossRef] [PubMed] Whitaker, M.J.; Quirk, R.A.; Howdle, S.M.; Shakesheff, K.M. Growth factor release from tissue engineering scaffolds. J. Pharm. Pharm. 2001, 53, 1427–1437. [CrossRef] Lee, K.Y.; Yuk, S.H. Polymeric protein delivery systems. Prog. Polym. Sci. 2007, 32, 669–697. [CrossRef] Wan, Y.; Wen, D. Preparation and characterization of porous conducting poly (DL-lactide) composite membranes. J. Membr. Sci. 2005, 246, 193–201. [CrossRef] Perinelli, D.R.; Campana, R.; Skouras, A.; Bonacucina, G.; Cespi, M.; Mastrotto, F.; Baffone, W.; Casettari, L. Chitosan loaded into a hydrogel delivery system as a strategy to treat vaginal co-infection. Pharmaceutics 2018, 10, 23. [CrossRef] [PubMed] Kim, S.E.; Park, J.H.; Cho, Y.W.; Chung, H.; Jeong, S.Y.; Lee, E.B.; Kwon, I.C. Porous chitosan scaffold containing microspheres loaded with transforming growth factor-β1: Implications for cartilage tissue engineering. J. Control. Release 2003, 91, 365–374. [CrossRef] Mohammed, M.A.; Syeda, J.T.M.; Wasan, K.M.; Wasan, E.K. An overview of chitosan nanoparticles and its application in non-parenteral drug delivery. Pharmaceutics 2017, 9, 53. [CrossRef] [PubMed] Gan, Q.; Wang, T. Chitosan nanoparticle as protein delivery carrier: Systematic examination of fabrication conditions for efficient loading and release. Colloids Surf. B Biointerfaces 2007, 59, 24–34. [CrossRef] [PubMed] Timin, A.S.; Gao, H.; Voronin, D.V.; Gorin, D.A.; Sukhorukov, G.B. Inorganic/organic multilayer capsule composition for improved functionality and external triggering. Adv. Mater. Interfaces 2017, 4, 1600338. [CrossRef] Luo, Y.; Wang, Q. Recent development of chitosan-based polyelectrolyte complexes with natural polysaccharides for drug delivery. Int. J. Biol. Macromol. 2014, 64, 353–367. [CrossRef] [PubMed] Parashar, P.; Rathor, M.; Dwivedi, M.; Saraf, S.A. Hyaluronic acid decorated naringenin nanoparticles: Appraisal of chemopreventive and curative potential for lung cancer. Pharmaceutics 2018, 10, 33. [CrossRef] [PubMed] Hemmati-Sadeghi, S.; Ringe, J.; Dehne, T.; Haag, R.; Sittinger, M. Hyaluronic acid influence on normal and osteoarthritic tissue-engineered cartilage. Int. J. Mol. Sci. 2018, 19, 1519. [CrossRef] [PubMed] Duceppe, N.; Tabrizian, M. Factors influencing the transfection efficiency of ultra low molecular weight chitosan/hyaluronic acid nanoparticles. Biomaterials 2009, 30, 2625–2631. [CrossRef] [PubMed] Lu, H.D.; Zhao, H.Q.; Wang, K.; Lv, L.L. Novel hyaluronic acid-chitosan nanoparticles as non-viral gene delivery vectors targeting osteoarthritis. Int. J. Pharm. 2011, 420, 358–365. [CrossRef] [PubMed] Fuente, M.D.L.; Seijo, B.; Alonso, M.J. Bioadhesive hyaluronan-chitosan nanoparticlescan transport genes across the ocular mucosa and transfect ocular tissue. Gene Ther. 2008, 15, 668–676. [CrossRef] [PubMed] Wan, Y.; Creber, K.A.M.; Peppley, B.; Bui, V.T. Ionic conductivity and tensile properties of hydroxyethyl and hydroxypropyl chitosan membranes. J. Polym. Sci. Part B Polym. Phys. 2004, 42, 1379–1397. [CrossRef] Feng, H.; Dong, C.M. Synthesis and characterization of phthaloyl-chitosan-g-poly(L-lactide) using an organic catalyst. Carbohydr. Polym. 2007, 70, 258–264. [CrossRef]

Pharmaceutics 2018, 10, 239

25.

26. 27.

28. 29.

30. 31. 32. 33. 34. 35.

36. 37. 38. 39. 40. 41.

42.

