Materials Science and Engineering C 51 (2015) 139–147
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Injectable hydrogels derived from phosphorylated alginic acid calcium complexes Han-Sem Kim, Minsoo Song ⁎, Eun-Jung Lee, Ueon Sang Shin ⁎ Department of Nanobiomedical Science & BK21 PLUS NBM Global Research Center for Regenerative Medicine, Dankook University, Chungnam, Cheonan 330-714, Republic of Korea Institute of Tissue Regeneration Engineering (ITREN), Dankook University, Chungnam, Cheonan 330-714, Republic of Korea
a r t i c l e
i n f o
Article history: Received 26 November 2014 Received in revised form 28 January 2015 Accepted 23 February 2015 Available online 25 February 2015 Keywords: Phosphorylated alginic acid calcium complexes In vivo injectable hydrogel Cell and drug delivery
a b s t r a c t Phosphorylation of sodium alginate salt (NaAlg) was carried out using H3PO4/P2O5/Et3PO4 followed by acid–base reaction with Ca(OAc)2 to give phosphorylated alginic acid calcium complexes (CaPAlg), as a water dispersible alginic acid derivative. The modified alginate derivatives including phosphorylated alginic acid (PAlg) and CaPAlg were characterized by nuclear magnetic resonance spectroscopy for 1H, and 31P nuclei, high resolution inductively coupled plasma optical emission spectroscopy, Fourier transform infrared spectroscopy, and thermogravimetric analysis. CaPAlg hydrogels were prepared simply by mixing CaPAlg solution (2 w/v%) with NaAlg solution (2 w/v%) in various ratios (2:8, 4:6, 6:4, 8:2) of volume. No additional calcium salts such as CaSO4 or CaCl2 were added externally. The gelation was completed within about 3–40 min indicating a high potential of hydrogel delivery by injection in vivo. Their mechanical properties were tested to be ≤6.7 kPa for compressive strength at break and about 8.4 kPa/mm for elastic modulus. SEM analysis of the CaPAlg hydrogels showed highly porous morphology with interconnected pores of width in the range of 100–800 μm. Cell culture results showed that the injectable hydrogels exhibited comparable properties to the pure alginate hydrogel in terms of cytotoxicity and 3D encapsulation of cells for a short time period. The developed injectable hydrogels showed suitable physicochemical and mechanical properties for injection in vivo, and could therefore be beneficial for the field of soft tissue engineering. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Injectable hydrogels with biodegradability have emerged as promising biomaterials for a wide variety of biomedical applications in the field of tissue engineering. Specifically, they allow encapsulation of biologically active molecules (drugs and genes) and cells, and patient friendly delivery into the body without painful surgery, due to their in situ formability [1–4]. Given their unique characteristics, injectable biodegradable hydrogels have been actively explored as drug delivery systems for controlled release of bioactive agents [5–7] and temporary extracellular matrices for tissue engineering [8–11]. Alginate is well known for chelating with divalent cations such as Ca2+, Mg2+, and Ba2+ to form hydrogels [13,14]. Alginate molecule is a linear copolymer covalently coupled through glycosidic linkages between two monomeric saccharide epimer units such as β-D mannuronic acid (M) and R-L-guluronic acid (G). Moreover, the molecular structure of the copolymer shows characteristic three-dimensional arrangements to create the homopolymeric blocks such as poly guluronate
⁎ Corresponding authors. Tel.: +82 41 550 3691; fax: +82 41 559 7911. E-mail addresses: [email protected] (M. Song), [email protected] (U.S. Shin).
http://dx.doi.org/10.1016/j.msec.2015.02.031 0928-4931/© 2015 Elsevier B.V. All rights reserved.
(GG) and poly mannuronate (MM). The hydrogel formation is driven by the interchain cross-linking by the cationic coordinations with carboxylate groups in two facing GG blocks (the egg-box model) [12]. Biomedical applications of alginate hydrogels are profound because of its biocompatibility and minimal toxicity [15,16]. Alginate hydrogels prepared by a traditional external setting technique using calcium salts such as CaCl2 and CaSO4 have been utilized in cell encapsulation (or immobilization) [17], drug delivery [18], wound healing dressing [19], and other applications in tissue engineering [20]. The significance of alginate as biomaterials have propelled research on inherent property enhancement and new property introduction through chemical modification [21,22]. For example, Grøndahl et al. have prepared phosphorylated alginate derivatives (PAlg) for the first time using the urea/phosphate method to evaluate the ability to induce hydroxyapatite nucleation and growth [23]. Thorough NMR studies have revealed that phosphorylation of M residues occurred mostly on the C-3 equatorial hydroxyl groups of the polysaccharide ring. However, Ca-crosslinking of phosphorylated alginates by externally added CaCl2 to form PAlg-hydrogel has failed. The reduction of molecular weight of phosphorylated alginic acid and conformational changes on phosphorylation were considered as reasons. Since calcium ions show high affinity for phosphates and carboxylates and do not cross-link with other phosphorylated alginic acid molecules (PAlg) [21], we envisioned that
140
H.-S. Kim et al. / Materials Science and Engineering C 51 (2015) 139–147
phosphorylated alginic acid calcium complexes (CaPAlg) as a new alginate derivative could be prepared without self-gelation and be efficiently utilized to form hydrogels by simple mixing with alginic acid sodium salts. Here the CaPAlg complexes provide Ca-ions and polysaccharide chains bearing PO4-ions internally. Unlike the traditional external setting technique, this strategy of internal setting technique would effectively provide calcium ions within the alginate molecules to give rise to uniform calcium ion concentrations throughout the modified molecules [24]. Herein, we report the preparation of CaPAlg and its application for in vivo injectable hydrogels. The phosphorylation of alginic acid sodium salts was attempted using H3PO4/P2O5/Et3PO4/hexanol method for the first time, and the calcium complexes (CaPAlg) were produced by acid– base reaction using Ca(OAc)2. A series of hydrogels was successfully prepared by screening various ratios of NaAlg and CaPAlg with varying setting times from ~3 to ~40 min and optimal mechanical strength. 1Hand 31P-NMR spectroscopy, FT-IR spectroscopy, ICP-OES, TGA, SEM analysis, compressive mechanical test, and in vitro cell test were employed to characterize the modified alginates and the prepared hydrogels.
2. Experimental 2.1. Materials Sodium alginate (NaAlg; Mw 1.2–1.9 × 105, M/G ratio 1.56), triethylphosphate, phosphorus pentoxide, and 1-hexanol were purchased from Sigma-Aldrich (USA), and phosphoric acid was purchased from J. T. Baker (USA). All purchased chemicals were used without further purification.
2.2. Preparation of phosphorylated alginic acid (PAlg) NaAlg was phosphorylated according to the procedure described by Amaral et al. for phosphorylation of chitosan [29]. For phosphorylation of sodium alginate using H3PO4/P3O5/Et3PO4/hexanol, the sodium alginate powder (1 g) was suspended to H3PO4 (43 mL) at 37 °C with stirring and was slowly added a separately prepared solution of P2O5 (28 g) and Et3PO4 (38 mL) in hexanol (40 mL). The mixture was stirred for 2 days (48 h) at 37 °C. The reaction mixture was poured into a beaker containing methanol of 4–500 mL to give white precipitates. The precipitates were filtered using polytetrafluoroethylene (PTFE) filter (0.5 μm pore size, 47 mm in diameter), rinsed with methanol (50 mL × 3), and dried in vacuum at room temperature for one day to give 700 mg of phosphorylated alginic acid (PAlg) as pale yellow granules in 72% yield. 2.3. Preparation of phosphorylated alginic acid calcium complexes (CaPAlg) The as-obtained PAlg powder (100 mg) was dissolved in distilled water (100 mL) and was adjusted to pH ~ 8 by slowly adding aqueous Ca(OAc)2 solution (1.0 N), by the exchange of protons of the phosphorylated alginic acid molecules with calcium ions. The solution was then dialyzed using dialysis tubing membrane (pore: 12,000 Da) for 2 days against distilled water to remove all the inorganic salts. The water was changed every 6 h. After freeze-drying, 97 mg of CaPAlg was obtained as a white powder, an 87% yield. 2.4. Preparation of in vitro injectable hydrogels from phosphorylated alginic acid calcium complex (CaPAlg) The synthesized CaPAlg was used in cross-linking of NaAlg to form a hydrogel simply by mixing immediately after short vortexing. An aqueous solution of NaAlg (2 wt.%) (solution A) and an aqueous solution of CaPAlg (2 wt.%) (solution B) were prepared separately; both (solutions A and B) were subsequently mixed in the ratios, as shown in (Table 2). Each mixture was fixated in polydimethylsiloxane (PDMS) mold (SYLGARD®) until it formed hydrogels. 2.5. Characterization of the prepared alginic acid derivatives and the in vitro injectable hydrogels Fourier transform infrared (FT-IR) spectroscopy was performed on a Perkin-Elmer Spectrum BXII FT-IR spectrometer. The IR spectra were recorded between 4000 and 500 cm−1, with 16 scans at 4 cm−1 resolution. Nuclear magnetic resonance (NMR) spectra were collected on a Bruker Ultrashield 500 PLUS NMR spectrometer for 1H, and 31P nuclei. High resolution inductively coupled plasma optical emission spectroscopy (ICP-OES) was carried out on the PAlg and CaPAlg samples to determine the degree of substitution (DS) values. Samples were digested using nitric acid and analyzed using a Perkin Elmer Optima 5300dV spectrometer, with a forward rate of 1.5 mL/min (temperature 24 ± 1 °C, relative humidity 29 ± 1%). Burgener PEEK Mira Mist nebulizer was used and the instrument was operated by taking the average of three 5 s integrations at 213.62 nm for PAlg samples and at 317.93 nm for CaPAlg samples. All samples were run in triplicate. Thermogravimetric analysis (TGA) was performed using SCINCO TGA N-1500 (sample weight 5 ± 0.1 mg, heating rate 10 °C/min, initial temperature rt). The microporous structure of the CaPAlg hydrogels was observed by scanning electron microscopy (SEM) in lyophilized hydrogel samples Table 1 The contents of P and Ca in PAlg and CaPAlg samples (100 mg × 3) by ICP-OES. Sample
Fig. 1. a) FT-IR spectra and b) thermograms of NaAlg, PAlg, and CaPAlg.
