Materials Science and Engineering C 59 (2016) 1038–1046
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Effect of ultrasonication on the fibril-formation and gel properties of collagen from grass carp skin Ying Jiang a, Haibo Wang b,⁎, Mingxia Deng b, Zhongwen Wang a, Juntao Zhang b, Haiyin Wang c, Hanjun Zhang b a b c
College of Food Science and Engineering, Wuhan Polytechnic University, Wuhan, Hubei, China Department of Chemical and Environmental Engineering, Wuhan Polytechnic University, Wuhan, Hubei, China Tongji Medical Collage, Huzhong University of Science and Technology, Wuhan, Hubei, China
a r t i c l e
i n f o
Article history: Received 23 April 2015 Received in revised form 25 September 2015 Accepted 3 November 2015 Available online 4 November 2015 Keywords: Collagen Ultrasonic treatment Fibril formation Collagen gels Properties
a b s t r a c t Controlling the fibril-formation process of collagen in vitro to fabricate novel biomaterials is a new area in the field of collagen research. This study aimed to determine the effect of ultrasonication on collagen fibril formation and the properties of the resulting collagen gels. Native collagen, extracted from the skin of grass carp, selfassembled under ultrasonic conditions (at different ultrasonic power and duration). The self-assembly kinetics, fibrillar morphology, and physical and cell growth-promoting properties of the collagen gels were analyzed and compared. The results showed that the self-assembly rate of collagen was increased by ultrasonication at the nucleation stage. The resulting fibrils exhibited smaller diameters and D-periodicity lengths than that of the untreated collagen samples (p b 0.05). The viscoelasticity and textural properties of collagen gels also changed after ultrasonication at the nucleation stage. Texture profile analysis and cell proliferation assays showed that ultrasonication produced softer collagen gel colloids, which were more suitable for cell proliferation than the untreated collagen gels. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Collagen represents the most abundant connective tissue protein in multicellular animals including humans. Its biocompatibility and biodegradability have led to its widespread use as a biomaterial [1–3]. Although there are at least 27 different types of collagen (types I– XXVII), the most commonly utilized collagen is type I, which is a triple helix that consists of three left-handed peptide chains intertwined in a right-handed manner [4–5]. In vivo, type I collagen undergo selfassembly into microfibrils and further into fibrils, which can form even larger bundles that are organized differently in different tissues. It has been known for some time that collagen in solution can spontaneously self-assemble to form fibrillar structures, similar to those found in collagen from native tissues under ideal conditions of pH, temperature, and ionic strength. These collagen fibers formed in vitro have mechanical and biological properties similar to those of native tissues, which make them useful in tissue engineering applications such as 3D scaffolds and artificial tendons and skin [6–11]. Many studies have shown that the self-assembly kinetics of collagen and the properties of the formed collagen fibrils are strongly affected by different assembly conditions including electrolytes, pH, etc. [12–15]. Moreover, collagen self-assembly is also influenced by external factors such as gravity and ⁎ Corresponding author at: Department of Chemical and Environmental Engineering, Wuhan Polytechnic University, Changqing garden, Wuhan, Hubei, China. E-mail address: [email protected] (H. Wang).
http://dx.doi.org/10.1016/j.msec.2015.11.007 0928-4931/© 2015 Elsevier B.V. All rights reserved.
magnetic and electrical fields. Chen et al. reported that magnetic fields can influence the fibril-formation process of collagen such that it forms aligned collagen fibrils [16]. Cheng et al. reported a novel electrochemical alignment technique to control the assembly of collagen molecules into highly oriented bundles at the macroscale [17]. The main goal of this work was to investigate the effect of ultrasonication on collagen fibril formation in vitro. Ultrasound technology is based on mechanical waves at a frequency above the threshold of human hearing (N 16 kHz). These waves travel either through the bulk of a material or on its surface at a speed that is characteristic of the nature of the wave and the material through which it is propagating [18–20]. Ultrasonic energy can be transferred to a material through a process called cavitation, which is the formation, growth, and violent collapse of cavities in water. The energy provided by cavitation in this so-called sonochemistry is approximately 10–100 kJ/mol, which is within the hydrogen bond energy scale [21–23]. Ultrasonic treatment has been widely used in cleaning, rehabilitative treatment, and extraction of bioactive components [24–26]. Christine et al. reported that therapeutic ultrasound improves the strength of repaired Achilles tendons in rats [25]. Ultrasonication has also been applied to the isolation of collagen [26]. However, thus far the effect of ultrasound on the selfassembly behavior of collagen in solution has not been well described. In this study, native collagen extracted from the skin of grass carp was self-assembled under ultrasonic conditions at different ultrasonic powers and duration times. In order to understand the effect of ultrasonication on collagen fibril formation, the self-assembly kinetics
Y. Jiang et al. / Materials Science and Engineering C 59 (2016) 1038–1046
