Poly(γ-glutamic acid) induced homogeneous mineralization of the poly(ethylene glycol)-co-2-hydroxyethyl methacrylate cryogel for potential application in bone tissue engineering

Abstract

Methacrylated poly(γ-glutamic acid) (mPGA) was introduced to the spongy poly(ethylene glycol)-co-2-hydroxyethyl methacrylate (PEG-HEMA) cryogels. A suitable amount (<0.1% w/v) of poly(γ-glutamic acid) (γ-PGA) in cryogels was found to promote homogeneous mineralization, while higher concentration of γ-PGA inhibited the mineralization. The deposited minerals mainly consisted of hydroxyapatite, carbonated apatite, and calcium phosphate, and were independent on the concentration of γ-PGA. The mineralized cryogels can support adhesion, viability, and migration of rat bone marrow stromal cells (rMSCs). The γ-PGA/HEMA-PEGDA cryogels showed their great potential as promising scaffolding materials in bone tissue engineering.

---

Introduction

Natural bone is a composite of collagen, carbonated apatite crystals, and cells. The hierarchically organized interconnected porous structure of spongy bone allows nutrient/metabolite transportation, cell migration, ECM deposition, and so on. In bone tissue engineering, it is critical to develop a scaffold that chemically and structurally mimics the native extracellular matrix of bone tissue. Mineral–polymer composite scaffold with interconnected macroporous structure may be a promising candidate for this purpose.

Recently, cryogel^1–3 has received increasing attention in tissue engineering due to its high porosity and interconnected macroporous structure. Varghese⁴ investigated poly(ethylene glycol) cryogels as scaffolds for cartilage tissue engineering. Their study indicated that the cryogels can support adhesion, viability, proliferation, and biosynthetic activity of seeded chondrocytes. The microarchitecture of PEG cryogel can also affect the osteogenic differentiation of human mesenchymal stem cells (hMSCs).⁵ Spongy cryogels can promote osteogenic differentiation to a greater extent than their columnar counterparts. Kumar⁶ prepared alginate/gelatin cryogels using glutaraldehyde or 1-ethyl-3(3-dimethylaminopropyl)-carbodiimide as crosslinkers, and found the type of crosslinker may affect the physiochemical properties of matrices and lead to different cell behaviors.

Stemming from the fact that anionic proteins can regulate the mineral phase deposition in natural bone,³⁰ the incorporation of anionic groups, such as carboxyl group, onto polymeric substrate has become a common approach to inducing mineralization on the surface. Chappard⁷ discovered that modification of poly(2-hydroxyethyl methacrylate) (pHEMA) with carboxymethyl groups could significantly increase the deposition of calcium. Bertozzi⁸ studied the effect of carboxyl groups on the mineralization of pHEMA hydrogels. Carboxyl groups may act as calcium binding-sites to initiate the heterogeneous nucleation and high-affinity growth of calcium apatite on the gel surface and extensive calcification inside the gel. A later study on substrates containing biomimetic mineral-nucleating amino acid ligands revealed that the structure and density of the ligands could effectively control the morphology and crystallinity of the mineral.⁹

Glutamic acid and aspartic acid residues are widely found with extraordinary abundance in non-collagenous proteins of bone tissue and take part in the bio-function of such proteins. For example, consecutive glutamyl residues play a major role in bone sialoprotein as a mineral nucleating agent in vivo.¹⁰ Similarly, crosslinked poly(γ-glutamic acid) (γ-PGA) has proved to induce the heterogeneous nucleation of hydroxyapatite in a simulated body environment.¹¹

In this work, the role of γ-PGA in mineralization of the spongy pHEMA cryogel scaffold was investigated; and the effect of concentration of γ-PGA on mineralization, as well as the adhesion, viability, and migration of rat bone marrow stromal cells (rMSCs) in the scaffolds were evaluated in detail.