PAlg CaPAlg
P
Ca
×104 (mg/kg)
DS (mol%)
×104 (mg/kg)
DS (mol%)
1.5 ± 0.1 0.31 ± 0.02
12.5 7.8
– 7.8 ± 0.1
– 112.5
H.-S. Kim et al. / Materials Science and Engineering C 51 (2015) 139–147
(Hitachi S-4300) at a voltage of 20 kV. SEM analysis was performed in duplicate. 2.6. Compression mechanical tests Uniaxial compression measurements were performed on the hydrogels prepared at 25 °C using a mechanical testing machine (INSTRON 3344, MA, USA) in compression mode. Hydrogel samples were cut to specimens with a diameter of 9.0 mm and thickness of 6.4 mm. All measurements have been tested for 1–2 min until break at a load rate (crosshead speed) of 10 mm/min. To obtain statistically reliable results all measurements were performed on 7 samples each of four hydrogel types, CaPAlg20–80. 2.7. Measurement of gelation time Determination of gelation time was applied using the well known vial inversion test (or the flow test) [25–28]. The gelation time was determined by the point at which the mixture of solutions A and B no longer flows upon inversion of the vial. Three replicate samples were measured at room temperature and an average value was taken. 2.8. In vitro cell test of cell-encapsulated hydrogels In order to evaluate the feasibility of the in vitro injectable CaPAlg hydrogels for use as a cell encapsulation material, the viability of MC3T3s encapsulated in the hydrogels was examined using LIVE/ DEAD® Viability/Cytotoxicity Kit (Invitrogen, Carlsbad, CA) via a fluorescence microscope and MTS assay. The sub-cultured MC3T3s at a density of 1 × 106 cells were encapsulated into the hydrogels of 1 ml, and cultured for up to 3 days. All procedure was carried out in laminated
141
lamellar flow hood. The medium was completely renewed every two days during culturing. After prescribed culturing time, the cellencapsulated hydrogels were placed in a new culture plate, and washed with phosphate buffered saline (PBS) solution 3 times. Cell viability and morphology were examined post-staining. To assess the encapsulation efficiency of the hydrogel, the cell survivability after encapsulation (0-day) was measured by direct cell-counting based on the Live/Dead staining. The metabolic activity of the cell for time in culture was evaluated using MTS ((3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4sulfophenyl)-2H-tetrazolium)) assay kit. After culturing, the hydrogels were placed in new 24-well plates and then MTS solution was added to each hydrogel. After 4 h of incubation at 37 °C, colorimetric measurements were performed on a micro reader with an optical density reading at 490 nm. All cell culture experiments were performed in triplicate and the data were compared by performing a statistical analysis using oneway ANOVA at a significance level of p b 0.01. 3. Results and discussion 3.1. Preparation of phosphorylated alginic acid calcium complexes (CaPAlg) Since the phosphorylation of cellulose using H3PO4/P2O5/Et3PO4/1hexanol was first attempted by Touey and Kingsfort for the hemi synthesis of water-soluble and non-degradable flame-proof textile [30], some similar methods using P2O5/MeSO3H or H3PO4/Urea/DMF have been employed for the phosphorylation of polysaccharide such as chitosan [31,32]. It is well known that phosphorylation of chitosan by P2O5/ MeSO3H or H3PO4/Urea/DMF results in reduced molecular weight of the polymer due to degradation caused by strong acidic conditions and difficulty in separating the final product when urea was included [31,32]. However, the H3PO4/P2O5/Et3PO4/1-hexanol method for the
Fig. 2. 1H-NMR (D2O) spectra of NaAlg, PAlg, and CaPAlg: Here, Gn and Mn indicate the protons at C-n (n = 1–5) for, respectively, G and M units. The colorful circle symbols were placed to facilitate the comparison.
142
H.-S. Kim et al. / Materials Science and Engineering C 51 (2015) 139–147
phosphorylation of cellulose proceeded under milder reaction conditions and caused low degradation of the polymer [11,33]. Moreover, the applicability of the product to the field of bone tissue engineering has been shown by in vitro and in vivo cytocompatibility [34]. Based on the aforementioned advantages of phosphorylation reaction conditions and its bio-applicability, we have attempted phosphorylation of NaAlg using H3PO4/P2O5/Et3PO4 method, for the first time (see scheme S1 of Supporting material). As shown in Fig. 1 (see the full spectra in Fig. S1 of Supporting material), the as-phosphorylated alginic acid was characterized by FT-IR and TGA on comparison with sodium alginate. The FT-IR spectra of unmodified and modified samples are shown in Fig. 1a. The IR spectrum of unmodified NaAlg showed characteristic bands: at 1614 and 1416 cm−1 for C_O (asymmetric and symmetric) stretching of carboxylate, and at 1300–1090 cm−1 for C\O(H) and C\O\C (ring) vibrational modes [35,36]. After phosphorylation to PAlg, several new peaks were observed at 1740, 1630, 1244, 1084, 948, and 928 cm−1 in the IR spectrum. Peaks at 1740 and 1630 cm−1 correspond to C_O stretching of carboxylic acids indicating carboxylate ions of sodium alginate were predominantly converted into the acid form. Other peaks from phosphate group such as P_O asymmetric stretching at 1244 cm−1 and P\O\C (aliphatic) and two P\O(H) vibrational modes at 948–928 cm−1 were apparently observed indicative of the presence of phosphate on the ring as well. Disappearance of a peak at 1300 cm−1 of sodium alginate after phosphorylation, strongly supports the fact that the phosphorylation has occurred at the hydroxyl groups of the alginate ring. Overall, the FT-IR spectrum of the modified alginic acid, PAlg using H3PO4/P2O5/Et3PO4/1-hexanol method was identical to previous data on preparations by the urea/phosphate method [23]. Calcium complexes (CaPAlg) of PAlg were prepared by cation exchange reaction of PAlg with Ca(OAc)2. As shown in Fig. 1a, it was noticeable that the FT-IR spectrum of CaPAlg complexes became quite different from that of PAlg, but similar to those of the NaAlg organic salts.
Table 2 Injectable hydrogels prepared by 2 wt.% of CaPAlg solution (Solution A) and 2 wt.% of NaAlg solution (Solution B). Entry no.
Solution A (mL)
Solution B (mL)
Hydrogel code-name
1 2 3 4
0.2 0.4 0.6 0.8
0.8 0.6 0.4 0.2
CaPAlg20 CaPAlg40 CaPAlg60 CaPAlg80
The characteristic peaks of phosphorous-related functional groups (P_O, P\O\C, and P\O(H)) at 1244–928 cm−1 showed no significant changes compared with PAlg indicating the phosphates were intact on the alginate ring. Especially, the C_O related broad peaks at 1608 and 1422 cm− 1 which were shifted from 1740 and 1630 cm−1 of PAlg strongly indicated that all carboxylic acids were converted back into the carboxylate anions which form calcium complex. In order to further study the composition difference of NaAlg, PAlg, and CaPAlg, thermal decomposition patterns were examined by TGA analysis as shown in Fig. 1b. The NaAlg sample used as a reference [37, 38], showed two steps of thermal decomposition after evaporation of water. The first step for the pyrolysis of hydrocarbon moiety occurred at about 200 °C. After slight decrease of residual carbohydrate pyrolysis over the range of 250–500 °C, the second step of pyrolysis occurred at ~600 °C, remaining about 20.7 wt.% of residual mass of sodium oxide. For PAlg, several complicated steps of thermal decomposition were observed. The first step for the pyrolysis of hydrocarbon moiety occurred clearly at about 200 °C. The following steps which occurred in the range of 300–600 °C and N 600 °C might have included additional carbohydrate pyrolysis, dehydration of phosphate groups to phosphorus pentoxide, and vaporization of phosphorus pentoxide, resulting in only 0.5 wt.% of remaining polymorphous polyphosphate as a glassy
Fig. 3. a) Schematic representation for the preparation of injectable hydrogel using CaPAlg (A: red) and NaAlg (B: blue) {the inset photos show the liquid form of the precursor solutions (A and B) and the injectable property of a solution mixture (A + B for CaPAlg50) after mixing (see Fig. 3S}; b) photos of hydrogel figures prepared from different volume ratios of CaPAlg and NaAlg solutions: before (top) and after (middle) freeze drying and gelation time (down).
H.-S. Kim et al. / Materials Science and Engineering C 51 (2015) 139–147
or amorphous form. Similar to PAlg case, several steps of decomposition were observed in the thermogram of CaPAlg. After the evaporation of water, the first step for the pyrolysis of hydrocarbon moiety between 200–300 °C, and the second step for additional carbohydrate pyrolysis including dehydration of phosphate groups to phosphorus pentoxide between 300–600 °C were clearly observed. The additional weight loss appeared in the high temperature range of N800 °C is presumably attributed to the continual dehydration process of Ca(OH)2 and CaP compounds which were possibly formed at the high temperature (e.g., Ca(OH)2 → CaO; Ca(H2PO4)2 → CaHPO4 → Ca3(PO4)2). The contents of P and Ca ion in PAlg and CaPAlg were measured by elemental analysis using ICP-OES. The measurements for each of both samples (100 mg) were performed in triplicate. The contents (in ppm and mol%) of phosphorus and calcium contained per 1 mol of cyclic monosaccharide unit are summarized in Table 1. On average, PAlg samples contained about 0.125 mol of P element per 1 mol of the cyclic monosaccharide units and CaPAlg samples contained N 1.13 mol of Ca element per 1 mol of the cyclic monosaccharide units, resulting in DS value of 12.5 and 112.5 mol%, respectively. The DS value (12.5 mol%) of PAlg was comparably higher than the previous value obtained by Grøndahl's method [23]. According to a recent report for the phosphorylation of chitosan [39], among 3 available phosphorylation methods, namely H3PO4/urea, H3PO4/Et3PO4/P2O5, and P2O5/MeSO3H methods, the second method (H3PO4/Et3PO4/P2O5) had the highest phosphorylated products yield. The DS value (112.5 mol%) of Ca ion for the CaPAlg sample was found to be much higher than that (12.5 mol%) of P element for PAlg sample. Theoretically, the functional groups such as carboxylic and phosphoric acids within the PAlg molecules will allow the cation exchange in the presence of Ca(OAc)2 to form the phosphorylated alginic acid calcium complexes. Here the DS value of Ca ions for the prepared CaPAlg molecules was calculated on the basis that all the carboxylic acid functional groups within the molecules exchange their protons with calcium ions. The obtained high DS value indicated that all two cyclic monosaccharide units have a calcium ion at the carboxylate functional group sites and additionally about twelve cyclic monosaccharide units have a calcium ion at the phosphate functional group site. Consequently,
143
these results imply that all carboxylate and phosphate functional groups could actively form strong ionic bonds with the divalent cations without any obvious gelation and the individual calcium ions could be evenly dispersed on the surface of the CaPAlg polymer molecules. These chemical characteristics support that the CaPAlg complexes are potentially suitable for the preparation of injectable hydrogel. 1 H-NMR studies were performed on both PAlg and CaPAlg to identify the possible sites of phosphorylation and cation exchange with Ca2+, as shown in Fig. 2 (see the original spectra in Fig. S2 of Supporting material). The analyses of 1H-NMR spectra for the unmodified and phosphorylated alginates have been well established in the literature [23, 40,41]. Major peaks of unmodified sodium alginate and phosphatemodified alginic acid prepared by us, were assigned based on the literature value as shown in Fig. 2 [23]. It was noticed that the entire spectrum was globally shifted to the downfield with significantly increased complexity upon phosphorylation and protonation. In a related study, Grøndahl et al. reported that phosphorylation of M residues occurred predominantly at the C-3 equatorial hydroxyl groups, and phosphorylation occurred for G residues as well, but it was difficult to determine the site of phosphorylation [23]. However, the C\H protons at C-2, -3, and -5 of both M and G units for the PAlg prepared by us were strongly shifted to downfield compared to that of NaAlg (Fig. 2), while those at C-1 and -4 were weakly shifted. These results may suggest many electrical changes at C-2, -3, and -5 by phosphorylation of the hydroxyl groups and by protonation of all the carboxylate groups at C-5. Compared to Grøndahl's results, the higher complexity of peaks in the 1 H NMR spectrum from the PAlg suggests that the H3PO4/P2O5/Et3PO4 method is more effective for the phosphorylation than the urea/H3PO4 method. Furthermore, Liu et al. recently reported that the same method (H3PO4/P2O5/Et3PO4) gave a higher degree of phosphorylation on the chitosan ring than the other available methods, i.e., H3PO4/urea and P2O5/MeSO3H [39]. Therefore, it was reasonably expected that H3PO4/ P2O5/Et3PO4 method would give rise to comparably higher degree of phosphorylation of alginate than the other methods. This significant change in 1H NMR spectrum directly corresponded to the relatively higher DS value (12.5 mol% in average) of PAlg as shown in ICP-OES
Fig. 4. SEM images (a–d) of CaPAlg hydrogels: a) CaPAlg20, b) CaPAlg40, c) CaPAlg60, and d) CaPAlg80.