of collagen, fibrillar morphology, and physical and cell growthpromoting properties of collagen gels were analyzed and compared. The results of this study will facilitate the development of novel collagen biomaterials and help improve the biological and physical properties of collagen-based materials. 2. Materials and methods 2.1. Collagen preparation from fish skin Grass carp (Ctenopharyngodon idellus) skins were obtained from a supermarket in Wuhan (Hubei, China). The skins were removed manually, washed with chilled tap water, placed in polyethylene bags, and stored at − 25 °C until use. Collagen was prepared as described by Yang et al. [27]. 2.2. Ultrasonic treatment during collagen self-assembly Collagen was dissolved to yield a 3 g L− 1 solution in 0.5 mol L− 1 acetic acid and dialyzed against phosphate buffer (200 mM, pH 7.4) at 4 °C for 2 days. Subsequently, the collagen solution was incubated in the sample cell of an ultrasonic processor (94F0104, Kudos, Shanghai, China) at different powers at 35 kHz for various duration. To prevent an increase in temperature induced by the ultrasonic treatment, the temperature of the sample cell was controlled by a circulating water bath, maintained at 30 °C. After termination of the ultrasonication process, the collagen samples were further incubated in the sample cell at 30 °C such that the total incubation time was 60 min. The resultant collagen gels under different conditions are abbreviated as C0W0m, C100W5m, C140W5m, C180W5m, C180W15m, and C180W60m, the details of which are shown in Table 1. 2.3. Effect of ultrasonic treatment on the fibrillogenic kinetics of collagen The turbidity vs. time curves of collagen self-assembly under different conditions of ultrasonic treatment were obtained according to the method of Aukkanit and Garnjanagoonchorn [28]. The turbidity was monitored every 5 min by the absorbance at 310 nm using a UVspectrophotometer (UV-2000, Unico, Shanghai, China). In order to further study the effect of ultrasonication on collagen assembly, real-time changes in the molecular morphologies of C0W0m and C180W5m at 5, 15 and 60 min in the self-assembly process were observed by atomic force microscopy (AFM, SPM9700, Shimadzu, Japan). After drawing sample solutions from C0W0m and C180W5m, a series of dilutions ranging from 0- to 30-fold were immediately prepared by adding aliquots of the samples to an appropriate volume of deionized water in an ice–water bath. Aliquots of 20 μL diluted samples were deposited onto freshly cleaved sheets of mica and then air dried at 4 °C. Imaging was performed in tapping mode, typically using a square pyramidal silicon nitride tip with a nominal spring constant of ~ 0.58 N m− 1. All samples were imaged at room temperature and the force-mode images were obtained simultaneously. Table 1 Abbreviations and incubation conditions for different samples. Samples
Ultrasonic power (W)
Ultrasonic time at 30 °C (min)
Incubate time at 30 °C in the absence of ultrasonic (min)
C0W0m C100W5m C140W5m C180W5m C180W15m C180W60m
0 100 140 180 180 180
0 5 5 5 15 60
60 55 55 55 45 0
1039
2.4. Electron microscopy The microstructure of the collagen fibrils formed in the different conditions was observed by scanning electron microscopy (SEM; 5600LV, JEOL, Tokyo, Japan) at an acceleration voltage of 15 kV. After all samples were incubated at 30 °C for 60 min, the collagen gel was fixed with 2.5% (v/v) formalin in phosphate buffer for 12 h and dehydrated in ethanol with stepwise concentrations of 30, 50, 70, 90 and 100% (v/v) for 15 min each [29]. The samples were then dried with critical-point carbon dioxide using a Balzers CPD020 (Furstentum, Liechtenstein), mounted on a stub, and coated with gold/palladium alloy. For each sample, at least ten SEM images were randomly captured and at least ten different fibrils were selected from each image. Fibril diameters were measured from SEM images and the mean of 100 fibrils was reported in each case. The collagen fibrils with D-periodicity were also viewed by transmission electron microscopy (TEM; TEM 200CX, JEOL) using an accelerating voltage of 80 kV. TEM samples were prepared by placing the collagen suspensions on copper grids with 200 mesh size and removing excess water by placing a piece of filter paper at the edge of the grid. Then, the fibrils were negatively stained with 1% phosphotungstic acid at pH 7.4 for 15 s and the stained grids were air dried. D-periodicities were measured from TEM images and the mean of 50 D-periodicities was reported in each case. 2.5. Viscoelasticity measurement of collagen gels Assays were performed in an oscillatory rheometer (AR 2000, TA Instruments, New Castle, DE, USA), using a parallel stainless steel plate with a diameter of 40 mm and a gap of 1 mm. Linear viscoelasticity range was determined at 1 Hz and 1% deformation was selected for frequency sweeps of all samples. In the rheological experiments, the collagen samples were incubated in the sample cell of the ultrasonicator with different ultrasonic powers at 30 °C for 5 min. Then, an appropriate volume (~ 3 mL) of the collagen samples was immediately transferred to the thermostatic inferior plate, the temperature of which was controlled by a Peltier module, and incubated for 55 min at 30 °C for gelation. Subsequently, the storage modulus (Gˊ) and loss modulus (G″) were recorded as functions of frequency ranging from 0.01 to 10 Hz. 2.6. Texture profile analysis (TPA) of collagen gels Collagen gels (2.0 cm diameter and 4.0 cm height) were compressed at 25 °C to 25% of their original height until rupture in a TA-XT2i Texture Analyzer (Stable Micro Systems, Surrey, UK) using a cylindrical probe (P/0.5) at a constant velocity of 1.0 mm s− 1. The texture of collagen gels was analyzed by a uniaxial compression test of two cycles (TPA mode) and the hardness of gels was obtained from the force vs. distance curve. In each determination, five gel samples were used and average values were calculated. 2.7. Structural analysis of collagen gels All the collagen gel samples were quickly frozen in liquid nitrogen (− 196 °C) for 20 min and then freeze-dried. Cylindrical sections, 5 mm in diameter, were removed from the freeze-dried collagen samples and coated with gold ions using an ion coater prior to observation. The structure of the collagen gel samples was examined by SEM (S-3000 N, Hitachi, Tokyo, Japan) using a 15 kV accelerating voltage. The parameters of collagen gel microstructure, i.e., pore length (L) and pore width (W) were quantized from those SEM images. For each sample, at least five SEM images were randomly captured and at least twenty different pores were selected
1040
Y. Jiang et al. / Materials Science and Engineering C 59 (2016) 1038–1046
from each image. The mean of the effective size of pores was calculated using the relationship: d = (L × W)0.5 [30].