Materials and experimental

Materials

Poly(ethylene glycol) diacrylate (PEGDA-3.4k, molar mass 3400 g mol⁻¹) and methacrylated γ-PGA (mPGA, nominal molar mass 1.0 × 10⁶ g mol⁻¹) were synthesized as previously reported,^12,13 with degrees of substitution of acryloyl or methacryloyl groups of 97.5% and 10.0% respectively. Simulated body fluid (SBF), 40 mM Ca²⁺/24 mM HPO₄²⁻ immersion solution, and phosphate-buffered saline (PBS) were prepared following established protocols.^10,14 2-Hydroxyethyl methacrylate (HEMA, 98%) and tetramethylethylenediamine (TEMED) were purchased from Aladdin Industrial Corporation. Ammonium persulphate (APS) was purchased from Guangzhou Chemical Reagent Factory (Guangzhou, China), and o-cresolphthalein complexone and vanadium ammonium molybdate were purchased from Shanghai Chemical Reagent Company (Shanghai, China). Papain and cysteine were purchased from Sigma-Aldrich Co. (USA). Live/Dead Viability/Cytotoxicity Kit and Quant-iT Picogreen dsDNA Assay kit were purchased from Invitrogen/Life Technologies (USA).

Growth medium containing high-glucose Dulbecco's modified Eagle medium (DMEM) (Hyclone, USA), 10% of fetal bovine serum (FBS) (PAA, Germany), and 1% of penicillin–streptomycin (Tianjin Haoyang Biological Manufacture Co., LTD, China) was used in the cell culture. rMSCs were isolated from the femurs and tibias of four week-old male Sprague Dawley rats as described elsewhere.¹⁵ Cells were cultured in growth medium. Cells at forth passage were used for experiments.

Preparation of cryogels

Cryogels were prepared in cylindrical polypropylene moulds (3 mm in diameter) as reported previously.^4,5 Briefly, a solution of 0.5 M HEMA, 5% (w/v) of PEGDA_3.4k, and predetermined amount of mPGA in PBS was chilled to 4 °C. To this chilled solution, 0.1% (v/v) of TEMED and 0.5% (w/v) of APS were added and vortexed. Immediately, 40 μL of the solution was transferred to each mold, and polymerized at −16 °C for 18 h. Following polymerization, the cryogels were thawed at room temperature and washed with deionized water for several times to remove the residual monomer/precursor and initiator (Scheme 1).

Scheme 1 The structures of HEMA, PEGDA & mPGA (a) and the process of cryogel preparation (b).

Mineralization

Cryogels were sterilized by immersion in 70% ethanol for 12 h, washed in sterile PBS for three times, and then immersed in sterile PBS for 24 h. Before mineralization, the sterile cryogels were immersed in 40 mM Ca²⁺/24 mM HPO₄²⁻ solution¹⁰ for 3 h, and rinsed with PBS for three times. These cryogels were then incubated in SBF for 32 h, and the SBF was removed and replenished by fresh SBF every 16 h. This 32 h process was considered one immersion cycle. After the first cycle of mineralization, the cryogels were washed with PBS for three times and mineralized for the second cycle.

Cryogels or mineralized cryogels were thoroughly rinsed with running DI water, and lyophilized before further characterization. The mineralized cryogels were presented as M-Px, where x stands for the percentage content of mPGA in the polymerization formulation; and the non-mineralized cryogels were termed Non-M-Px in the following context.

Fourier transform infrared spectroscopy (FTIR)

The chemical structure of the cryogels with different concentration of γ-PGA were characterized by attenuated total reflectance Fourier transform infrared spectroscopy (ATR-FTIR). The spectra were recorded using a Bruker Vertex 70 spectrometer (Bruker, Germany) over a range of 4000–400 cm⁻¹. The mineralized cryogels was also examined following the same method.