144
H.-S. Kim et al. / Materials Science and Engineering C 51 (2015) 139–147
analysis. In addition, the 31P-NMR spectrum of PAlg also tested as shown in Fig. S2 (see Supporting material), was identical with Grøndahl's results, of a strong signal at −0.01 ppm and a low intensity signal at −10.83 ppm, which is known as pyrophosphate [23]. No other signals were observed. Finally, 1H NMR spectrum of CaPAlg obtained after cation exchange of PAlg with calcium ions maintained the high complexity of PAlg caused by phosphorylation, while the entire spectrum was shifted back to upfield similar to NaAlg case probably due to the formation of calcium salt. When compared to the spectrum of NaAlg, the C-2 and C-3 related protons of both M and G units for the CaPAlg were still to be weakly shifted to downfield due to the phosphorylation (Fig. 2), while those at C-1, C-4 and C-5 almost returned to the chemical shift values of NaAlg. The general splitting patterns of proton peaks also were similar to those of NaAlg, but the fine patterns were still more complex than those of NaAlg due to the presence of phosphate groups (compare the peaks in the colorful circle symbols). Intramolecular and intermolecular complex networkings between the deprotonated phosphate and carboxylate groups via divalent cations probably affect the splitting patterns of proton peaks. Consequently, it could be clearly considered that the phosphorylation of the hydroxyl groups at C-2 and C-3 and the formation of calcium salt at C-2, C-3, and C-5 such as the phosphoric acid calcium salt and carboxylic acid calcium salt cause the downfield shift of protons at C-2 and C-3 for both G and M units and the more complex splitting of the proton peaks. 3.2. Preparation of injectable hydrogels from phosphorylated alginic acid calcium complex (CaPAlg) Unlike the traditional method to form hydrogels where calcium ions are diffused externally into the alginate solution [21], we have successfully prepared hydrogels by internal diffusion technique where phosphorylated alginic acid molecules bearing calcium ions (CaPAlg) form complicated networks between themselves and NaAlg molecules by using their calcium ions. Schematic representation for the preparation and the formation of injectable hydrogel using CaPAlg and NaAlg is depicted in Fig. 3a (see the time-dependent gelation property of CaPAlg50 in Fig. S3 of Supporting material). Hydrogel preparation was performed by simply mixing CaPAlg solution with NaAlg solution in different ratios as shown in Table 2. Gelation times for each sample and their morphological appearances before and after freeze drying were depicted in Fig. 3b. For example, the CaPAlg/NaAlg combinations of CaPAlg20 and CaPAlg80 hydrogels showed relatively longer gelation times (N40 min and N30 min, respectively), because one of the most important components (for example, Ca2 + and non-modified alginate chain of NaAlg) for gel-formation was insufficient in the two combinations. The optimum results were obtained from the CaPAlg/NaAlg combinations of CaPAlg40 and CaPAlg60 hydrogels showing effective gelation within 3–10 min and then good fixation in PDMS mold. To obtain the most optimal conditions for setting time and gel-formation, the preparation of CaPAlg50 hydrogel was accomplished by more fine tuning the concentration of both components as follows: 1 mL of 2 wt.% CaPAlg solution + 1 mL of 2 wt.% NaAlg solution. In that case, the amount of internally supplied Ca-ions is considered to be about 1.56 mg (0.04 mmol) within about 38.44 mg of alginate component (PAlg + NaAlg). In order to examined the possibility of gelation using the equimolar amount of NaAlg (38.44 mg/1.5 mL) and CaCl2 (0.04 mmol/0.5 mL or 4.44 mg/0.5 mL) that the CaPAlg50 hydrogel contain, we tried gelation by external addition (a traditional method), but no gelation was observed at all (see NaAlg/CaCl2 in Fig. 3b). In order to achieve optimal hydrogel formation by the external supply of Ca-ion, usually, at least 17.32 mg (0.156 mmol, about 4 times more amount than the case of CaPAlg) of CaCl2 is necessary. In this case since the gelation occurs promptly it is considered that transferring the mixture solution under the skin through a syringe is inappropriate. These encouraging results from CaPAlg40–60, such as effective gelation by the minimum quantity of Ca-ion and the optimal gelation times enabling injection under the
skin may be caused by (i) the different routes of Ca ion supply (internal vs. external), (ii) the uniform dispersibility of Ca-ions on the modified alginate molecules, (iii) the different mechanism of Ca-ion transfer from carboxylat (and/or phosphate) functional groups of CaPAlg molecules to carboxylate functional groups of NaAlg molecules, and (iv) the direct participation of the polymeric counter ions of CaPAlg bearing phosphate and carboxylate anions in networking with the normal alginate molecules. Moreover, the use of CaPAlg to produce injectable alginate-based hydrogels showed easy gelation without significant change in the volume of produced hydrogels. This also may be an important characteristic for favorable cell encapsulation since the encapsulated cells can be free from physical stress that could be potentially induced by volume change during the gelation of a hydrogel. The SEM images of the alginate hydrogels revealed highly interconnected pores cross-linked by calcium ion as shown in Fig. 4. As the amount of CaPAlg component within the hydrogels increased from 20 to 80 v/v%, the hydrogels showed several important characteristic trends as follows: i) a decrease of pore size, ii) a decrease of regularity of pore shape, and iii) an increase of roughness of pore walls. Indeed, CaPAlg20 hydrogel shows regularly networked pore structures with average pore size ~800 μm in length that are similar to the one formed by NaAlg and CaCl2. However, the other hydrogels, CaPAlg40 ~ 80, showed round-shaped pore structures with an increased irregularity and roughness and the smaller sizes ranged from 320–400 μm in diameter. The increased irregularity in the pore shapes upon increasing CaPAlg concentration was probably caused by irregular participation of P-modified alginate in the networking process with pristine alginate molecules. Overall, SEM analysis showed that hydrogel
Fig. 5. a) Compression stress–strain curves of CaPAlg40–80 hydrogels and b) the straight lines fits to the experimental data in the initial deformation range (represents the elastic modulus).
H.-S. Kim et al. / Materials Science and Engineering C 51 (2015) 139–147
CaPAlg40–60, which have shown optimal gelation time (about 3–10 min) and pore structures of 300–400 μm in diameter, may provide favorable microenvironments for cell growth for tissue engineering applications. The stress–strain curves of compression tests of CaPAlg40 and CaPAlg60 hydrogels at 25 °C are presented in Fig. 5a, whereas Fig. 5b shows the initial part of experimental curves. As seen in Fig. 5a, the incorporation of 20 and 40 wt.% CaPAlg increased the ultimate compression strength (σ) at break from about 0.6 to 6.7 kPa; however, at the same time, the increased incorporation of 60 and 80 wt.% CaPAlg component reduced the maximum compression strength (σ) of CaPAlg40 at break to, respectively, about 3.5 and 2.3 kPa. The increase of the compression strength (σ) at break from CaPAlg20 to CaPAlg40 is obviously due to the higher networking by the higher Ca ion concentration, whereas the reduction of the compression strength (σ) at break from CaPAlg40 to CaPAlg80 is obviously due to the lower networking by the lower concentration of unmodified alginate component of NaAlg. The elastic modulus (initial slope at lineal region) of the strongest hydrogels of CaPAlg40 and CaPAlg60 were determined to be 8.4 ± 0.412 kPa/mm and 4.8 ± 0.385 kPa/mm, respectively. The elastic modulus values may suggest that the CaPAlg40 hydrogel is effectively networked by the higher concentrations of Ca ion and unmodified alginate components. Consequently, the incorporation of
145
about 40 wt.% CaPAlg component into 60 wt.% NaAlg component may be the most effective for the improvement of the mechanical properties of the CaPAlg hydrogels. 3.3. Cellular responses The biocompatibility of the injectable hydrogel CaPAlg50, which has the middle composition of CaPAlg40–60, was assessed by determining cellular morphology, monitoring survivability and metabolic activity of cells encapsulated into the hydrogels. For this, the biocompatibility of the NaAlg hydrogel prepared using 50 mM CaCl2 solution was compared as a control. Fig. 6 shows the fluorescence images of cells encapsulated within the hydrogel during 3-day culture. The green spots indicate live cells. Through the fluorescence images, we investigated the cell dispersion, the cell morphology and survivability in the hydrogels. At 4 h post encapsulation (0-day), the round-shaped cells with good distribution were observed in all hydrogels, no significant different in the cell survivability between the hydrogels. During 3 days of culture period, the live cells after encapsulation were almost alive without spreading within the NaAlg hydrogel, while the CaPAlg50 hydrogel revealed some dead cells. Fig. 7 shows the cell survivability at 4 h post encapsulation and metabolic activity for several culture periods determined by MTS assay for quantitative analysis. The NaAlg hydrogel and the
Fig. 6. The fluorescence images of cells encapsulated within the hydrogels, a) pure alginate hydrogel and b) CaPAlg50 hydrogel, during different culture times (4 h and 1–3 days). The green spots indicate live cells.