3. Results and discussion 3.1. Kinetics of collagen self-assembly
2.8. Cell culture assays The ability to promote cell proliferation of collagen fibril gels, treated under different ultrasonic powers, was assayed by the method of Jia et al. [31]. NIH 3T3 fibroblasts (CCTCC, Wuhan, China) were maintained in Dulbecco's modified Eagle's medium (DMEM, BD Biosciences, San Jose, CA, USA) containing 10% (v/v) fetal bovine serum, and 1% (w/v) penicillin-streptomycin. Cells were grown in 5% CO2 at 37 °C. Collagen samples were dissolved to yield a 3 g L−1 solution in 0.5 mol L− 1 acetic acid and dialyzed against phosphate buffer (200 mmol L− 1, pH 7.4) at 4 °C for 2 days. After that, the collagen samples were incubated in the sample cell of the ultrasonicator with different powers at 35 kHz for 5 min. The collagen samples were then immediately dispensed into a 96-well tissue culture plate (50 μL per well) and incubated for 55 min at 30 °C to form collagen gels. Subsequently, the wells were sterilized with 75% ethanol for 3–4 h, followed by solvent exchange with phosphate-buffered saline (PBS) six times. The sterilized collagen samples were hydrated with DMEM for 12 h, and the DMEM was carefully removed prior to cell seeding. After treatment with 0.25% trypsin, the cell density was adjusted to 100,000 cells/mL and seeded on each well (200 μL/well). Those wells into which no tested collagen sample was placed were used as the control. The plate was incubated at 37 °C in a humidified 5% CO2 atmosphere and the cell morphology was periodically monitored by phase-contrast microscopy. After incubation for 3 days, cell layers were rinsed with PBS, and 50 μL of MTT (Amresco, Solon, OH, USA) solution (12 mmol L−1) was added to each well. After 4 h of incubation at 37 °C, the MTT solution was removed, and the produced insoluble formazan crystals were dissolved in 150 μL of dimethyl sulfoxide (Amresco) followed by an absorbance reading at 570 nm. The higher absorbance values indicate higher cell viability. 2.9. Statistical analysis All data were subjected to analysis of variance (ANOVA), and the differences between means were evaluated by Duncan's multiple-range test. The SPSS statistics program (SPSS 11.0 for Windows, SPSS Inc., Chicago, IL, USA) was used for data analysis.
The kinetics of collagen fibrillogenesis was determined by turbidity measurement, which is the most commonly used method in the study of collagen self-assembly. The typical turbidity–time curves of collagen fibrillogenesis were sigmoidal with a lag phase, a growth phase, and a plateau phase. The lag phase is the first step without absorbance change, corresponding to the nucleation of collagen fibrils, and is also called the nucleation stage. The second step is a growth phase with rapid absorbance change, representing the growing cores of fibrils, and the third step is a maturity phase with a constant value of absorbance, indicating the formation of three-dimensional networks of fibrils [32]. The turbidity curves of collagen assembly with different powers of ultrasonic treatment are shown in Fig. 1a. The C0W0m sample displayed a typical sigmoidal curve with a lag phase of ~5 min, a growth phase of ~5–45 min, and maturity phase of ~60 min. However, the self-assembly curves of other collagen samples with ultrasonic treatment showed shorter lag and growth phases compared with C0W0m, and the lag and growth phases of C100W5m, C140W5m, and C180W5m decreased gradually in that order. These results suggest that ultrasonic treatment can increase the collagen fibrillogenic rate, and this effect is more obvious at higher ultrasonic powers. The turbidity curves of collagen assembly with different durations of ultrasonication at 180 W are shown in Fig. 1b. We found that the selfassembly curves for C180W5m, C180W15m, and C180W60m almost overlapped, which suggests that the duration of ultrasonication had no effect on the kinetics of collagen self-assembly. That is to say, the fibrillogenic rate of collagen can be accelerated by ultrasonication only at the beginning of collagen assembly. 3.2. AFM observation In order to further illustrate the impact of ultrasonication on the kinetics of collagen self-assembly, real-time changes in collagen morphology at different self-assembly stages were observed by AFM. As can be seen from the AFM images (Fig. 2), the morphology of the ultrasonicated collagen samples exhibited a visible difference from C0W0m at the nucleation (5 min) and growth (15 min) stages. In the nucleation stage, C0W0m was composed mainly of collagen monomers of ~ 1.5 nm in diameter and ~ 300 nm in length (Fig. 2a), but the ultrasonicated sample (C180W5m) was mainly composed of small
Fig. 1. Turbidity–time curves of collagen assembly with (a) different powers of ultrasonic treatment for 5 min, (b) different durations of ultrasonic treatment at 180 W.