Equilibrium water content (EWC)

EWC of cryogels or mineralized cryogels (n = 3) were measured gravimetrically. Cryogels or mineralized cryogels with saturated amount of water were weighted and lyophilized. EWC was calculated according to the following equation:

EWC% = [(m_w − m_d)/m_w] × 100 (1)

where m_w is the mass of water-saturated cryogel or mineral–cryogel composite, and m_d is the mass of lyophilized cryogel or mineral–cryogel composite.

Optical microscopy

Paraffin-embedded cryogels were sectioned with thickness of 10 μm on a Leica RM2245 microtome following reported protocol.⁵ The non-mineralized cryogel sections was stained with crystal violet, and mineralized cryogel was stained with alizarin red S. The sections were observed on microscope (Leica DM2500M, Germany) under bright field, and the pore size and the thickness of the pore wall were measured manually on the images of the sections using an associated image analyzing software (Image-Pro plus), three representative fields were measured for each sample.

Scanning electron microscopy-energy dispersive spectroscopy (SEM-EDS)

The microstructural topology and composition of cryogels and mineralized cryogels were analyzed by scanning electron microscopy (SEM) with associated energy dispersive spectroscopy (EDS) (Thermal field emission environmental SEM-EDS-EBSD, Quanta 400 FEG, FEI/OXFORD/HKL, France). Cryogel or mineral–cryogel composite was flash-frozen in liquid N₂ and lyophilized. The surface and cross-section were sputter-coated with gold prior to observation.

Porosity

The porosity of the cryogels was approximated according to Archimedes's principle using a gravity bottle.^6,16 The mass of lyophilized cryogel (m_dry) and the mass of the gravity bottle filled with cyclohexane (m₁) were measured. Then, the sample was submerged in cyclohexane in the bottle under reduced pressure to remove residual gas in the cryogel, and the total mass of the bottle was recorded as m₂. The cryogel was then taken out and weighed (m_wet). Measurements were done in triplicates for each sample and the porosity was calculated via the following equation:

Ca and P content

Mineralized cryogels (n = 3 per group) were immersed in 1.0 mL of 0.5 M HCl and vigorously vortexed overnight at 4 °C. Then the cryogels were homogenized and the suspension was adjusted to 10 mL using ultrapure water and vigorously vortexed for two hours at 37 °C, and then centrifuged. The absorbance of calcium–cresolphthalein complexone in the supernatant at 570 nm was measured to derive the content of calcium.^17–20 The content of phosphorus was determined similarly at 355 nm using vanadium ammonium molybdate.^21,22

X-ray diffraction

Mineralized cryogels were homogenized and lyophilized. The dry powder was compressed into thin film and analyzed on a Rigaku X-ray diffractometer (Cu Kα1), with beam energy of 36 kV, beam current of 30 mA, rotating rate of 4° min⁻¹, and angular range (2θ) from 5° to 80°.

Cell culture

Sterile cryogels were immersed in growth medium at 37 °C, 5% CO₂ for 24 h prior to cell seeding. Prior to cell seeding, cryogels were taken out from the medium and dried in air under sterile conditions for 60 min, then sterilized again under UV light (253.7 nm, 20 W) for 30 min. 3 × 10⁵ cells (rMSCs, passage 4) in 30 μL of growth medium were then seeded on top of the cryogel. The constructs were then incubated without additional medium at 37 °C, 5% CO₂ for four hours to allow cell attachment. Following this, these cryogels were incubated in 2 mL of growth medium for 24 h, 7 days, and 21 days at 37 °C, 5% CO₂, and the growth medium was changed every 48 hours. Samples were collected for analysis at different time points as detailed below.

Cell viability

After 24 h culture, the constructs were cut into thin slices, washed with PBS, and treated with the Live/Dead Viability/Cytotoxicity Kit,¹⁰ which contains 0.5 μL of calcein-AM and 2.0 μL of ethidium homodimer-1 in 1.0 mL of DMEM. After 30 min of incubation, the samples were rinsed with PBS and images were obtained on a fluorescence microscope (OLYMPUS IX71, OLYMPUS, Japan) to distinguish live cells (stained green by calcein-AM) from dead cells (stained red by ethidium homodimer).

