Enhanced bioactivity of transform growth factor-β1 from sulfated chitosan microspheres for in vitro chondrogenesis of mesenchymal stem cells

Materials

Chitosan (deacetylation degree 85 %, Mη = 2.5 × 10⁴) was a product from Haidebei Bioengineering Co. Ltd. (Qingdao, China). Recombinant human TGF-β1 was purchased from Peprotech (New Jersey, USA). Dichloroacetic acid (DCAA), N,N-dimethylformamide (DMF), sodium borohydride, glutaraldehyde (GA), isopropyl alcohol, and petroleum ether were purchased from Sinopharm Chemical Reagent Co. Ltd. (Shanghai, China). All other chemicals were of regent grade and were used as received.

3,6-O-sulfated chitosan was prepared according to Ronge Xing’s methods with modification [36]. Briefly, sulfating reagent was prepared by adding 5 mL of HClSO₃ dropwise with stirring to 20 mL DMF which was previously cooled at 0∼4°C. 25 mL sulfating reagent were added to 60 mL chitosan solution containing 2 mL DCAA with swirling. Then the reaction was run at 65 °C for 2 h, and then washed with 95 % ethanol and water, respectively. The solution was dialyzed against deionized water, and then lyophilized to get 3,6-O-sulfated chitosan. The number average molecular weight and the sulfur content of the obtained sulfated chitosan was 2.5 × 10⁴ and 20.7 %, respectively.

Preparation of SCMs

SCMs were fabricated by membrane emulsification with some modifications [25]. Briefly, sulfated chitosan dissolved in water with the concentration of 1.5 %. A definite volume of sulfated chitosan solution was pressed though the glass membrane with uniform pore size of 1.9 μm into oil phase (liquid paraffin and Span-80 in a volume ratio of 24:1) under nitrogen pressure (5∼15 kPa). The volume ratio of aqueous solution and liquid paraffin is 1:75. At the end of emulsification, the emulsion was cross-linked by glutaraldehyde with same volume of sulfated chitosan solution at 37 °C for 4 h. After washed three times with petroleum ether, isopropyl alcohol and Milli-Q water respectively, sodium borohydride was added to react with the remaining of aldehyde group in consideration of toxicity of glutaraldehyde. SCMs were collected by lyophilization and stored at 4 °C for further experiments.

Loading of TGF-β1

TGF-β1 was loaded by immersing 50 μg sterilized SCMs into 150 μL 0.1 % bull serum albumin/phosphate-buffered saline solution (0.1 % BSA/PBS, pH=7.2) containing 100 ng TGF-β1 at room temperature for 30 min with shaking. Then the mixture was incubated at room temperature for 30 min and remove supernatant by centrifugation to obtain TGF-β1/SCMs. To determine the loading efficiency of TGF-β1, TGF-β1 concentration of supernatant was assayed by enzyme-linked immunosorbent assay kit (ELISA, eBioscience, San Diego, USA).

Characterization of SCMs

The morphology of SCMs before and after adsorption of TGF-β1 was measured by scanning electron microscope (SEM, Hitachi S-4800, Tokyo, Japan). The mean diameters of SCMs and TGF-β1/SCMs were determined using a dynamic lighting scattering (DLS) method by a Particle Size Analyzer (Brookhaven, USA) at 25 °C and a fixed light angle of 90°. The ζ potential of the microspheres was tested using a ζ Potential Analyzer (DelsaTM, Beckman Coulter, Brea, USA).

TGF-1 release study

TGF-β1/SCMs were incubated with 120 μL 0.1 % BSA/PBS in capped tube under static condition. At each time point, sample was centrifuged for 6 min at 8000 rpm to collect the supernatant. Then 120 μL fresh 0.1 % BSA/PBS was added to each sample, which was incubated until the next time point. The collected supernatant was stored at –80 °C until further analysis. ELISA was used to determine the concentration of the released TGF-β1.There samples were analyzed for each time point.