Y. Jiang et al. / Materials Science and Engineering C 59 (2016) 1038–1046
1041
Fig. 2. AFM images of collagen morphology at different self-assembly stages obtained in tapping mode. (a), (b), and (c) are C0W0m samples that were incubated at 30 °C for 5, 15, and 60 min, respectively. (d), (e), and (f) are C180W5m samples that were incubated at 30 °C for 5, 15, and 60 min, respectively.
collagen aggregates, i.e. microfibers, of ~ 10–30 nm in diameter and ~ 600 nm in length (Fig. 2d). These results suggest that the ultrasonicated collagen had already completed nucleation by the beginning of self-assembly (after 5 min), which is consistent with the turbidity–time curve. In the growth stage of fibrils (15 min), the dominant components of C0W0m (Fig. 2b) and C180W5m (Fig. 2e) were microfibers, but the microfiber diameter of C180W5m (~ 20–60 nm) was greater than that of C0W0m (~ 15–35 nm). In this stage, the network structure of collagen fibrils had predominantly already formed, but the network structure of C180W5m was looser and more regular than that of C0W0m. In the maturity stage of fibrils (60 min), C0W0m and C180W5m were comprised of mature collagen fibers (Fig. 2c, f) and the network structure of collagen fibrils had completely formed. There was no apparent difference in the fibril morphology of these two samples, but the network structure of C0W0m was denser than that of C180W5m, which is similar to the growth stage of fibrils.
According to the results of turbidity measurement and AFM analysis, ultrasonic treatment at the nucleation stage of collagen fibrillogenesis can accelerate collagen self-assembly. The nucleation phase of collagen fibrillogenesis is at the beginning of collagen assembly (~ 5 min for C0W0m). At this phase, ultrasonication can remove gas attached to the surface of collagen molecules, thereby facilitating collagen– collagen interactions. On the other hand, ultrasonication can produce a strong mechanical oscillation through cavitation. Within the cavitation bubbles and the immediate surrounding area, violent shock waves are produced, which can increase collagen movement and the probability of molecular interaction [33]. In addition, it has been known that ultrasonic processing can promote the ions in the solution to disperse more evenly, which may also be one of the reasons for acceleration of collagen self-assembly. It has been reported that divalent phosphate ions, which is the main components of phosphate buffer, binding on collagen molecules appear to form salt bridges within
1042
Y. Jiang et al. / Materials Science and Engineering C 59 (2016) 1038–1046
regions of high excess positive charge in collagen molecular, which can promote the collagen assembling process [34]. Therefore, ultrasonication during the nucleation phase caused an increase in the collagen-assembly rate, and increasing the ultrasonic power increases the growth rate of collagen fibrillogenesis. However, collagen assembly is a spontaneous and orderly process of molecular arrangement after formation of the collagen fibril nucleus [7]. Thus, extending the time of ultrasonication after the nucleation stage of collagen fibrillogenesis has no discernible effect on the
kinetics of collagen self-assembly. Furthermore, we found that ultrasonication for extended times led to delamination of the assembly product and heterogeneous collagen gels. 3.3. Fibril morphologies SEM images of self-assembled collagen fibrils are shown in Fig. 3, in which it can be seen that all of the collagen fibrils were interwoven into
Fig. 3. SEM images of self-assembled collagen fibrils of (A) C0W0m, (B) C100W5m, (C) C140W5m, and (D) C180W5m. The fibril-diameter distribution of (a) C0W0m, (b) C100W5m, (c) C140W5m, and (d) C180W5m. Histograms of diameter distribution measured from SEM images. The x-axis is the diameter in nm, and the y-axis is the frequency. Average diameters and standard deviations are in nm.
Y. Jiang et al. / Materials Science and Engineering C 59 (2016) 1038–1046
a network that is the typical structure of collagen fibrils after fibrillogenesis. However, the network structure of collagen fibrils treated by ultrasonication at the nucleation stage was looser in comparison with C0W0m, which is consistent with the AFM results. The fibril diameter distribution shown in Fig. 3a–d indicates that the mean fibril diameters of the different collagen samples were 102.9 ± 25.3 nm (C0W0m), 92.8 ± 24 nm (C100W5m), 81.8 ± 18.7 nm (C140W5m), and 79.7 ± 14.9 nm (C180W5m) (p b 0.05), and showed a gradual downward trend. We observed the D-periodicity structure of collagen fibrils by TEM (Fig. 4), which showed that all the collagen fibrils had an obvious periodic banding pattern (D-periodicity) along each fibril, which is a characteristic fingerprint of collagen fibrils [17]. The mean lengths of Dperiodicity measured from TEM images were 66.1 ± 2.6 nm (C0W0m), 65.0 ± 2.1 nm (C100W5m), 62.2 ± 3.2 nm (C140W5m), and 62.5 ± 2.0 nm (C180W5m). Compared with C0W0m, the length of D-periodicity of C140W5m and C180W5m significantly decreased (p b 0.05). As can be seen from the results of SEM and TEM, the ultrasonic treatment did not affect the formation of collagen fibers and D-periodicity. However, a marked reduction in fibril diameter and D-periodicity length was observed in the ultrasonicated collagen samples when compared with C0W0m, and these decreases were directly related to increases in ultrasonic power. Moreover, the fibril-diameter distribution of ultrasonicated collagen was more homogeneous, which is confirmed by the results in Fig. 3a–d and the standard deviation values of the mean fibril diameters. These results indicate that ultrasonication at the nucleation stage of collagen assembly may influence the fibril-formation process of collagen. The assembly of collagen into fibrils is an orderly behavior driven by the loss of solvent molecules from the surface of the protein. It was reported that the morphology and distribution of mature collagen fibrils are decided by those of ‘early fibrils’, which appear to be intermediates