Cell-laden constructs (n = 3) were harvested at 0, 7, and 21 days for determination of the content of DNA.⁵ Constructs were frozen at −80 °C, lyophilized, weighed, and homogenized. The homogenized constructs were digested in papain solution (2.375 U mL⁻¹ of papain, 10 mM of cysteine in PBE buffer) for 16 h at 60 °C. DNA content was then measured using Quant-iT Picogreen dsDNA Assay kit according to manufacturer's guidance.

Immunohistochemical staining

Constructs were collected after three-week's culture and fixed in 4% paraformaldehyde for 24 hours. The fixed constructs were dehydrated, embedded in paraffin and cut into 10 μm thick sections. Sections were stained for F-actin using Actin-Tracker Green (Beyotime, China), then the sections were immersed in DAPI solution (1 μg mL⁻¹) for 1 min and washed with PBS for 30 min, and observed under fluorescent microscope.

Statistical analysis

All quantitative data were presented as mean ± SD. T-test was performed to assess the statistical significance between groups. Values of p < 0.05 were accepted as statistically significant. Significance level was presented as either * (p < 0.05) or ** (p < 0.01).

Result and discussion

Chemical & morphological analysis of cryogels

The incorporation of γ-PGA into the backbone of cryogel was confirmed by ATR-FTIR (Fig. 1a). A band at ∼1600 cm⁻¹ for stretching vibration of C=C bond from mPGA disappeared in the spectrum of cryogels, indicating the consumption of C=C bond during cryogelation. At the same time, peaks for amide I (1650 cm⁻¹) and amide II (1550 cm⁻¹) appeared in the spectrum of cryogel matrix with γ-PGA. The disappearance of the peak for C=C bond and appearance of the amide peaks suggested that γ-PGA was successfully incorporated into the cryogel matrix via co-valent bond.

Fig. 1 ATR-FTIR spectra (a) and XRD spectra (b) of mPGA and cryogels before and after mineralization.

The interconnectivity of the porous structure of the cryogels were examined under SEM and optical microscope (Fig. 2). Cryogels with different concentrations of γ-PGA (0, 0.01, 0.1, 0.25, 0.5% w/v) have similar average pore sizes with diameters about 50 μm (Table S1 and Fig. S1†). The thickness of the wall of the pores was found to be in the range of 11.9–19.3 μm, while no statistical difference was observed at different γ-PGA concentrations. Porosity of the cryogels with different concentration of γ-PGA is similarly above 80% (Fig. S2†). These findings suggested that the concentration of γ-PGA did not significantly affect the pore parameters of cryogel.

Fig. 2 Optical microscope images of crystal violet stained cryogel sections (a, b, and c) and scanning electron microscopy images of mineralized cryogels (inner: d, e, and f; surface: g, h, and i; ×500) with different concentration of mPGA (0, 0.1, 0.5% w/v).

After mineralization, the interconnectivity of cryogels was maintained (Fig. 2) with very little variation in the structure of the pores as compared with that of non-mineralized cryogel (Table S1, Fig. S1 and S2†). ATR-FTIR analysis was performed on the surface of the mineralized cryogels (Fig. 1a). After mineralization, the intensity of peaks for the backbone of the polymeric matrix scaffold (P0 and P0.5) decreased greatly, while the bands of PO₄³⁻ groups (ν₁ ∼ 963 cm⁻¹, ν₃ ∼ 1036 cm⁻¹) appeared. Additionally, bands at ∼870 cm⁻¹ (ν₂) due to CO₃²⁻ can be found,^23,24 suggesting that CO₃²⁻ groups have been incorporated into the minerals. In the XRD patterns, significant intensity of peaks at (002) and peaks at (211), (112), (300) in the mineralized cryogels indicated that the minerals in the cryogels have similar crystal lattice to HA standard (ICDD 9-432).²⁵