Bioactivity of TGF-β1 with the storage time

TGF-β1/SCMs were incubated with 120 μL 0.1 % BSA/PBS in capped tube under static condition. At each time point, the sample was centrifuged for 6 min at 8000 rpm. 120 μL fresh 0.1 % BSA/PBS was added and then incubated for 4 h with shaking. After incubation, the sample was centrifuged for 6 min at 8000 rpm again, following by collection of supernatant. The collected supernatant was stored at –80 °C for further analysis. The concentration of the released TGF-β1 was measured by ELISA. There samples were analyzed for each time point.

Cell culture

MSCs were isolated from bone marrow aspirate of New Zealand rabbits as describe previously [5]. Briefly, appropriate volume rabbit bone marrow was aspirated into a 10 mL sterile syringe containing 5000 U heparin, and then diluted with Dulbecco’s modified Eagle’s medium (DMEM, GIBCO, Wu han, China). The mixture was centrifuged two times with a speed of 2000 rpm for 20 min. The precipitate cell pellet was cultured in low-glucose DMEM containing 10 % bovine serum (GIBCO, Wu han, China), 100 U/mL penicillin and 100 μg/mL streptomycin. After incubation at 37 °C under 5 % CO₂ for 5 days, non-adherent cells were removed by washing twice with the basal medium. Medium was changed every two days until reaching 90 % confluence. MSCs were retrieved by treatment with 0.25 % trypsin and employed at passages 2 or 3.

Cytotoxicity of SCMs

MSCs were seeded in 96-well plate at a destiny of 1.0 × 10⁴ cells/well. After 24 h, culture medium was replaced with 200 μL culture medium containing different concentration of SCMs (0 μg, 1 μg, 2 μg, 5 μg and 10 μg/1.0 × 10⁴ cells). At each time point, cell viability was performed by MTT assay. Samples were run in quintuplicate for each experiment.

Cell proliferation

Transwells model was used to evaluate cell proliferation behaviors [37]. MSCs were seeded in the bottom layer of 12-well plate at a destiny of 1.5 × 10⁴ cells/well and then incubated at 37 °C under 5 % CO₂ for 1 day. After one day’s incubation, 2 mL culture medium containing TGF-β1/SCMs (2 μg/ml SCMs, 5 ng/ml TGF-β1) were added. The culture mediums containing the same concentration of SCMs or TGF-β1 respectively were chosen as the controls. The culture medium was changed twice per week. At each time point, the cell plate was washed twice with PBS and stored at –20 °C for further analysis. DNA content was performed by PicoGreen assay (Molecular Probes, Eugene, USA) according to the instructions of the manufacturer. Samples were run in quadruplicate for each experiment.

Quantitative analysis of GAG secretion

The quantity of the secreted GAGs was assayed by Alcian blue dye assay. At each point, 200 μL papain solutions was added to the samples for digestion, and then the mixture was incubated with 200 μL of 1.4 mg/mL Alcian blue dye for 10 min. The adsorption of the mixture at 490 nm was measured by a spectrophotometer, with chondroitin sulfate C (Sigma, Santa Clara, USA) as a standard.

Quantitative reverse transcription-polymerase chain reaction (qRT-PCR)

qRT-PCR was used to detect the expression of chondrogenesis-related genes, i.e., type II collagen and aggrecan. The samples were prepared as described in section of cell proliferation. At day 14, the cell layer was detached from each well using 2.5 % trypsine. Total RNAs were extracted from each sample by RNeasy Mini Kit (Qiagen, Duesseldorf, Germany), and 1 μg of total RNAs was reverse-transcribed to complementary DNA (cDNA) with Omniscript RT Kit (Qiagen). Glyceraldehydes-3-phosphate dehydrogenase (GAPDH) was chosen as the internal control gene. Table S1 list the primer sequence of analyzed genes and GAPDH. An ABI 7300 real-time PCR system (Applied Biosystems, Eugene, USA) with SYBR Green PCR master mix (Applied Biosystems, Eugene, USA) was used to carry out qRT-PCR and the conditions was 15 s at 90 °C, 60 s at 60 °C. Fluorescence intensity was recorded for 40 cycles [11]. The qRT-PCR assays were run in triplicate for each experiment.