146
H.-S. Kim et al. / Materials Science and Engineering C 51 (2015) 139–147
microenvironment for 3 day encapsulation of cells and the initial cell survivability for 0–1 day. Further study regarding application of CaPAlg hydrogels in the field of soft tissue engineering are currently ongoing in our laboratory. Acknowledgments This study was supported by grants (2009-0093829 (Priority Research Centers Program) and 2012-003905) from the National Research Foundation of Korea. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.msec.2015.02.031. References
Fig. 7. a) The initial cell survivability for 0–1 day and b) the cell proliferation rate in normal alginate hydrogel (NaAlg) and CaPAlg50 hydrogel for 3 days, which was normalized by the proliferation level of 0-day NaAlg hydrogel.
injectable hydrogel CaPAlg50 possessed similar initial cell survivability of approximately 95%, but the cell proliferation rate in each hydrogel for longer time, which was normalized by the proliferation level of 0day NaAlg hydrogel, was slightly different. During 3-day culture, the encapsulated cells within the NaAlg hydrogel were continuously growing (viability of 125%), whereas the cell proliferation within the injectable CaPAlg50 hydrogels decreased (viability of 71%). Consequently, the results revealed that the injectable CaPAlg40–60 hydrogels provide favorable microenvironment for 3 day encapsulation of cells and the initial cell survivability for 0–1 day. Further studies regarding improvement of the cell proliferation level of the injectable CaPAlg hydrogels for a long time period and their applications for cartilage and soft tissue engineering are currently ongoing in our laboratories. 4. Conclusions We have prepared phosphorylated alginate (PAlg) using H3PO4/ P2O5/Et3PO4/hexanol method under mild reaction conditions. PAlg was utilized in ion exchange with minimum amount of Ca(OAc)2 to form CaPAlg. Injectable hydrogel materials were successfully prepared from CaPAlg and NaAlg solutions via simple mixing to give uniform ion concentrations throughout by “internal setting” technique without externally adding calcium salts. The preparation of PAlg and CaPAlg was confirmed by instrumental analyses using 1H- and 31P-NMR, FTIR, TGA, and ICP-MS. The injectable CaPAlg hydrogels (especially CaPAlg40–60) showed optimal gelation time (about 3–40 min) and less regularly interconnected pore structures of 300–400 μm in diameter. Their mechanical properties were tested to be ≤6.7 kPa for compressive strength at break and about 8.4 kPa/mm for elastic modulus. Cell culture assay revealed that CaPAlg hydrogels provide favorable
[1] K.H. Bae, L.-S. Wang, M. Kurisawa, Injectable biodegradable hydrogels: progress and challenges, J. Mater. Chem. B 1 (2013) 5371–5388. [2] H. Singh, in: M. Ramalingam, A. Tiwari, S. Ramakrishna, H. Kobayashi (Eds.), Injectable in situ gelling hydrogels as biomaterials, Integrated Biomaterials for Biomedical Technology, 1st ed.Wiley, 2012, pp. 359–396. [3] K. Nagahama, A. Takahashi, Y. Ohya, Biodegradable polymers exhibiting temperature-responsive sol–gel transition as injectable biomedical materials, React. Funct. Polym. 73 (2013) 979–985. [4] A.A. Amini, L.S. Nair, Injectable hydrogels for bone and cartilage repair, Biomed. Mater. 7 (2012) 024105/1–024105/13. [5] M.M. Pakulska, B.G. Ballios, M.S. Shoichet, Injectable hydrogels for central nervous system therapy, Biomed. Mater. 7 (2012) 024101/1–024101/13. [6] D.J. Overstreet, D. Dutta, S.E. Stabenfeldt, B.L. Vernon, Injectable hydrogels, J. Polym. Sci. B Polym. Phys. 50 (2012) 881–903. [7] D. Macaya, M. Spector, Injectable hydrogel materials for spinal cord regeneration: a review, Biomed. Mater. 7 (2012) 012001/1–012001/22. [8] Y. Li, J. Rodrigues, H. Tomas, Injectable and biodegradable hydrogels: gelation, biodegradation and biomedical applications, Chem. Soc. Rev. 41 (2012) 2193–2221. [9] C.T. Huynh, M.K. Nguyen, D.S. Lee, Injectable block copolymer hydrogels: achievements and future challenges for biomedical applications, Macromolecules 44 (2011) 6629–6636. [10] J.-F.T. Lutz, Thermo-switchable materials prepared using the oligoethylene glycol methacrylate-platform, Adv. Mater. 23 (2011) 2237–2243. [11] B. Balakrishnan, R. Banerjee, Biopolymer-based hydrogels for cartilage tissue engineering, Chem. Rev. 111 (2011) 4453–4474. [12] K.Y. Lee, D.J. Mooney, Hydrogels for tissue engineering, Chem. Rev. 101 (2001) 1869–1879. [13] D. Seliktar, Designing cell-compatible hydrogels for biomedical applications, Science 336 (2012) 1124–1128. [14] A.M. Kloxin, A.M. Kasko, C.N. Salinas, K.S. Anseth, Photodegradable hydrogels for dynamic tuning of physical and chemical properties, Science 324 (2009) 59–63. [15] G. Orive, A.M. Carcaboso, R.M. Hernández, A.R. Gascόn, J.L. Pedraz, Biocompatibility evaluation of different alginates and alginate-based microcapsules, Biomacromolecules 6 (2005) 927–931. [16] G. Orive, S. Ponce, R.M. Hernandez, A.R. Gascon, M. Igartua, J.L. Pedraz, Biocompatibility of microcapsules for cell immobilization elaborated with different type of alginates, Biomaterials 23 (2002) 3825–3831. [17] O. Smidsrød, G. Skjåk-Bræk, Alginate as immobilization matrix for cells, Trends Biotechnol. 8 (1990) 71–78. [18] A. Shilpa, S.S. Agrawal, A.R. Ray, Controlled delivery of drugs from alginate matrix, J. Macromol. Sci. Polym. Rev. C43 (2003) 187–221. [19] J.S. Boateng, K.H. Matthews, H.N.E. Stevens, G.M. Eccleston, Wound healing dressings and drug delivery systems: a review, J. Pharm. Sci. 97 (2008) 2892–2923. [20] J.L. Drury, R.G. Dennis, D. Mooney, The tensile properties of alginate hydrogels, Biomaterials 25 (2004) 3187–3199. [21] S.N. Pawar, K.J. Edgar, Alginate derivatization: a review of chemistry, properties and applications, Biomaterials 33 (2012) 3279–3305. [22] S.N. Pawar, K.J. Edgar, Alginate esters via chemoselective carboxyl group modification, Carbohydr. Polym. 98 (2013) 1288–1296. [23] R.J. Coleman, G. Lawrie, L.K. Lambert, M. Whittaker, K.S. Jack, L. Grøndahl, Phosphorylation of alginate: synthesis, characterization, and evaluation of in vitro mineralization capacity, Biomacromolecules 12 (2011) 889–897. [24] G. Skjåk-Bræk, H. Grasdalen, O. Smidsrød, Inhomogeneous polysaccharide ionic gels, Carbohydr. Polym. 10 (1989) 31–54. [25] E.F. Banwell, E.S. Abelardo, D.J. Adams, M.A. Birchall, A. Corrigan, A.M. Donald, Rational design and application of responsive alpha-helical peptide hydrogels, Nat. Mater. 8 (2009) 596–600. [26] B.S. Luisi, K.D. Rowland, B. Moulton, Coordination polymer gels: synthesis, structure and mechanical properties of amorphous coordination polymers, Chem. Commun. 27 (2007) 2802–2804. [27] Y. Yu, Y. Ma, Breath figure fabrication of honeycomb films with small molecules through hydrogen bond mediated self-assembly, Soft Matter 7 (2011) 884–891.
H.-S. Kim et al. / Materials Science and Engineering C 51 (2015) 139–147 [28] K. Chawla, T. Yu, S.W. Liao, Z. Guan, Biodegradable and biocompatible synthetic saccharide–peptide hydrogels for three-dimensional cell culture, Biomacromolecules 12 (2011) 560–567. [29] I.F. Amaral, P.L. Granja, M.A. Barbosa, Chemical modification of chitosan by phosphorylation: an XPS, FT-IR and SEM study, J. Biomater. Sci. Polym. Ed. 16 (2005) 1575–1593. [30] J.D. Reid, L.W. Mazzeno, Preparation and properties of cellulose phosphates, Ind. Eng. Chem. 41 (1949) 2828–2831. [31] N. Nishi, S.I. Nishimura, A. Ebina, A. Tsutsumi, S. Tokura, Preparation and characterization of water soluble chitin phosphate, Int. J. Biol. Macromol. 6 (1984) 53–54. [32] D.R. Khanal, K. Miyatake, Y. Okamoto, T. Shinobu, M. Morimoto, H. Saimoto, Y. Shigemasa, S. Tokura, S. Minami, Phosphated chitin (P-chitin) exerts protective effects by restoring the deformability of polymorphonuclear neutrophil (PMN) cells, Carbohydr. Polym. 48 (2002) 305–311. [33] P.L. Granja, L. Pouységu, M. Pétraud, B. De Jéso, C. Baquey, M. Barbosa, Cellulose phosphates as biomaterials. I. Synthesis and characterization of highly phosphorylated cellulose gels, J. Appl. Polym. Sci. 82 (2001) 3341–3353. [34] J.C. Fricain, P.L. Granja, M.A. Barbosa, B. De Jéso, N. Barthe, C. Baquey, Cellulose phosphates as biomaterials. In vivo biocompatibility studies, Biomaterials 23 (2002) 971–980.
147
[35] C. Sartori, D.S. Finch, R.B., G.K., Determination of cation content of alginate thin films by FT IR spectroscopy, Polymer 38 (1997) 43–51. [36] G. Lawrie, I. Keen, B. Drew, A. Chandler-Temple, L. Rintoul, P. Fredericks, L. Grøndahl, Interactions between alginate and chitosan biopolymers characterized using FTIR and XPS, Biomacromolecules 8 (2007) 2533–2541. [37] T.S. Pathak, J.S. Kim, S.J. Lee, D.J. Baek, K.J. Paeng, Preparation of alginic acid and metal alginate from algae and their comparative study, J. Polym. Environ. 16 (2008) 198–204. [38] P. Sundarrajan, P. Eswaran, A. Marimuthu, L.B. Subhadra, P. Kannaiyan, One pot synthesis and characterization of alginate stabilized semiconductor nanoparticles, Bull. Korean Chem. Soc. 33 (2012) 3218–3224. [39] K. Wang, Q. Liu, Chemical structure analyses of phosphorylated chitosan, Carbohydr. Res. 386 (2014) 48–56. [40] H. Grasdalen, High-field, 1H-NMR spectroscopy of alginate: sequential structure and linkage conformations, Carbohydr. Res. 118 (1983) 255–260. [41] H. Grasdalen, B. Larsen, O. Smidsrød, 13C-NMR studies of monomeric composition and sequence in alginate, Carbohydr. Res. 89 (1981) 179–191.