Due to its strong hydrophilicity, incorporating γ-PGA into the cryogel matrix could lead to higher EWC. The EWCs of the cryogels were similarly higher than 93% at concentration of γ-PGA lower than 0.25%; however, the EWC decreased greatly when the concentration of mPGA reached 0.25% w/v, this could probably be attributed to the increased crosslink density due to the increased amount of methacryloyl groups brought by γ-PGA. During the mineralization, carboxyl groups were consumed in chelation with calcium ion and induction of minerals, and the minerals formed in this process could in turn reduce the accessibility of carboxyl groups to water molecules, i.e. displaying a decrease in EWC.

The mineral deposited in the γ-PGA-modified cryogel was stained using Alizarin red S (Fig. 3). With 0.1% (w/v) of γ-PGA, the whole section was stained red, indicating the minerals were deposited homogenously in the cryogel; however, cryogels with γ-PGA concentration of 0, 0.01, and 0.5% showed relative lighter staining in the center of the sections. Minerals on the surface or inside the cryogels may be originated from either the precipitation from the metastable solution or the growth of the crystal nucleus induced by glutamyl residues on the backbone of cryogel. The staining on the surface of cryogel matrix with 0 or 0.01% (w/v) of γ-PGA probably mainly came from the precipitation of the metastable solution and the growth of such precipitation; once the surface is covered by these precipitates and nucleated minerals, less minerals formed inside cryogels, which was later confirmed by the SEM observation (Fig. 5). Meanwhile, the minerals which covered the surface of cryogels can hinder the mineral deposition in the interior space of the cryogels. As a result, less minerals were found inside the cryogels. However, as the concentration of γ-PGA increased, the mineral crystal nucleation and the growth of the crystal were promoted. Mineralization not only took place on the surface of the cryogel but also in the interior space of the matrix. Here, carboxyl groups on the cryogel matrix acted as Ca²⁺ binding-site and induced the heterogeneous nucleation of hydroxyapatite.^7–9,11 The amount and adhesion strength of minerals in the cryogels were governed by the density of carboxyl groups on the surface of the matrix. Kawashita²⁶ has demonstrated the catalytic effect of carboxyl groups on apatite nucleation, and acceleration of apatite nucleation from released Ca²⁺ ions. In this work, the apatite nucleation may also be catalyzed and accelerated by the carboxyl groups in the cryogel matrix following the similar process. After being treated with 40 mM Ca²⁺/24 mM HPO₄²⁻ solution, a large number of the Ca²⁺ ions was combined with the carboxyl groups, which may develop high local concentration of Ca²⁺ ions and facilitate the nucleation of apatite in situ. With increasing concentration of γ-PGA, the amount of minerals firstly increased slightly and then decreased, which was in accordance with the quantification of Ca and P (Fig. 4a and b). Comparable amount of mineral was found in each group with mPGA concentration ranging from 0–0.1%. This could probably be attributed to the large amount of minerals from the precipitation of the metastable immersion solution onto the superficial area of the cryogel.

Fig. 3 Alizarin red S staining of mineralized cryogels.

Fig. 4 Ca & P content in the cryogels after mineralization. (a) Ca content & (b) P content from UV measurement; (c) SEM-EDS images.

It has been reported that excessive carboxyl groups may inhibit the nucleation of mineral in the case of polyaspartic acid.²⁷ In this study, maximum Ca and P contents were obtained at concentration of γ-PGA of 0.1% w/v, and were significantly higher than those in cryogels with 0.5% w/v of γ-PGA (P < 0.05). In other words, mineralization was inhibited at higher concentration of γ-PGA than 0.1% w/v, and the minerals were found mainly located on the surface of the cryogels with 0.1% or higher concentration of γ-PGA due to the growth of precipitation.