1043
in the formation of mature fibrils [7,12]. In general, the collagen molecules gathered randomly into microfibers at the nucleation stage of assembly, resulting in a nonuniform distribution of microfibril diameters. When the collagen microfibrils assembled further into fibrils, the fibril diameters were also heterogeneous. However, ultrasonication at the nucleation phase of self-assembly resulted in an increase in the collision rate of collagen as well as an increase in collagen self-assembly by the cavitation effect, which may lead to greater homogeneity of microfibers formed during the nucleation phase compared with untreated collagen. As a result, when the number of total collagen molecules is constant, increasing the number of collagen microfibers caused a reduction in the mean fibril diameter as well as diameter heterogeneity. As shown in the TEM images, the length of D-periodicity of C140W5m and C180W5m markedly decreased compared with that of C0W0m, which indicated that the molecular arrangement in those fibrils also changed. It has been shown that the structure of collagen fibrils is determined by the number and shape of the nuclei formed during the lag period, which can be dramatically influenced by the assembly conditions [29]. According to this view, changes in the length of D-periodicity of fibrils formed under ultrasonic conditions may be attributable to the influence of ultrasound on the number and shape of the nuclei; however, more data are needed to confirm this hypothesis. 3.4. Viscoelasticity of collagen gels The viscoelasticity of the collagen fibril gels was analyzed by using an oscillatory rheometer and the frequency–dependence curves of G′ and G″ moduli of different collagen gels are shown in Fig. 5. The G′ modulus describes the elasticity and the G″ modulus reflects the dissipated energy as a characteristic viscosity. As can be seen in Fig.5, the G′ moduli of all samples remained virtually unchanged over the range of measured frequencies and were much larger than the corresponding G″, which is consistent with the typical rheological properties of the gel structure
Fig. 4. TEM images of self-assembled collagen fibrils of (a) C0W0m, (b) C100W5m, (c) C140W5m, and (d) C180W5m.
1044
Y. Jiang et al. / Materials Science and Engineering C 59 (2016) 1038–1046
Fig. 5. Rheological spectra of different collagen fibril gels: (a) storage modulus (G′); (b) loss modulus (G″).
[35]. Compared with C0W0m, the G′ and G″ moduli of other samples treated with ultrasonication tended to decrease with increases in ultrasonic power, which indicates that the elasticity and viscosity of the collagen gels decreased after ultrasonication during self-assembly. Hydrogels generally form a random network of hydrophilic and hydrophobic polymers, wherein a large amount of water is retained [36–40]. Collagen gels formed by self-assembly is also a kind of hydrogel and its viscoelasticity depends on the structure and distribution of collagen fibrils. According to our SEM observations, the fibrils in the ultrasonicated collagen gels were looser and more tenuous compared with C0W0m, which is likely the main reason for the lower G′ and G″ moduli of those gels. 3.5. Textural properties of collagen gels The hardness of different collagen gels was analyzed by a texture analyzer and the results are shown in Fig. 6. The hardness of different gels was 97.6 ± 3.2 N (C0W0m), 68.9 ± 4.3 N (C100W5m), 65.8 ± 4.1 N
(C140W5m), and 62.1 ± 3.1 N (C180W5m). A gradual downward trend along with increases in ultrasonic power was observed, which indicates that a “softer” gel can be obtained after ultrasonication during the nucleation phase of collagen fibrillogenesis. The texture of gels is closely related to their network structural properties, such as the density and size and distribution uniformity of gel pores. The SEM images of different collagen gels are shown in Fig. 7, which shows that the C0W0m gel was compact and had homogeneously sized pores. However, more heterogeneous pore structures with larger pore size (Fig. 6) can be observed in C100W5m, C140W5m and C180W5m, with increases in ultrasonic power. Studies have shown that the water-holding capacity of protein gels increases after ultrasonication because a greater number of hydrophilic moieties of amino acids become exposed to the surrounding water [41], resulting in a more heterogeneous pore structure. These changes in pore structure are the primary cause of differences in collagen gel texture. Collagen gels are widely used in biomedical materials such as scaffolds in tissue engineering, delivery matrices for cells or genes in gene therapy, and injectable carriers for drug delivery [42]. The requirements for the textural properties of collagen gels in different application fields are not identical. For example, collagen gels with sufficient strength and elasticity for use as scaffolds for mechanical support may not be suitable for applications that require a soft, low-viscosity gel such as injectable drug carriers. In this study, ultrasonic treatment during collagen self-assembly was found to cause a reduction in strength and viscosity of collagen gel, making it a potentially useful tool for controlling the physical properties of collagen gels to meet the diverse needs of practical application.
3.6. Cell culture assays
Fig. 6. Hardness and mean pore size of different collagen fibril gels.
The in vitro ability to promote cell proliferation of the different collagen gel samples was investigated using NIH 3 T3 fibroblasts and an MTT assay. As shown in Fig. 8, all collagen gels exhibited a significant ability to promote cell growth compared with the nocollagen control (p b 0.05), which is consistent with the results of another study [43]. Moreover, C180W5m and C140W5m displayed a greater ability to promote cell growth than C0W0m (p b 0.05), although this was not true for C100W5m (p N 0.05).
Y. Jiang et al. / Materials Science and Engineering C 59 (2016) 1038–1046
1045
Fig. 7. SEM images of collagen gels of (a) C0W0m, (b) C100W5m, (c) C140W5m, and (d) C180W5m.