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Injectable hydrogels derived from phosphorylated alginic acid calcium complexes Han-Sem Kim, Minsoo Song ⁎, Eun-Jung Lee, Ueon Sang Shin ⁎ Department of Nanobiomedical Science & BK21 PLUS NBM Global Research Center for Regenerative Medicine, Dankook University, Chungnam, Cheonan 330-714, Republic of Korea Institute of Tissue Regeneration Engineering (ITREN), Dankook University, Chungnam, Cheonan 330-714, Republic of Korea
a r t i c l e
i n f o
Article history: Received 26 November 2014 Received in revised form 28 January 2015 Accepted 23 February 2015 Available online 25 February 2015 Keywords: Phosphorylated alginic acid calcium complexes In vivo injectable hydrogel Cell and drug delivery
a b s t r a c t Phosphorylation of sodium alginate salt (NaAlg) was carried out using H3PO4/P2O5/Et3PO4 followed by acid–base reaction with Ca(OAc)2 to give phosphorylated alginic acid calcium complexes (CaPAlg), as a water dispersible alginic acid derivative. The modified alginate derivatives including phosphorylated alginic acid (PAlg) and CaPAlg were characterized by nuclear magnetic resonance spectroscopy for 1H, and 31P nuclei, high resolution inductively coupled plasma optical emission spectroscopy, Fourier transform infrared spectroscopy, and thermogravimetric analysis. CaPAlg hydrogels were prepared simply by mixing CaPAlg solution (2 w/v%) with NaAlg solution (2 w/v%) in various ratios (2:8, 4:6, 6:4, 8:2) of volume. No additional calcium salts such as CaSO4 or CaCl2 were added externally. The gelation was completed within about 3–40 min indicating a high potential of hydrogel delivery by injection in vivo. Their mechanical properties were tested to be ≤6.7 kPa for compressive strength at break and about 8.4 kPa/mm for elastic modulus. SEM analysis of the CaPAlg hydrogels showed highly porous morphology with interconnected pores of width in the range of 100–800 μm. Cell culture results showed that the injectable hydrogels exhibited comparable properties to the pure alginate hydrogel in terms of cytotoxicity and 3D encapsulation of cells for a short time period. The developed injectable hydrogels showed suitable physicochemical and mechanical properties for injection in vivo, and could therefore be beneficial for the field of soft tissue engineering. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Injectable hydrogels with biodegradability have emerged as promising biomaterials for a wide variety of biomedical applications in the field of tissue engineering. Specifically, they allow encapsulation of biologically active molecules (drugs and genes) and cells, and patient friendly delivery into the body without painful surgery, due to their in situ formability [1–4]. Given their unique characteristics, injectable biodegradable hydrogels have been actively explored as drug delivery systems for controlled release of bioactive agents [5–7] and temporary extracellular matrices for tissue engineering [8–11]. Alginate is well known for chelating with divalent cations such as Ca2+, Mg2+, and Ba2+ to form hydrogels [13,14]. Alginate molecule is a linear copolymer covalently coupled through glycosidic linkages between two monomeric saccharide epimer units such as β-D mannuronic acid (M) and R-L-guluronic acid (G). Moreover, the molecular structure of the copolymer shows characteristic three-dimensional arrangements to create the homopolymeric blocks such as poly guluronate
⁎ Corresponding authors. Tel.: +82 41 550 3691; fax: +82 41 559 7911. E-mail addresses: [email protected] (M. Song), [email protected] (U.S. Shin).
http://dx.doi.org/10.1016/j.msec.2015.02.031 0928-4931/© 2015 Elsevier B.V. All rights reserved.
(GG) and poly mannuronate (MM). The hydrogel formation is driven by the interchain cross-linking by the cationic coordinations with carboxylate groups in two facing GG blocks (the egg-box model) [12]. Biomedical applications of alginate hydrogels are profound because of its biocompatibility and minimal toxicity [15,16]. Alginate hydrogels prepared by a traditional external setting technique using calcium salts such as CaCl2 and CaSO4 have been utilized in cell encapsulation (or immobilization) [17], drug delivery [18], wound healing dressing [19], and other applications in tissue engineering [20]. The significance of alginate as biomaterials have propelled research on inherent property enhancement and new property introduction through chemical modification [21,22]. For example, Grøndahl et al. have prepared phosphorylated alginate derivatives (PAlg) for the first time using the urea/phosphate method to evaluate the ability to induce hydroxyapatite nucleation and growth [23]. Thorough NMR studies have revealed that phosphorylation of M residues occurred mostly on the C-3 equatorial hydroxyl groups of the polysaccharide ring. However, Ca-crosslinking of phosphorylated alginates by externally added CaCl2 to form PAlg-hydrogel has failed. The reduction of molecular weight of phosphorylated alginic acid and conformational changes on phosphorylation were considered as reasons. Since calcium ions show high affinity for phosphates and carboxylates and do not cross-link with other phosphorylated alginic acid molecules (PAlg) [21], we envisioned that
140
H.-S. Kim et al. / Materials Science and Engineering C 51 (2015) 139–147
phosphorylated alginic acid calcium complexes (CaPAlg) as a new alginate derivative could be prepared without self-gelation and be efficiently utilized to form hydrogels by simple mixing with alginic acid sodium salts. Here the CaPAlg complexes provide Ca-ions and polysaccharide chains bearing PO4-ions internally. Unlike the traditional external setting technique, this strategy of internal setting technique would effectively provide calcium ions within the alginate molecules to give rise to uniform calcium ion concentrations throughout the modified molecules [24]. Herein, we report the preparation of CaPAlg and its application for in vivo injectable hydrogels. The phosphorylation of alginic acid sodium salts was attempted using H3PO4/P2O5/Et3PO4/hexanol method for the first time, and the calcium complexes (CaPAlg) were produced by acid– base reaction using Ca(OAc)2. A series of hydrogels was successfully prepared by screening various ratios of NaAlg and CaPAlg with varying setting times from ~3 to ~40 min and optimal mechanical strength. 1Hand 31P-NMR spectroscopy, FT-IR spectroscopy, ICP-OES, TGA, SEM analysis, compressive mechanical test, and in vitro cell test were employed to characterize the modified alginates and the prepared hydrogels.
2. Experimental 2.1. Materials Sodium alginate (NaAlg; Mw 1.2–1.9 × 105, M/G ratio 1.56), triethylphosphate, phosphorus pentoxide, and 1-hexanol were purchased from Sigma-Aldrich (USA), and phosphoric acid was purchased from J. T. Baker (USA). All purchased chemicals were used without further purification.
2.2. Preparation of phosphorylated alginic acid (PAlg) NaAlg was phosphorylated according to the procedure described by Amaral et al. for phosphorylation of chitosan [29]. For phosphorylation of sodium alginate using H3PO4/P3O5/Et3PO4/hexanol, the sodium alginate powder (1 g) was suspended to H3PO4 (43 mL) at 37 °C with stirring and was slowly added a separately prepared solution of P2O5 (28 g) and Et3PO4 (38 mL) in hexanol (40 mL). The mixture was stirred for 2 days (48 h) at 37 °C. The reaction mixture was poured into a beaker containing methanol of 4–500 mL to give white precipitates. The precipitates were filtered using polytetrafluoroethylene (PTFE) filter (0.5 μm pore size, 47 mm in diameter), rinsed with methanol (50 mL × 3), and dried in vacuum at room temperature for one day to give 700 mg of phosphorylated alginic acid (PAlg) as pale yellow granules in 72% yield. 2.3. Preparation of phosphorylated alginic acid calcium complexes (CaPAlg) The as-obtained PAlg powder (100 mg) was dissolved in distilled water (100 mL) and was adjusted to pH ~ 8 by slowly adding aqueous Ca(OAc)2 solution (1.0 N), by the exchange of protons of the phosphorylated alginic acid molecules with calcium ions. The solution was then dialyzed using dialysis tubing membrane (pore: 12,000 Da) for 2 days against distilled water to remove all the inorganic salts. The water was changed every 6 h. After freeze-drying, 97 mg of CaPAlg was obtained as a white powder, an 87% yield. 2.4. Preparation of in vitro injectable hydrogels from phosphorylated alginic acid calcium complex (CaPAlg) The synthesized CaPAlg was used in cross-linking of NaAlg to form a hydrogel simply by mixing immediately after short vortexing. An aqueous solution of NaAlg (2 wt.%) (solution A) and an aqueous solution of CaPAlg (2 wt.%) (solution B) were prepared separately; both (solutions A and B) were subsequently mixed in the ratios, as shown in (Table 2). Each mixture was fixated in polydimethylsiloxane (PDMS) mold (SYLGARD®) until it formed hydrogels. 2.5. Characterization of the prepared alginic acid derivatives and the in vitro injectable hydrogels Fourier transform infrared (FT-IR) spectroscopy was performed on a Perkin-Elmer Spectrum BXII FT-IR spectrometer. The IR spectra were recorded between 4000 and 500 cm−1, with 16 scans at 4 cm−1 resolution. Nuclear magnetic resonance (NMR) spectra were collected on a Bruker Ultrashield 500 PLUS NMR spectrometer for 1H, and 31P nuclei. High resolution inductively coupled plasma optical emission spectroscopy (ICP-OES) was carried out on the PAlg and CaPAlg samples to determine the degree of substitution (DS) values. Samples were digested using nitric acid and analyzed using a Perkin Elmer Optima 5300dV spectrometer, with a forward rate of 1.5 mL/min (temperature 24 ± 1 °C, relative humidity 29 ± 1%). Burgener PEEK Mira Mist nebulizer was used and the instrument was operated by taking the average of three 5 s integrations at 213.62 nm for PAlg samples and at 317.93 nm for CaPAlg samples. All samples were run in triplicate. Thermogravimetric analysis (TGA) was performed using SCINCO TGA N-1500 (sample weight 5 ± 0.1 mg, heating rate 10 °C/min, initial temperature rt). The microporous structure of the CaPAlg hydrogels was observed by scanning electron microscopy (SEM) in lyophilized hydrogel samples Table 1 The contents of P and Ca in PAlg and CaPAlg samples (100 mg × 3) by ICP-OES. Sample
Fig. 1. a) FT-IR spectra and b) thermograms of NaAlg, PAlg, and CaPAlg.