14 of 14

Liu, L.; Shi, A.; Guo, S.; Fang, Y.; Chen, S.; Li, J. Preparation of chitosan-g-polylactide graft copolymers via self-catalysis of phthaloylchitosan and their complexation with DNA. React. Funct. Polym. 2010, 70, 301–305. [CrossRef] Kim, M.S.; Ahn, S.M.; Moon, A. In vitro bioassay for transforming growth factor-β using XTT method. Arch. Pharm. Res. 2002, 25, 903–909. [CrossRef] [PubMed] Wu, H.; Zhang, J.; Xiao, B.; Zan, X.; Gao, J.; Wan, Y. N-(2-hydroxypropyl)-3-trimethylammonium chitosan-poly(ε-caprolactone) copolymers and their antibacterial activity. Carbohydr. Polym. 2011, 83, 824–830. [CrossRef] Yao, F.; Chen, W.; Wang, H.; Liu, H.; Yao, K.; Sun, P.; Lin, H.A. Study on cytocompatible poly(chitosan-g-L-lactic acid). Polymer 2003, 44, 6435–6441. [CrossRef] Cacciotti, I.; Ciocci, M.; Giovanni, E.D.; Nanni, F.; Melino, S. Hydrogen sulfide-releasing fibrous membranes: Potential patches for stimulating human stem cells proliferation and viability under oxidative stress. Int. J. Mol. Sci. 2018, 19, 2368. [CrossRef] [PubMed] Desai, K.G.H. Chitosan nanoparticles prepared by ionotropic gelation: An overview of recent advances. Crit. Rev. Ther. Drug Carrier Syst. 2016, 33, 107–158. [CrossRef] [PubMed] Mao, S.; Guo, C.; Shi, Y.; Li, L.C. Recent advances in polymeric microspheres for parenteral drug delivery-part 1. Expert Opin. Drug Deliv. 2012, 9, 1161–1176. [CrossRef] [PubMed] Muzzarelli, R.A.A. Chitosan composites with inorganics, morphogenetic proteins and stem cells, for bone regeneration. Carbohydr. Polym. 2011, 83, 1433–1445. [CrossRef] Almeida, J.F.; Fonseca, A.; Baptista, C.M.S.G.; Leite, E.; Gil, M.H. Immobilization of drugs for glaucoma treatment. J. Mater. Sci. Mater. Med. 2007, 18, 2309–2317. [CrossRef] [PubMed] Jiang, Z.; Yu, Y.; Wu, H. Preparation of CS/GPTMS hybrid molecularly imprinted membrane for efficient chiral resolution of phenylalanine isomers. J. Membr. Sci. 2006, 280, 876–882. [CrossRef] Lee, J.E.; Kim, K.E.; Kwon, I.C.; Ahn, H.J.; Lee, S.H.; Cho, H.; Kim, H.J.; Seong, S.C.; Lee, M.C. Effects of the controlled-released TGF-β1 from chitosan microspheres on chondrocytes cultured in a collagen/ chitosan/glycosaminoglycan scaffold. Biomaterials 2004, 25, 4163–4173. [CrossRef] [PubMed] George, M.; Abraham, T.E. Polyionic hydrocolloids for the intestinal delivery of protein drugs: Alginate and chitosan—A review. J. Control. Release 2006, 114, 1–14. [CrossRef] [PubMed] Kumari, A.; Yadav, S.K.; Yadav, S.C. Biodegradable polymeric nanoparticles based drug delivery systems. Colloids Surf. B Biointerfaces 2010, 75, 1–18. [CrossRef] [PubMed] Peppas, L.B. Polymers in controlled drug delivery. Med. Plast. Biomater. 1997, 4, 34–44. Giannoni, P.; Hunziker, E.B. Release kinetics of transforming growth factor-β1 from fibrin clots. Biotechnol. Bioeng. 2003, 83, 121–123. [CrossRef] [PubMed] DeFail, A.J.; Chu, C.R.; Izzo, N.; Marra, K.G. Controlled release of bioactive TGF-β1 from microspheres embedded within biodegradable hydrogels. Biomaterials 2006, 27, 1579–1585. [CrossRef] [PubMed] Parker, J.A.T.C.; Brunner, G.; Walboomers, X.F.; Von Den Hoff, J.W.; Maltha, J.C.; Jansen, J.A. Release of bioactive transforming growth factor β3 from microtextured polymer surfaces in vitro and in vivo. Tissue Eng. 2002, 8, 853–861. [CrossRef] [PubMed] Nicoll, S.B.; Denker, A.E.; Tuan, R.S. In vitro characterization of transforming growth factor-beta1-loaded composites of biodegradable polymer and mesenchymal cells. Cells Mater. 1995, 5, 231–244. © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).