NIH 3 T3 fibroblasts are ~ 50–80 μm long and ~ 5–15 μm in diameter at the widest point (images not shown). Structural observation of collagen gels by SEM showed that the mean pore size of C180W5m and C140W5m was ~ 5.10–5.93 μm, which was more suitable for the migration of NIH3T3 cells than that of C0W0m (mean pore size, ~ 4.22 μm). It has been reported that the ability of collagen gels to regulate cell growth is influenced by physical parameters of its network structure such as pore size [44]. Generally, if the pores in the gel are too small, cell migration is limited, resulting in the formation of cellular capsules around the edges of the scaffold. In this study, the higher cell viability observed in the C180W5m and C140W5m samples may be at least partly attributable to the greater heterogeneity of their pore size compared with that of C0W0m. 4. Conclusion
Fig. 8. Cell growth after 3 days of culture on C0W0m, C100W5m, C140W5m, C180W5m, and control (no collagen). Bars (mean ± standard deviation, n = 8) with different letters have mean values that are significantly different (p b 0.05).
In this study, changes in the fibrillogenic kinetics, fibrillar morphology, and texture and biocompatibility of collagen gels after ultrasonic treatment under different conditions were measured. The present results indicate that the collagen fibrillogenic rate can be increased by ultrasonic processing at the nucleation stage of assembly and the diameters and D-periodicity lengths of the resulting fibrils were smaller
1046
Y. Jiang et al. / Materials Science and Engineering C 59 (2016) 1038–1046
and more homogeneous than those of the untreated samples. Moreover, compared with untreated collagen fibril gels, ultrasonically treated collagen gels were softer in texture and had a larger pore size, along with increased ability to promote cell proliferation. Table 1 Abbreviations and incubation conditions for different samples. Acknowledgments This study was financially supported by the National Natural Science Foundation of China (No. 21376183), Project of Application and Foundation Study of Wuhan, China (2013020501010177) and by Innovation Team Program of Hubei province, China (No. T201208). References [1] C.H. Lee, A. Singla, Y. Lee, Biomedical applications of collagen, Int. J. Pharm. 221 (2001) 1–22. [2] E.A.A. Neel, L. Bozec, J.C. Knowles, O. Syed, V. Mudera, R. Day, J.K. Hyun, Collagen — emerging collagen based therapies hit the patient, Adv. Drug Deliv. Rev. 65 (2013) 429–456. [3] S. Zeng, C. Zhang, H. Ling, P. Yang, P. Hong, Z. Jiang, Isolation and characterization of acid-solubilised collagen from the skin of Nile tilapia (Oreochromis niloticus), Food Chem. 116 (2009) 879–883. [4] B. Brodsky, J.A.M. Ramshaw, The collagen triple-helix structure, Matrix Biol. 15 (1997) 545–554. [5] S. Xu, A. Liu, Q. Chen, M. Lv, M. Yonese, H. Liu, Self-assembly nano-structure of type I collagen adsorbed on Gemini surfactant LB monolayers, Colloid. Surface. B 70 (2009) 124–131. [6] D.A. Cisneros, C. Hung, C.M. Franz, D.J. Muller, Observing growth steps of collagen self-assembly by time-lapse high-resolution atomic force microscopy, J. Struct. Biol. 154 (2006) 232–245. [7] K.E. Kadler, D.F. Holmes, J.A. Trotter, J.A. Chapman, Collagen fibril formation, Biochem. J. 316 (1996) 1–11. [8] J.K. Rainey, C.K. Wen, M.C. Goh, Hierarchical assembly and the onset of banding in fibrous long spacing collagen revealed by atomic force microscopy, Matrix Biol. 21 (2002) 647–660. [9] N.N. Fathima, A. Dhathathreyan, T. Ramasami, J. Krägel, R. Miller, Degree of crosslinking of collagen at interfaces: adhesion and shear rheological indicators, Int. J. Biol. Macromol. 48 (2011) 67–73. [10] S.W. Chang, B.P. Flynn, J.W. Ruberti, M.J. Buehler, Molecular mechanism of force induced stabilization of collagen against enzymatic breakdown, Biomaterials 33 (2012) 3852–3859. [11] K.L. Yung, C.L. Deng, Comparison of physical–chemical properties of type I collagen from different species, Food Chem. 99 (2006) 244–251. [12] M. Yan, B. Li, X. Zhao, S. Qin, Effect of concentration, pH and ionic strength on the kinetic self-assembly of acid-soluble collagen from walleye pollock (Theragra chalcogramma) skin, Food Hydrocolloid. 29 (2012) 199–204. [13] Y. Nomura, S. Toki, Y. Ishii, K. Shirai, Improvement of the material property of shark type I collagen by composing with pig type I collagen, J. Agric. Food Chem. 48 (2000) 6332–6336. [14] I. Bae, K. Osatomi, A. Yoshida, A. Yamaguchi, K. Tachibana, T. Oda, et al., Characteristics of a self-assembled fibrillar gel prepared from red stingray collagen, Fish. Sci. 75 (2009) 765–770. [15] Y. Li, A. Asadi, M.R. Margo, D.P. Elliot, pH effects on collagen fibrillogenesis in vitro: electrostatic interactions and phosphate binding, Mat. Sci. Eng. C-Mater. 29 (2009) 1643–1649. [16] S. Chen, N. Hirota, M. Okuda, M. Takeguchi, H. Kobayashi, N. Hanagata, T. Ikoma, Microstructures and rheological properties of tilapia fish-scale collagen hydrogels with aligned fibrils fabricated under magnetic fields, Acta Biomater. 7 (2011) 644–652. [17] X. Cheng, U.A. Gurkan, C.J. Dehen, M.P. Tate, H.W. Hillhouse, G.J. Simpson, O. Akkus, An electrochemical fabrication process for the assembly of anisotropically oriented collagen bundles, Biomaterials 29 (2008) 3278–3288. [18] D. Knorr, M. Zenker, V. Heinz, D.U. Lee, Applications and ultrasonics in food potential of processing, Trends Food Sci. Tech. 15 (2004) 261–266.