PAlg CaPAlg
P
Ca
×104 (mg/kg)
DS (mol%)
×104 (mg/kg)
DS (mol%)
1.5 ± 0.1 0.31 ± 0.02
12.5 7.8
– 7.8 ± 0.1
– 112.5
H.-S. Kim et al. / Materials Science and Engineering C 51 (2015) 139–147
(Hitachi S-4300) at a voltage of 20 kV. SEM analysis was performed in duplicate. 2.6. Compression mechanical tests Uniaxial compression measurements were performed on the hydrogels prepared at 25 °C using a mechanical testing machine (INSTRON 3344, MA, USA) in compression mode. Hydrogel samples were cut to specimens with a diameter of 9.0 mm and thickness of 6.4 mm. All measurements have been tested for 1–2 min until break at a load rate (crosshead speed) of 10 mm/min. To obtain statistically reliable results all measurements were performed on 7 samples each of four hydrogel types, CaPAlg20–80. 2.7. Measurement of gelation time Determination of gelation time was applied using the well known vial inversion test (or the flow test) [25–28]. The gelation time was determined by the point at which the mixture of solutions A and B no longer flows upon inversion of the vial. Three replicate samples were measured at room temperature and an average value was taken. 2.8. In vitro cell test of cell-encapsulated hydrogels In order to evaluate the feasibility of the in vitro injectable CaPAlg hydrogels for use as a cell encapsulation material, the viability of MC3T3s encapsulated in the hydrogels was examined using LIVE/ DEAD® Viability/Cytotoxicity Kit (Invitrogen, Carlsbad, CA) via a fluorescence microscope and MTS assay. The sub-cultured MC3T3s at a density of 1 × 106 cells were encapsulated into the hydrogels of 1 ml, and cultured for up to 3 days. All procedure was carried out in laminated
141
lamellar flow hood. The medium was completely renewed every two days during culturing. After prescribed culturing time, the cellencapsulated hydrogels were placed in a new culture plate, and washed with phosphate buffered saline (PBS) solution 3 times. Cell viability and morphology were examined post-staining. To assess the encapsulation efficiency of the hydrogel, the cell survivability after encapsulation (0-day) was measured by direct cell-counting based on the Live/Dead staining. The metabolic activity of the cell for time in culture was evaluated using MTS ((3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4sulfophenyl)-2H-tetrazolium)) assay kit. After culturing, the hydrogels were placed in new 24-well plates and then MTS solution was added to each hydrogel. After 4 h of incubation at 37 °C, colorimetric measurements were performed on a micro reader with an optical density reading at 490 nm. All cell culture experiments were performed in triplicate and the data were compared by performing a statistical analysis using oneway ANOVA at a significance level of p b 0.01. 3. Results and discussion 3.1. Preparation of phosphorylated alginic acid calcium complexes (CaPAlg) Since the phosphorylation of cellulose using H3PO4/P2O5/Et3PO4/1hexanol was first attempted by Touey and Kingsfort for the hemi synthesis of water-soluble and non-degradable flame-proof textile [30], some similar methods using P2O5/MeSO3H or H3PO4/Urea/DMF have been employed for the phosphorylation of polysaccharide such as chitosan [31,32]. It is well known that phosphorylation of chitosan by P2O5/ MeSO3H or H3PO4/Urea/DMF results in reduced molecular weight of the polymer due to degradation caused by strong acidic conditions and difficulty in separating the final product when urea was included [31,32]. However, the H3PO4/P2O5/Et3PO4/1-hexanol method for the
Fig. 2. 1H-NMR (D2O) spectra of NaAlg, PAlg, and CaPAlg: Here, Gn and Mn indicate the protons at C-n (n = 1–5) for, respectively, G and M units. The colorful circle symbols were placed to facilitate the comparison.
142
H.-S. Kim et al. / Materials Science and Engineering C 51 (2015) 139–147
phosphorylation of cellulose proceeded under milder reaction conditions and caused low degradation of the polymer [11,33]. Moreover, the applicability of the product to the field of bone tissue engineering has been shown by in vitro and in vivo cytocompatibility [34]. Based on the aforementioned advantages of phosphorylation reaction conditions and its bio-applicability, we have attempted phosphorylation of NaAlg using H3PO4/P2O5/Et3PO4 method, for the first time (see scheme S1 of Supporting material). As shown in Fig. 1 (see the full spectra in Fig. S1 of Supporting material), the as-phosphorylated alginic acid was characterized by FT-IR and TGA on comparison with sodium alginate. The FT-IR spectra of unmodified and modified samples are shown in Fig. 1a. The IR spectrum of unmodified NaAlg showed characteristic bands: at 1614 and 1416 cm−1 for C_O (asymmetric and symmetric) stretching of carboxylate, and at 1300–1090 cm−1 for C\O(H) and C\O\C (ring) vibrational modes [35,36]. After phosphorylation to PAlg, several new peaks were observed at 1740, 1630, 1244, 1084, 948, and 928 cm−1 in the IR spectrum. Peaks at 1740 and 1630 cm−1 correspond to C_O stretching of carboxylic acids indicating carboxylate ions of sodium alginate were predominantly converted into the acid form. Other peaks from phosphate group such as P_O asymmetric stretching at 1244 cm−1 and P\O\C (aliphatic) and two P\O(H) vibrational modes at 948–928 cm−1 were apparently observed indicative of the presence of phosphate on the ring as well. Disappearance of a peak at 1300 cm−1 of sodium alginate after phosphorylation, strongly supports the fact that the phosphorylation has occurred at the hydroxyl groups of the alginate ring. Overall, the FT-IR spectrum of the modified alginic acid, PAlg using H3PO4/P2O5/Et3PO4/1-hexanol method was identical to previous data on preparations by the urea/phosphate method [23]. Calcium complexes (CaPAlg) of PAlg were prepared by cation exchange reaction of PAlg with Ca(OAc)2. As shown in Fig. 1a, it was noticeable that the FT-IR spectrum of CaPAlg complexes became quite different from that of PAlg, but similar to those of the NaAlg organic salts.
Table 2 Injectable hydrogels prepared by 2 wt.% of CaPAlg solution (Solution A) and 2 wt.% of NaAlg solution (Solution B). Entry no.
Solution A (mL)
Solution B (mL)
Hydrogel code-name
1 2 3 4
0.2 0.4 0.6 0.8
0.8 0.6 0.4 0.2
CaPAlg20 CaPAlg40 CaPAlg60 CaPAlg80
The characteristic peaks of phosphorous-related functional groups (P_O, P\O\C, and P\O(H)) at 1244–928 cm−1 showed no significant changes compared with PAlg indicating the phosphates were intact on the alginate ring. Especially, the C_O related broad peaks at 1608 and 1422 cm− 1 which were shifted from 1740 and 1630 cm−1 of PAlg strongly indicated that all carboxylic acids were converted back into the carboxylate anions which form calcium complex. In order to further study the composition difference of NaAlg, PAlg, and CaPAlg, thermal decomposition patterns were examined by TGA analysis as shown in Fig. 1b. The NaAlg sample used as a reference [37, 38], showed two steps of thermal decomposition after evaporation of water. The first step for the pyrolysis of hydrocarbon moiety occurred at about 200 °C. After slight decrease of residual carbohydrate pyrolysis over the range of 250–500 °C, the second step of pyrolysis occurred at ~600 °C, remaining about 20.7 wt.% of residual mass of sodium oxide. For PAlg, several complicated steps of thermal decomposition were observed. The first step for the pyrolysis of hydrocarbon moiety occurred clearly at about 200 °C. The following steps which occurred in the range of 300–600 °C and N 600 °C might have included additional carbohydrate pyrolysis, dehydration of phosphate groups to phosphorus pentoxide, and vaporization of phosphorus pentoxide, resulting in only 0.5 wt.% of remaining polymorphous polyphosphate as a glassy
Fig. 3. a) Schematic representation for the preparation of injectable hydrogel using CaPAlg (A: red) and NaAlg (B: blue) {the inset photos show the liquid form of the precursor solutions (A and B) and the injectable property of a solution mixture (A + B for CaPAlg50) after mixing (see Fig. 3S}; b) photos of hydrogel figures prepared from different volume ratios of CaPAlg and NaAlg solutions: before (top) and after (middle) freeze drying and gelation time (down).
H.-S. Kim et al. / Materials Science and Engineering C 51 (2015) 139–147
or amorphous form. Similar to PAlg case, several steps of decomposition were observed in the thermogram of CaPAlg. After the evaporation of water, the first step for the pyrolysis of hydrocarbon moiety between 200–300 °C, and the second step for additional carbohydrate pyrolysis including dehydration of phosphate groups to phosphorus pentoxide between 300–600 °C were clearly observed. The additional weight loss appeared in the high temperature range of N800 °C is presumably attributed to the continual dehydration process of Ca(OH)2 and CaP compounds which were possibly formed at the high temperature (e.g., Ca(OH)2 → CaO; Ca(H2PO4)2 → CaHPO4 → Ca3(PO4)2). The contents of P and Ca ion in PAlg and CaPAlg were measured by elemental analysis using ICP-OES. The measurements for each of both samples (100 mg) were performed in triplicate. The contents (in ppm and mol%) of phosphorus and calcium contained per 1 mol of cyclic monosaccharide unit are summarized in Table 1. On average, PAlg samples contained about 0.125 mol of P element per 1 mol of the cyclic monosaccharide units and CaPAlg samples contained N 1.13 mol of Ca element per 1 mol of the cyclic monosaccharide units, resulting in DS value of 12.5 and 112.5 mol%, respectively. The DS value (12.5 mol%) of PAlg was comparably higher than the previous value obtained by Grøndahl's method [23]. According to a recent report for the phosphorylation of chitosan [39], among 3 available phosphorylation methods, namely H3PO4/urea, H3PO4/Et3PO4/P2O5, and P2O5/MeSO3H methods, the second method (H3PO4/Et3PO4/P2O5) had the highest phosphorylated products yield. The DS value (112.5 mol%) of Ca ion for the CaPAlg sample was found to be much higher than that (12.5 mol%) of P element for PAlg sample. Theoretically, the functional groups such as carboxylic and phosphoric acids within the PAlg molecules will allow the cation exchange in the presence of Ca(OAc)2 to form the phosphorylated alginic acid calcium complexes. Here the DS value of Ca ions for the prepared CaPAlg molecules was calculated on the basis that all the carboxylic acid functional groups within the molecules exchange their protons with calcium ions. The obtained high DS value indicated that all two cyclic monosaccharide units have a calcium ion at the carboxylate functional group sites and additionally about twelve cyclic monosaccharide units have a calcium ion at the phosphate functional group site. Consequently,
143
these results imply that all carboxylate and phosphate functional groups could actively form strong ionic bonds with the divalent cations without any obvious gelation and the individual calcium ions could be evenly dispersed on the surface of the CaPAlg polymer molecules. These chemical characteristics support that the CaPAlg complexes are potentially suitable for the preparation of injectable hydrogel. 1 H-NMR studies were performed on both PAlg and CaPAlg to identify the possible sites of phosphorylation and cation exchange with Ca2+, as shown in Fig. 2 (see the original spectra in Fig. S2 of Supporting material). The analyses of 1H-NMR spectra for the unmodified and phosphorylated alginates have been well established in the literature [23, 40,41]. Major peaks of unmodified sodium alginate and phosphatemodified alginic acid prepared by us, were assigned based on the literature value as shown in Fig. 2 [23]. It was noticed that the entire spectrum was globally shifted to the downfield with significantly increased complexity upon phosphorylation and protonation. In a related study, Grøndahl et al. reported that phosphorylation of M residues occurred predominantly at the C-3 equatorial hydroxyl groups, and phosphorylation occurred for G residues as well, but it was difficult to determine the site of phosphorylation [23]. However, the C\H protons at C-2, -3, and -5 of both M and G units for the PAlg prepared by us were strongly shifted to downfield compared to that of NaAlg (Fig. 2), while those at C-1 and -4 were weakly shifted. These results may suggest many electrical changes at C-2, -3, and -5 by phosphorylation of the hydroxyl groups and by protonation of all the carboxylate groups at C-5. Compared to Grøndahl's results, the higher complexity of peaks in the 1 H NMR spectrum from the PAlg suggests that the H3PO4/P2O5/Et3PO4 method is more effective for the phosphorylation than the urea/H3PO4 method. Furthermore, Liu et al. recently reported that the same method (H3PO4/P2O5/Et3PO4) gave a higher degree of phosphorylation on the chitosan ring than the other available methods, i.e., H3PO4/urea and P2O5/MeSO3H [39]. Therefore, it was reasonably expected that H3PO4/ P2O5/Et3PO4 method would give rise to comparably higher degree of phosphorylation of alginate than the other methods. This significant change in 1H NMR spectrum directly corresponded to the relatively higher DS value (12.5 mol% in average) of PAlg as shown in ICP-OES
Fig. 4. SEM images (a–d) of CaPAlg hydrogels: a) CaPAlg20, b) CaPAlg40, c) CaPAlg60, and d) CaPAlg80.