The Ca/P ratios of the mineral in the cryogels were quantified to be 1.63, 1.66, and 1.64 for P0, P0.1, and P0.5, respectively, which were close to that of hydroxyapatite (1.67).²⁸ However, the Ca/P ratios on the surface of the mineralized cryogels as determined via SEM-EDS (Fig. 4c) were 1.48, 1.96, and 1.57 for P0, P0.1, and P0.5 cryogels, respectively. The discrepancy may come from the fact that EDS only detects the local composition of mineral, which may be greatly influenced by the presence of calcium carbonate.

More minerals were found on the surface than in the interior space of the cryogels (Fig. 5). Different morphologies of the minerals could be observed in different regions of the same cryogel; this could be attributed to the different stages of the mineral growth, which has little correlation with the amount of carboxyl groups but highly depends on the local concentration and composition of the SBF. Similar phenomenon had been reported by Mooney,²⁹ who discovered that the Ca/P ratio and crystal morphology on the surface of poly(α-hydroxy ester) were dependent not on the polymer surface characteristics but on the composition of the SBF. In this case, the cryogels were soaked in 40 mM Ca²⁺/24 mM HPO₄²⁻ solution prior to immersion in SBF to increase to the regional concentration of calcium and phosphate ions inside the cryogels, and consequently to accelerate the growth of the minerals. This process may change the formulation of SBF in the mineralization, and then influenced the chemical composition and morphology of the resultant mineral phase.³⁰ With the aid of 40 mM Ca²⁺/24 mM HPO₄²⁻ solution, the crystals growth not only originated from the amorphous minerals which was induced by the carboxyl groups on the surfaces of the scaffold, but also from the precipitate from the metastable solution.

Fig. 5 SEM images of mineralized cryogels.

In general, we confirm that the minerals in the cryogels have similar composition, mainly consisting of hydroxyapatite, carbonated apatite, and calcium phosphate, and were independent on the concentration of γ-PGA. However, the amount of minerals were closely related to the concentration of γ-PGA. The amount of minerals increased with the concentration of γ-PGA (<0.1% w/v) initially, and decreased when the concentration of mPGA was above 0.1% w/v.

In order to further confirm the role of γ-PGA in mineralization, cryogels with different concentrations of γ-PGA were immersed in SBF for four weeks without pre-soaking in the 40 mM Ca²⁺/24 mM HPO₄²⁻ solution, the SBF was changed with every other day to ensure sufficient supply of ions. Under this slow process of mineralization, tiny amount of Ca and P were detected in the mineralized cryogels via colorimetric assay (Fig. S3a and b†), even though no P was detected by SEM-EDS (data not shown). The amount of Ca increased slightly with the increasing concentration of γ-PGA. There was no significant difference in P content between each group. However, more mineral nucleation was found on the pore wall of cryogel with 0.1% w/v of γ-PGA as evidenced by SEM (Fig. S3c†) and OM observations (Fig. S3d†) as compared with those with 0% w/v and 0.5% w/v of γ-PGA. This phenomenon further reveals that, at the early stage of mineralization, suitable amount (<0.1% w/v) of γ-PGA promotes the mineral nucleation, while larger amount (>0.1% w/v) of γ-PGA binds to and stabilizes the pre-nucleation clusters, forming loosely packed, diffusive structures that slowly aggregate and densify, and finally inhibits mineral nucleation, as described by Sommerdijk.²⁷

Cell viability & migration

In the M-P0.1 cryogel, cells mainly adhered to surface of the cryogels after 24 h cell culture, and there were more live cells (stained green) than dead ones (stained red) (Fig. 6c). This indicated that mineralized cryogels are cytocompatible. After 7 days' culture, no statistical differences between the DNA content in M-P0, M-P0.1, and Non-M-P0.1 cryogels was observed (Fig. 6b). However, after 21 days' cell culture, DNA content in “M-P0.1” group was significantly higher than that in “M-P0” group (P < 0.05) (Fig. 6a), suggesting that minerals on the surface of the cryogels can better support the long term culture of rMSCs, this may be attributed to the osteoconductive and osteoinductive properties of apatites.³⁰ However, if we look at each group, the cells did not proliferate during the time scale of study (Fig. 6a). This could be because of the high seeding density of the cells (10⁷ cells per mL) and the small pore size of the cryogels (∼50 μm) that hinders the proliferation of rMSCs. Future work on cryogels with pore size around 300–500 μm should be explored to achieve better cell response.