[19] M.J.W. Povey, Ultrasonics of food, Contemp. Phy. 39 (1998) 467–478. [20] T.J. Mason, L. Paniwnyk, J.P. Lorimer, The uses of ultrasound in food technology, Ultrason. Sonochem. 3 (1996) S253–S260. [21] P.C.S.F. Tischer, M.R. Sierakowski, H. Westfahl Jr., C.A. Tischer, Nanostructural reorganization of bacterial cellulose by ultrasonic treatment, Biomacromolecules 11 (2010) 1217–1224. [22] M.A. Bahattab, Polystyrene latex synthesized in presence of ultrasonic initiation, J. Appl. Polym. Sci. 121 (2011) 2535–2542. [23] A.C. Soria, M. Villamiel, Effect of ultrasound on the technological properties and bioactivity of food: a review, Trends Food Sci. Tech. 21 (2010) 323–331. [24] M. Locke, E. Nussbaum, Continuous and pulsed ultrasound do not increase heat shock protein 72 content, Ultrasound Med. Biol. 27 (2001) 1413–1419. [25] O.Y.N.G. Christine, Y.F.N.G. Gabriel, K.N.S. Edwina, C.P.L. Mason, Therapeutic ultrasound improves strength of achilles tendon repair in rats, Ultrasound Med. Biol. 29 (2003) 1501–1506. [26] H.K. Kim, Y.H. Kim, H.J. Park, N.H. Lee, Application of ultrasonic treatment to extraction of collagen from the skins of sea bass Lateolabrax japonicus, Fish. Sci. 79 (2013) 849–856. [27] H. Yang, H. Wang, Y. Zhao, H. Wang, H. Zhang, Effect of heat treatment on the enzymatic stability of grass carp skin collagen and its ability to form fibrilsin vitro, J. Sci. Food Agric. 95 (2015) 329–336. [28] N. Aukkanit, W. Garnjanagoonchorn, Temperature effects on type I pepsinsolubilised collagen extraction from silver-line grunt skin and its in vitro fibril self-assembly, J. Sci. Food Agric. 90 (2010) 2627. [29] L. Sang, X. Wang, Z. Chen, J. Lu, Z. Gu, X. Li, Assembly of collagen fibrillar networks in the presence of alginate, Carbohyd. Polym. 82 (2010) 1264–1270. [30] S. Zmora, R. Glicklis, S. Cohen, Tailoring the pore architecture in 3-D alginate scaffolds by controlling the freezing regime during fabrication, Biomaterials 23 (2002) 4087–4094. [31] Y. Jia, H. Wang, H. Wang, Y. Li, M. Wang, J. Zhou, Biochemical properties of skin collagens isolated from black carp (Mylopharyngodon piceus), Food Sci. Biotechnol. 21 (2012) 1585–1592. [32] P. Noitup, M.T. Morrissey, W. Garnjanagoonchorn, In vitro self-assembly of silverline grunt type I collagen: effects of collagen concentrations, pH and temperatures on collagen self-assembly, J. Food Biochem. 30 (2006) 547–555. [33] W. Chen, H. Yu, Y. Liu, P. Chen, M. Zhang, Y. Hai, Individualization of cellulose nanofibers from wood using high-intensity ultrasonication combined with chemical pretreatments, Carbohyd. Polym. 83 (2011) 1804–1811. [34] E.L. Mertz, S. Leikin, Interactions of inorganic phosphate and sulfate anions with collagen, Biochemistry 43 (2004) 14901. [35] D. Velegol, F. Lanni, Cell traction forces on soft biomaterials. I. Microtechnology of type I collagen gels, Biophys. J. 81 (2001) 1786–1792. [36] M.E. Francis-Sedlak, S. Uriel, J.C. Larson, H.P. Greisler, D.C. Venerus, E.M. Brey, Characterization of type I collagen gels modified by glycation, Biomaterials 30 (2009) 1851–1856. [37] C.A. Miles, N.C. Avery, V.V. Rodin, A.J. Bailey, The increase in denaturation temperature following cross-linking of collagen is caused by dehydration of fibres, J. Mol. Biol. 346 (2005) 551–556. [38] C.A. Miles, T.J. Sims, N.P. Camacho, A.J. Bailey, The role of the alpha 2 chain in the stabilization of the collagen type I heterotrimer: a study of the type I homotrimer in oim mouse tissues, J. Mol. Biol. 321 (2002) 797–805. [39] P. Angele, J. Abke, R. Kujat, H. Faltermeier, D. Schumanna, M. Nerlicha, et al., Influence of different collagen species on physico-chemical properties of crosslinked collagen matrix, Biomaterials 25 (2004) 2831–2841. [40] H.O. Ho, Y.T. Tsai, R.N. Chen, M.T. Sheu, Viscoelastic characterizations of acellular dermal matrix (ADM) preparations for use as injectable implants, J. Biomed. Mater. Res. 70A (2004) 83–96. [41] G. Kresic, V. Lelas, A.R. Jambrak, Z. Herceg, S.R. Brncic, Influence of novel food processing technologies on the rheological and thermophysical properties of whey proteins, J. Food Eng. 87 (2008) 64–73. [42] S. Kyle, A. Aggeli, E. Ingham, M.J. McPherson, Production of self-assembling biomaterials for tissue engineering, Trends Biotechnol. 27 (2009) 423–433. [43] G.V.N. Rathna, Gelatin hydrogels: enhanced biocompatibility, drug release and cell viability, J. Mater. Sci.: Mater. Med. 19 (2008) 2351–2358. [44] P. Friedl, E.B. Brocker, The biology of cell locomotion within three-dimensional extracellular matrix, Cell. Mol. Life Sci. 57 (2000) 41–64.