144
H.-S. Kim et al. / Materials Science and Engineering C 51 (2015) 139–147
analysis. In addition, the 31P-NMR spectrum of PAlg also tested as shown in Fig. S2 (see Supporting material), was identical with Grøndahl's results, of a strong signal at −0.01 ppm and a low intensity signal at −10.83 ppm, which is known as pyrophosphate [23]. No other signals were observed. Finally, 1H NMR spectrum of CaPAlg obtained after cation exchange of PAlg with calcium ions maintained the high complexity of PAlg caused by phosphorylation, while the entire spectrum was shifted back to upfield similar to NaAlg case probably due to the formation of calcium salt. When compared to the spectrum of NaAlg, the C-2 and C-3 related protons of both M and G units for the CaPAlg were still to be weakly shifted to downfield due to the phosphorylation (Fig. 2), while those at C-1, C-4 and C-5 almost returned to the chemical shift values of NaAlg. The general splitting patterns of proton peaks also were similar to those of NaAlg, but the fine patterns were still more complex than those of NaAlg due to the presence of phosphate groups (compare the peaks in the colorful circle symbols). Intramolecular and intermolecular complex networkings between the deprotonated phosphate and carboxylate groups via divalent cations probably affect the splitting patterns of proton peaks. Consequently, it could be clearly considered that the phosphorylation of the hydroxyl groups at C-2 and C-3 and the formation of calcium salt at C-2, C-3, and C-5 such as the phosphoric acid calcium salt and carboxylic acid calcium salt cause the downfield shift of protons at C-2 and C-3 for both G and M units and the more complex splitting of the proton peaks. 3.2. Preparation of injectable hydrogels from phosphorylated alginic acid calcium complex (CaPAlg) Unlike the traditional method to form hydrogels where calcium ions are diffused externally into the alginate solution [21], we have successfully prepared hydrogels by internal diffusion technique where phosphorylated alginic acid molecules bearing calcium ions (CaPAlg) form complicated networks between themselves and NaAlg molecules by using their calcium ions. Schematic representation for the preparation and the formation of injectable hydrogel using CaPAlg and NaAlg is depicted in Fig. 3a (see the time-dependent gelation property of CaPAlg50 in Fig. S3 of Supporting material). Hydrogel preparation was performed by simply mixing CaPAlg solution with NaAlg solution in different ratios as shown in Table 2. Gelation times for each sample and their morphological appearances before and after freeze drying were depicted in Fig. 3b. For example, the CaPAlg/NaAlg combinations of CaPAlg20 and CaPAlg80 hydrogels showed relatively longer gelation times (N40 min and N30 min, respectively), because one of the most important components (for example, Ca2 + and non-modified alginate chain of NaAlg) for gel-formation was insufficient in the two combinations. The optimum results were obtained from the CaPAlg/NaAlg combinations of CaPAlg40 and CaPAlg60 hydrogels showing effective gelation within 3–10 min and then good fixation in PDMS mold. To obtain the most optimal conditions for setting time and gel-formation, the preparation of CaPAlg50 hydrogel was accomplished by more fine tuning the concentration of both components as follows: 1 mL of 2 wt.% CaPAlg solution + 1 mL of 2 wt.% NaAlg solution. In that case, the amount of internally supplied Ca-ions is considered to be about 1.56 mg (0.04 mmol) within about 38.44 mg of alginate component (PAlg + NaAlg). In order to examined the possibility of gelation using the equimolar amount of NaAlg (38.44 mg/1.5 mL) and CaCl2 (0.04 mmol/0.5 mL or 4.44 mg/0.5 mL) that the CaPAlg50 hydrogel contain, we tried gelation by external addition (a traditional method), but no gelation was observed at all (see NaAlg/CaCl2 in Fig. 3b). In order to achieve optimal hydrogel formation by the external supply of Ca-ion, usually, at least 17.32 mg (0.156 mmol, about 4 times more amount than the case of CaPAlg) of CaCl2 is necessary. In this case since the gelation occurs promptly it is considered that transferring the mixture solution under the skin through a syringe is inappropriate. These encouraging results from CaPAlg40–60, such as effective gelation by the minimum quantity of Ca-ion and the optimal gelation times enabling injection under the
skin may be caused by (i) the different routes of Ca ion supply (internal vs. external), (ii) the uniform dispersibility of Ca-ions on the modified alginate molecules, (iii) the different mechanism of Ca-ion transfer from carboxylat (and/or phosphate) functional groups of CaPAlg molecules to carboxylate functional groups of NaAlg molecules, and (iv) the direct participation of the polymeric counter ions of CaPAlg bearing phosphate and carboxylate anions in networking with the normal alginate molecules. Moreover, the use of CaPAlg to produce injectable alginate-based hydrogels showed easy gelation without significant change in the volume of produced hydrogels. This also may be an important characteristic for favorable cell encapsulation since the encapsulated cells can be free from physical stress that could be potentially induced by volume change during the gelation of a hydrogel. The SEM images of the alginate hydrogels revealed highly interconnected pores cross-linked by calcium ion as shown in Fig. 4. As the amount of CaPAlg component within the hydrogels increased from 20 to 80 v/v%, the hydrogels showed several important characteristic trends as follows: i) a decrease of pore size, ii) a decrease of regularity of pore shape, and iii) an increase of roughness of pore walls. Indeed, CaPAlg20 hydrogel shows regularly networked pore structures with average pore size ~800 μm in length that are similar to the one formed by NaAlg and CaCl2. However, the other hydrogels, CaPAlg40 ~ 80, showed round-shaped pore structures with an increased irregularity and roughness and the smaller sizes ranged from 320–400 μm in diameter. The increased irregularity in the pore shapes upon increasing CaPAlg concentration was probably caused by irregular participation of P-modified alginate in the networking process with pristine alginate molecules. Overall, SEM analysis showed that hydrogel
Fig. 5. a) Compression stress–strain curves of CaPAlg40–80 hydrogels and b) the straight lines fits to the experimental data in the initial deformation range (represents the elastic modulus).
H.-S. Kim et al. / Materials Science and Engineering C 51 (2015) 139–147
CaPAlg40–60, which have shown optimal gelation time (about 3–10 min) and pore structures of 300–400 μm in diameter, may provide favorable microenvironments for cell growth for tissue engineering applications. The stress–strain curves of compression tests of CaPAlg40 and CaPAlg60 hydrogels at 25 °C are presented in Fig. 5a, whereas Fig. 5b shows the initial part of experimental curves. As seen in Fig. 5a, the incorporation of 20 and 40 wt.% CaPAlg increased the ultimate compression strength (σ) at break from about 0.6 to 6.7 kPa; however, at the same time, the increased incorporation of 60 and 80 wt.% CaPAlg component reduced the maximum compression strength (σ) of CaPAlg40 at break to, respectively, about 3.5 and 2.3 kPa. The increase of the compression strength (σ) at break from CaPAlg20 to CaPAlg40 is obviously due to the higher networking by the higher Ca ion concentration, whereas the reduction of the compression strength (σ) at break from CaPAlg40 to CaPAlg80 is obviously due to the lower networking by the lower concentration of unmodified alginate component of NaAlg. The elastic modulus (initial slope at lineal region) of the strongest hydrogels of CaPAlg40 and CaPAlg60 were determined to be 8.4 ± 0.412 kPa/mm and 4.8 ± 0.385 kPa/mm, respectively. The elastic modulus values may suggest that the CaPAlg40 hydrogel is effectively networked by the higher concentrations of Ca ion and unmodified alginate components. Consequently, the incorporation of
145
about 40 wt.% CaPAlg component into 60 wt.% NaAlg component may be the most effective for the improvement of the mechanical properties of the CaPAlg hydrogels. 3.3. Cellular responses The biocompatibility of the injectable hydrogel CaPAlg50, which has the middle composition of CaPAlg40–60, was assessed by determining cellular morphology, monitoring survivability and metabolic activity of cells encapsulated into the hydrogels. For this, the biocompatibility of the NaAlg hydrogel prepared using 50 mM CaCl2 solution was compared as a control. Fig. 6 shows the fluorescence images of cells encapsulated within the hydrogel during 3-day culture. The green spots indicate live cells. Through the fluorescence images, we investigated the cell dispersion, the cell morphology and survivability in the hydrogels. At 4 h post encapsulation (0-day), the round-shaped cells with good distribution were observed in all hydrogels, no significant different in the cell survivability between the hydrogels. During 3 days of culture period, the live cells after encapsulation were almost alive without spreading within the NaAlg hydrogel, while the CaPAlg50 hydrogel revealed some dead cells. Fig. 7 shows the cell survivability at 4 h post encapsulation and metabolic activity for several culture periods determined by MTS assay for quantitative analysis. The NaAlg hydrogel and the
Fig. 6. The fluorescence images of cells encapsulated within the hydrogels, a) pure alginate hydrogel and b) CaPAlg50 hydrogel, during different culture times (4 h and 1–3 days). The green spots indicate live cells.