Statistical analysis

Data are expressed as mean±SD. Differences between experimental groups are assessed by ANOVA. p-values <0.05 were considered statistically significant difference.

Characterizations of SCMs

To investigate the effects of TGF-β1 loading on the morphology and size of SCMs, SEM images of SCMs before and after TGF-β1 loading were displayed in Fig. 1. It is shown that SCMs had a uniform size of 1.9 ± 0.04 μm and smooth morphology without obvious cracks or wrinkles (Fig. 1a). Figure 1b displayed the size distribution of SCMs and TGF-β1/SCMs in water, which exhibited a unimodal and narrow peak at 5.4 μm and 5.7 μm, respectively (Fig. 1b). There were no significant differences in size distribution between SCMs and TGF-β1/SCMs. The size difference of SCMs at dry state and in solution could be contributed to microspheres swell after uptaken a large amount of water. As shown in Fig. 1c, after loading of TGF-β1, the morphology and size of TGF-β1/SCMs did not show significant difference in comparison with SCMs. ζ-potentials of the microspheres before and after TGF-β1 loading were compared in Fig. 1d. Before loading, SCMs had a ζ-potential of –25 mV because of the negative charge of sulfated chitosan. Due to the positive charge of TGF-β1, the ζ-potential of TGF-β1/SCMs significantly increased to -4 mV, indicating the successful loading of TGF-β1 into the microspheres (Fig. 1d).

Fig. 1

SEM image (a) of SCMs. (b) and (d) is the size distribution and ζ-potentials of SCMs before and after adsorption of TGF-1 respectively. The insert is the SEM image with higher magnification, (c) is the SEM image of SCMs after adsorption of TGF-1.

TGF-β1 release behavior

The effects of TGF-β1 released from a carrier on MSCs differentiation are dependent on many factors, among which its release behavior is considered as the crucial one [9]. A cumulative release profile was plotted to evaluate the release behavior of TGF-β1 from TGF-β1/SCMs (Fig. 2a). It was shown that TGF-β1 was released in a controlled manner, except for the slight burst release in the initial stage. The profile showed that the release speed was much higher before day 5, while 28 % of TGF-β1 were released from TGF-β1/SCMs. From day 5 to day 16, only a small quantity of TGF-β1 was gradually released from SCMs. Correlated to the total loading amount, 35 % of TGF-β1 was released up to 16 days.

Fig. 2

(a) Cumulative release profile of TGF-1 from SCMs. (b) The amount of the released TGF-1 after different storage time.

Bioactivity of TGF-β1 with the storage time

To evaluate the ability of SCMs to protect TGF-β1 bioactivity during the long-term storage, the amount of the freshly released TGF-β1 with the storage time was measured (Fig. 2b). After stored for 1 day, 24 % TGF-β1 with preserved bioactivity were released from SCMs after 4 h incubation. At the storage time of day 3 and day 5, no significant difference was observed in the released amount of the bioactive TGF-β1 compared to that of day 1. With the increase of the storage time, the amount of released TGF-β1 decreased gradually. 13 % of TGF-β1 could be still detected after stored in SCMs for 14 days. It indicates that the bioactivity of TGF-β1 loaded in SCMs could be preserved for much longer than that in medium, which could be attributed to the synergetic effect of microspheres and sulfated chitosan.

Cytotoxicity of SCMs

To determine the appropriate feeding concentration of SCMs in the following differentiation studies, the viabilities of MSCs cocultured with the different concentrations of SCMs were measured by MTT assay (Fig. 3). At day 1 and day 2, no obvious dependence of cell viabilities on the feeding concentration of SCMs was found. With increase of culture time, the cell viability gradually increased for all the samples with regardless of the concentration of SCMs. From day 3, the effects of the SCMs concentration on the cells viability became obvious. It is shown that with the increase of the feeding concentration of SCMs, the cells viability gradually decreased. The lowest value of cell viability was obtained when 10 μg/1.0 × 10⁴ cells SCMs was added. However, no obvious deterioration of cell viability was observed when the concentration of SCMs is lower than 2 μg/1.0 × 10⁴ cells, which was considered as the upper limit of the safe concentration in the following studies.