Fig. 6 (a) DNA content of cells seeded on mineralized cryogels. (b) DNA content in cryogels after 7 days' culture. (c) Live/dead assay for cells seeded on mineralized cryogel. Statistical differences are represented by * between each group, p < 0.05. (d) Cytoskeleton staining of cells seeded on mineralized cryogel (M-P0.1) after 21 days' culture.

Cytoskeleton staining was performed to evaluate the migration of cells from the surface to the interior space of the cryogels (Fig. 6d). After 21 days' culture, more cells in the M-P0.1 cryogel were observed as compared with those in other groups. The staining of F-actin (stained green) also indicated that cells preferentially adhered to the cryogel where minerals were fully developed. After 21 days' culture, a larger amount of cells migrated into and were distributed uniformly inside the M-P0.1 cryogels. However, cells were mainly distributed on the top of “M-P0” and “Non-M-P0.1” cryogels. This phenomenon indicated that minerals can not only support the adhesion but also the migration of cells. Take account of the Alizarin red S staining of M-P0.1 cryogel (Fig. 3), it seems that the uniform distribution of minerals may have positive consequences on biological response of cells in terms of better viability and benefit the migration of cells to bulk of the matrix.

Conclusion

In summary, polymeric cryogels with interconnected porous structure and more than 80% of porosity were successfully prepared. Suitable amount (<0.1% w/v) of γ-PGA can promote uniform mineralization due to the chelation of carboxyl groups to Ca²⁺ ions. However, γ-PGA at concentration higher than 0.1% w/v may inhibit the mineral nucleation. Mineralization in SBF was accelerated when the cryogels were previously immersed in 40 mM Ca²⁺/24 mM HPO₄²⁻ solution. Under this mineralization process, the minerals in the cryogels display similar composition and mainly consist of hydroxyapatite, carbonated apatite, and calcium phosphate. Meanwhile, study found that mineralized cryogels with 0.1% w/v of γ-PGA better supported adhesion, viability, and migration of rMSC. Here, γ-PGA/HEMA-PEGDA cryogels can be a promising candidate in bone tissue engineering.

Acknowledgements

This work is financially supported by the Program for New Century Excellent Talents in University (Grant no. NCET-09-0818) of the Ministry of Education of China, the Natural Science Foundation of China (Grant no. 21004080), the Program for Industry, University & Research Institute Collaboration of Guangdong Province (Grant no. 2012B091100452), and the Science and Technology Planning Project of Guangdong Province (no. 2011A06090101).