146
H.-S. Kim et al. / Materials Science and Engineering C 51 (2015) 139–147
microenvironment for 3 day encapsulation of cells and the initial cell survivability for 0–1 day. Further study regarding application of CaPAlg hydrogels in the field of soft tissue engineering are currently ongoing in our laboratory. Acknowledgments This study was supported by grants (2009-0093829 (Priority Research Centers Program) and 2012-003905) from the National Research Foundation of Korea. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.msec.2015.02.031. References
Fig. 7. a) The initial cell survivability for 0–1 day and b) the cell proliferation rate in normal alginate hydrogel (NaAlg) and CaPAlg50 hydrogel for 3 days, which was normalized by the proliferation level of 0-day NaAlg hydrogel.
injectable hydrogel CaPAlg50 possessed similar initial cell survivability of approximately 95%, but the cell proliferation rate in each hydrogel for longer time, which was normalized by the proliferation level of 0day NaAlg hydrogel, was slightly different. During 3-day culture, the encapsulated cells within the NaAlg hydrogel were continuously growing (viability of 125%), whereas the cell proliferation within the injectable CaPAlg50 hydrogels decreased (viability of 71%). Consequently, the results revealed that the injectable CaPAlg40–60 hydrogels provide favorable microenvironment for 3 day encapsulation of cells and the initial cell survivability for 0–1 day. Further studies regarding improvement of the cell proliferation level of the injectable CaPAlg hydrogels for a long time period and their applications for cartilage and soft tissue engineering are currently ongoing in our laboratories. 4. Conclusions We have prepared phosphorylated alginate (PAlg) using H3PO4/ P2O5/Et3PO4/hexanol method under mild reaction conditions. PAlg was utilized in ion exchange with minimum amount of Ca(OAc)2 to form CaPAlg. Injectable hydrogel materials were successfully prepared from CaPAlg and NaAlg solutions via simple mixing to give uniform ion concentrations throughout by “internal setting” technique without externally adding calcium salts. The preparation of PAlg and CaPAlg was confirmed by instrumental analyses using 1H- and 31P-NMR, FTIR, TGA, and ICP-MS. The injectable CaPAlg hydrogels (especially CaPAlg40–60) showed optimal gelation time (about 3–40 min) and less regularly interconnected pore structures of 300–400 μm in diameter. Their mechanical properties were tested to be ≤6.7 kPa for compressive strength at break and about 8.4 kPa/mm for elastic modulus. Cell culture assay revealed that CaPAlg hydrogels provide favorable
[1] K.H. Bae, L.-S. Wang, M. Kurisawa, Injectable biodegradable hydrogels: progress and challenges, J. Mater. Chem. B 1 (2013) 5371–5388. [2] H. Singh, in: M. Ramalingam, A. Tiwari, S. Ramakrishna, H. Kobayashi (Eds.), Injectable in situ gelling hydrogels as biomaterials, Integrated Biomaterials for Biomedical Technology, 1st ed.Wiley, 2012, pp. 359–396. [3] K. Nagahama, A. Takahashi, Y. Ohya, Biodegradable polymers exhibiting temperature-responsive sol–gel transition as injectable biomedical materials, React. Funct. Polym. 73 (2013) 979–985. [4] A.A. Amini, L.S. Nair, Injectable hydrogels for bone and cartilage repair, Biomed. Mater. 7 (2012) 024105/1–024105/13. [5] M.M. Pakulska, B.G. Ballios, M.S. Shoichet, Injectable hydrogels for central nervous system therapy, Biomed. Mater. 7 (2012) 024101/1–024101/13. [6] D.J. Overstreet, D. Dutta, S.E. Stabenfeldt, B.L. Vernon, Injectable hydrogels, J. Polym. Sci. B Polym. Phys. 50 (2012) 881–903. [7] D. Macaya, M. Spector, Injectable hydrogel materials for spinal cord regeneration: a review, Biomed. Mater. 7 (2012) 012001/1–012001/22. [8] Y. Li, J. Rodrigues, H. Tomas, Injectable and biodegradable hydrogels: gelation, biodegradation and biomedical applications, Chem. Soc. Rev. 41 (2012) 2193–2221. [9] C.T. Huynh, M.K. Nguyen, D.S. Lee, Injectable block copolymer hydrogels: achievements and future challenges for biomedical applications, Macromolecules 44 (2011) 6629–6636. [10] J.-F.T. Lutz, Thermo-switchable materials prepared using the oligoethylene glycol methacrylate-platform, Adv. Mater. 23 (2011) 2237–2243. [11] B. Balakrishnan, R. Banerjee, Biopolymer-based hydrogels for cartilage tissue engineering, Chem. Rev. 111 (2011) 4453–4474. [12] K.Y. Lee, D.J. Mooney, Hydrogels for tissue engineering, Chem. Rev. 101 (2001) 1869–1879. [13] D. Seliktar, Designing cell-compatible hydrogels for biomedical applications, Science 336 (2012) 1124–1128. [14] A.M. Kloxin, A.M. Kasko, C.N. Salinas, K.S. Anseth, Photodegradable hydrogels for dynamic tuning of physical and chemical properties, Science 324 (2009) 59–63. [15] G. Orive, A.M. Carcaboso, R.M. Hernández, A.R. Gascόn, J.L. Pedraz, Biocompatibility evaluation of different alginates and alginate-based microcapsules, Biomacromolecules 6 (2005) 927–931. [16] G. Orive, S. Ponce, R.M. Hernandez, A.R. Gascon, M. Igartua, J.L. Pedraz, Biocompatibility of microcapsules for cell immobilization elaborated with different type of alginates, Biomaterials 23 (2002) 3825–3831. [17] O. Smidsrød, G. Skjåk-Bræk, Alginate as immobilization matrix for cells, Trends Biotechnol. 8 (1990) 71–78. [18] A. Shilpa, S.S. Agrawal, A.R. Ray, Controlled delivery of drugs from alginate matrix, J. Macromol. Sci. Polym. Rev. C43 (2003) 187–221. [19] J.S. Boateng, K.H. Matthews, H.N.E. Stevens, G.M. Eccleston, Wound healing dressings and drug delivery systems: a review, J. Pharm. Sci. 97 (2008) 2892–2923. [20] J.L. Drury, R.G. Dennis, D. Mooney, The tensile properties of alginate hydrogels, Biomaterials 25 (2004) 3187–3199. [21] S.N. Pawar, K.J. Edgar, Alginate derivatization: a review of chemistry, properties and applications, Biomaterials 33 (2012) 3279–3305. [22] S.N. Pawar, K.J. Edgar, Alginate esters via chemoselective carboxyl group modification, Carbohydr. Polym. 98 (2013) 1288–1296. [23] R.J. Coleman, G. Lawrie, L.K. Lambert, M. Whittaker, K.S. Jack, L. Grøndahl, Phosphorylation of alginate: synthesis, characterization, and evaluation of in vitro mineralization capacity, Biomacromolecules 12 (2011) 889–897. [24] G. Skjåk-Bræk, H. Grasdalen, O. Smidsrød, Inhomogeneous polysaccharide ionic gels, Carbohydr. Polym. 10 (1989) 31–54. [25] E.F. Banwell, E.S. Abelardo, D.J. Adams, M.A. Birchall, A. Corrigan, A.M. Donald, Rational design and application of responsive alpha-helical peptide hydrogels, Nat. Mater. 8 (2009) 596–600. [26] B.S. Luisi, K.D. Rowland, B. Moulton, Coordination polymer gels: synthesis, structure and mechanical properties of amorphous coordination polymers, Chem. Commun. 27 (2007) 2802–2804. [27] Y. Yu, Y. Ma, Breath figure fabrication of honeycomb films with small molecules through hydrogen bond mediated self-assembly, Soft Matter 7 (2011) 884–891.
H.-S. Kim et al. / Materials Science and Engineering C 51 (2015) 139–147 [28] K. Chawla, T. Yu, S.W. Liao, Z. Guan, Biodegradable and biocompatible synthetic saccharide–peptide hydrogels for three-dimensional cell culture, Biomacromolecules 12 (2011) 560–567. [29] I.F. Amaral, P.L. Granja, M.A. Barbosa, Chemical modification of chitosan by phosphorylation: an XPS, FT-IR and SEM study, J. Biomater. Sci. Polym. Ed. 16 (2005) 1575–1593. [30] J.D. Reid, L.W. Mazzeno, Preparation and properties of cellulose phosphates, Ind. Eng. Chem. 41 (1949) 2828–2831. [31] N. Nishi, S.I. Nishimura, A. Ebina, A. Tsutsumi, S. Tokura, Preparation and characterization of water soluble chitin phosphate, Int. J. Biol. Macromol. 6 (1984) 53–54. [32] D.R. Khanal, K. Miyatake, Y. Okamoto, T. Shinobu, M. Morimoto, H. Saimoto, Y. Shigemasa, S. Tokura, S. Minami, Phosphated chitin (P-chitin) exerts protective effects by restoring the deformability of polymorphonuclear neutrophil (PMN) cells, Carbohydr. Polym. 48 (2002) 305–311. [33] P.L. Granja, L. Pouységu, M. Pétraud, B. De Jéso, C. Baquey, M. Barbosa, Cellulose phosphates as biomaterials. I. Synthesis and characterization of highly phosphorylated cellulose gels, J. Appl. Polym. Sci. 82 (2001) 3341–3353. [34] J.C. Fricain, P.L. Granja, M.A. Barbosa, B. De Jéso, N. Barthe, C. Baquey, Cellulose phosphates as biomaterials. In vivo biocompatibility studies, Biomaterials 23 (2002) 971–980.
147
[35] C. Sartori, D.S. Finch, R.B., G.K., Determination of cation content of alginate thin films by FT IR spectroscopy, Polymer 38 (1997) 43–51. [36] G. Lawrie, I. Keen, B. Drew, A. Chandler-Temple, L. Rintoul, P. Fredericks, L. Grøndahl, Interactions between alginate and chitosan biopolymers characterized using FTIR and XPS, Biomacromolecules 8 (2007) 2533–2541. [37] T.S. Pathak, J.S. Kim, S.J. Lee, D.J. Baek, K.J. Paeng, Preparation of alginic acid and metal alginate from algae and their comparative study, J. Polym. Environ. 16 (2008) 198–204. [38] P. Sundarrajan, P. Eswaran, A. Marimuthu, L.B. Subhadra, P. Kannaiyan, One pot synthesis and characterization of alginate stabilized semiconductor nanoparticles, Bull. Korean Chem. Soc. 33 (2012) 3218–3224. [39] K. Wang, Q. Liu, Chemical structure analyses of phosphorylated chitosan, Carbohydr. Res. 386 (2014) 48–56. [40] H. Grasdalen, High-field, 1H-NMR spectroscopy of alginate: sequential structure and linkage conformations, Carbohydr. Res. 118 (1983) 255–260. [41] H. Grasdalen, B. Larsen, O. Smidsrød, 13C-NMR studies of monomeric composition and sequence in alginate, Carbohydr. Res. 89 (1981) 179–191.