Fig. 3

Viability of MSCs cocultured with different amounts of SCMs. Data were expressed as mean±SD (n = 5). *Denotes p-values <0.05.

Cell proliferation

In order to avoid the impact of interaction between SCMs and cells, transwells model was used to determine the efficiency of released TGF-β1 on cells behaviors. Firstly, the proliferation behaviors of MSCs alone (Blank) or cocultured with SCMs, TGF-β1 and TGF-β1/SCMs were assayed respectively by counting cells number with culture time (Fig. 4). For the groups of Blank and SCMs, the number of cells increased gradually with the culture time during the 14-days cultivation. For the groups of TGF-β1 and TGF-β1/SCMs, a significant increase of cells number could be found from day 3 to day 7, but no further increase at day 14. At all the time point, the TGF-β1 released from TGF-β1/SCMs showed almost the same effect on the proliferation of MSCs cocultured with free TGF-β1. At day 14, the number of MSCs in Blank group showed the highest value in comparison to the groups of TGF-β1 and TGF-β1/SCMs.

Fig. 4

Number of MSCs cultured in different conditions for different days. Data were expressed as mean±SD (n = 4). *Denotes p-values <0.05.

GAGs secretion

As one of the main components of cartilage ECM, GAGs is frequently selected as criteria to judge the degree of MSCs differentiation into chondrocytes [31]. As shown in Fig. 5, the amounts of GAGs secreted by MSCs cultured in different conditions were measured. For all the groups, the amounts of GAGs increased gradually with the prolongation of culture time. At day 14, the highest level of GAGs production (up to 50 mg/mL GAGs in the medium) was obtained in the group of TGF-β1/SCMs, the value of which is almost twice of that of blank group. Additionally, TGF-β1/SCMs had a significant impact on the production of GAGs in comparison to free TGF-β1 during culture period. Interestingly, the addition of TGF-β1 in free model had no obvious effect on the GAGs secretion compared to the groups of blank and SCMs. The similar results were also obtained from the results of GAGs staining by Alcian blue (data not shown). The promoting effect of TGF-β1/SCMs on GAGs secretion proved that, the chondrogenic differentiation of MSCs was preceded more deeply under the stimulating of released TGF-β1, indicating the protection of SCMs.

Fig. 5

Quantitative analysis of GAGs secreted by MSCs cultured in different conditions. Data were expressed as mean±SD (n = 3). *Denotes p-values <0.05.

Expression of chondrogenesis-related genes

The expression of cartilage-related proteins such as COL II and aggrecan is indispensable to realize the restoration of damaged cartilage [5]. To assay the chondrogenesis of MSCs cocultured at different conditions, the expression of chondrogenesis-related genes such as collagen II and aggrecan was determined by qRT-PCR. As shown in Fig. 6, after 14-days cultivation, both collagen II and aggrecan were highly expressed at the groups with the additional of TGF-β1 compared to the others (Fig. 6). As shown in Fig. 6a, the medium containing TGF-β1/SCMs had a higher level of collagen II expression in comparison to that containing free TGF-β1. In the TGF-β1 free groups, the presence of SCMs had no obvious effect on stimulating the expression of collagen II in comparison to the blank group. Similar to the results of collagen II, aggrecan expression was also up-regulated by the presence of TGF-β1/SCMs (Fig. 6b).

Fig. 6

qRT-PCR analysis for collagen II (a) and aggrecan (b) gene expression of MSCs after 14 days culture. Results were presented as relative quantity compared to GAPDH gene expression. Data are expressed as mean±SD (n = 3). *Denotes p-values <0.05.