Notes and references

1. V. I. Lozinsky, I. Y. Galaev, F. M. Plieva, I. N. Savina, H. Jungvid and B. Mattiasson, Trends Biotechnol., 2003, 21, 445.
2. F. M. Plieva, I. Y. Galaev and B. Mattiasson, J. Sep. Sci., 2007, 30, 1657.
3. S. A. Bencherif, R. W. Sands, D. Bhatta, P. Arany, C. S. Verbeke, D. A. Edwards and D. J. Mooney, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 19590.
4. Y. Hwang, N. Sangaj and S. Varghese, Tissue Eng., Part A, 2010, 16, 3033.
5. A. Phadke, Y. S. Hwang, S. H. Kim, S. H. Kim, T. Yamaguchi, K. Masuda and S. Varghese, Eur. Cells Mater., 2013, 25, 114.
6. A. Tripathi, T. Vishnoi, D. Singh and A. Kumar, Macromol. Biosci., 2013, 13, 838.
7. R. Filmon, F. Grizon, M. F. Basle and D. Chappard, Biomaterials, 2002, 23, 3053.
8. J. Song, E. Saiz and C. R. Bertozzi, J. Am. Chem. Soc., 2003, 125, 1236.
9. J. Song, V. Malathon and C. R. Bertozzi, J. Am. Chem. Soc., 2005, 127, 3366.
10. A. Phadke, C. Zhang, Y. S. Hwang, K. Vecchio and S. Varghese, Biomacromolecules, 2010, 11, 2060.
11. A. Sugino, T. Miyazaki and C. Ohtsuki, J. Mater. Sci.: Mater. Med., 2008, 19, 2269.
12. L. Cai and S. F. Wang, Polymer, 2010, 51, 164.
13. W. Zeng, W. K. Hu, H. Li, Y. H. Jing, H. Kang, Q. Jiang and C. Zhang, Chin. J. Polym. Sci., 2014, 32, 1507.
14. A. Oyane, H. M. Kim, T. Furuya, T. Kokubo, T. Miyazaki and T. Nakamura, J. Biomed. Mater. Res., Part A, 2003, 65, 188.
15. K. S. Kim, J. Y. Lee, Y. M. Kang, E. S. Kim, G. H. Kim, S. Dal Rhee, H. G. Cheon, J. H. Kim, B. H. Min, H. B. Lee and M. S. Kim, Biomaterials, 2010, 31, 1104.
16. A. Tripathi and A. Kumar, Macromol. Biosci., 2011, 11, 22.
17. H. Zhou and T. Z. Mao, J. Environ. Health, 1999, 16, 228.
18. B. Qi and T. Yin, Phys. Test. Chem. Anal., Part B, 1998, 34, 406.
19. H. Shin, J. S. Temenoff, G. C. Bowden, K. Zygourakis, M. C. Farach-Carson, M. J. Yaszemski and A. G. Mikos, Biomaterials, 2005, 26, 3645.
20. J. van den Dolder, G. N. Bancroft, V. I. Sikavitsas, P. H. M. Spauwen, J. A. Jansen and A. G. Mikos, J. Biomed. Mater. Res., Part A, 2002, 64, 235.
21. H. Yang and L. Zhang, J. Shanxi Univ. Sci. Technol., 2007, 25, 71.
22. Z. J. Yu, T. Z. Li and Y. Liu, Chem. Anal. Meterage, 2001, 10, 10.
23. A. Slosarczyk, Z. Paszkiewicz and C. Paluszkiewicz, J. Mol. Struct., 2005, 744, 657.
24. M. Vignoles, G. Bonel, D. W. Holcomb and R. A. Young, Calcif. Tissue Int., 1988, 43, 33.
25. M. Sadat-Shojai, M. Atai and A. Nodehi, J. Braz. Chem. Soc., 2011, 22, 571.
26. M. Kawashita, M. Nakao, M. Minoda, H. M. Kim, T. Beppu, T. Miyamoto, T. Kokubo and T. Nakamura, Biomaterials, 2003, 24, 2477.
27. F. Nudelman, K. Pieterse, A. George, P. H. H. Bomans, H. Friedrich, L. J. Brylka, P. A. J. Hilbers, G. de With and N. A. J. M. Sommerdijk, Nat. Mater., 2010, 9, 1004.
28. F. Z. Cui, Biomineralization, Tsinghua University Press, Bejing, 2007, vol. 05, pp. 17–18.
29. W. L. Murphy and D. J. Mooney, J. Am. Chem. Soc., 2002, 124, 1910.
30. J. D. Kretlow and A. G. Mikos, Tissue Eng., 2007, 13, 927.

---

Footnote

† Electronic supplementary information (ESI) available: Porosity, pore size and distribution, pore morphology. See DOI: 10.1039/c4ra15893h

---