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Effect of ultrasonication on the fibril-formation and gel properties of collagen from grass carp skin Ying Jiang a, Haibo Wang b,⁎, Mingxia Deng b, Zhongwen Wang a, Juntao Zhang b, Haiyin Wang c, Hanjun Zhang b a b c
College of Food Science and Engineering, Wuhan Polytechnic University, Wuhan, Hubei, China Department of Chemical and Environmental Engineering, Wuhan Polytechnic University, Wuhan, Hubei, China Tongji Medical Collage, Huzhong University of Science and Technology, Wuhan, Hubei, China
a r t i c l e
i n f o
Article history: Received 23 April 2015 Received in revised form 25 September 2015 Accepted 3 November 2015 Available online 4 November 2015 Keywords: Collagen Ultrasonic treatment Fibril formation Collagen gels Properties
a b s t r a c t Controlling the fibril-formation process of collagen in vitro to fabricate novel biomaterials is a new area in the field of collagen research. This study aimed to determine the effect of ultrasonication on collagen fibril formation and the properties of the resulting collagen gels. Native collagen, extracted from the skin of grass carp, selfassembled under ultrasonic conditions (at different ultrasonic power and duration). The self-assembly kinetics, fibrillar morphology, and physical and cell growth-promoting properties of the collagen gels were analyzed and compared. The results showed that the self-assembly rate of collagen was increased by ultrasonication at the nucleation stage. The resulting fibrils exhibited smaller diameters and D-periodicity lengths than that of the untreated collagen samples (p b 0.05). The viscoelasticity and textural properties of collagen gels also changed after ultrasonication at the nucleation stage. Texture profile analysis and cell proliferation assays showed that ultrasonication produced softer collagen gel colloids, which were more suitable for cell proliferation than the untreated collagen gels. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Collagen represents the most abundant connective tissue protein in multicellular animals including humans. Its biocompatibility and biodegradability have led to its widespread use as a biomaterial [1–3]. Although there are at least 27 different types of collagen (types I– XXVII), the most commonly utilized collagen is type I, which is a triple helix that consists of three left-handed peptide chains intertwined in a right-handed manner [4–5]. In vivo, type I collagen undergo selfassembly into microfibrils and further into fibrils, which can form even larger bundles that are organized differently in different tissues. It has been known for some time that collagen in solution can spontaneously self-assemble to form fibrillar structures, similar to those found in collagen from native tissues under ideal conditions of pH, temperature, and ionic strength. These collagen fibers formed in vitro have mechanical and biological properties similar to those of native tissues, which make them useful in tissue engineering applications such as 3D scaffolds and artificial tendons and skin [6–11]. Many studies have shown that the self-assembly kinetics of collagen and the properties of the formed collagen fibrils are strongly affected by different assembly conditions including electrolytes, pH, etc. [12–15]. Moreover, collagen self-assembly is also influenced by external factors such as gravity and ⁎ Corresponding author at: Department of Chemical and Environmental Engineering, Wuhan Polytechnic University, Changqing garden, Wuhan, Hubei, China. E-mail address: [email protected] (H. Wang).
http://dx.doi.org/10.1016/j.msec.2015.11.007 0928-4931/© 2015 Elsevier B.V. All rights reserved.
magnetic and electrical fields. Chen et al. reported that magnetic fields can influence the fibril-formation process of collagen such that it forms aligned collagen fibrils [16]. Cheng et al. reported a novel electrochemical alignment technique to control the assembly of collagen molecules into highly oriented bundles at the macroscale [17]. The main goal of this work was to investigate the effect of ultrasonication on collagen fibril formation in vitro. Ultrasound technology is based on mechanical waves at a frequency above the threshold of human hearing (N 16 kHz). These waves travel either through the bulk of a material or on its surface at a speed that is characteristic of the nature of the wave and the material through which it is propagating [18–20]. Ultrasonic energy can be transferred to a material through a process called cavitation, which is the formation, growth, and violent collapse of cavities in water. The energy provided by cavitation in this so-called sonochemistry is approximately 10–100 kJ/mol, which is within the hydrogen bond energy scale [21–23]. Ultrasonic treatment has been widely used in cleaning, rehabilitative treatment, and extraction of bioactive components [24–26]. Christine et al. reported that therapeutic ultrasound improves the strength of repaired Achilles tendons in rats [25]. Ultrasonication has also been applied to the isolation of collagen [26]. However, thus far the effect of ultrasound on the selfassembly behavior of collagen in solution has not been well described. In this study, native collagen extracted from the skin of grass carp was self-assembled under ultrasonic conditions at different ultrasonic powers and duration times. In order to understand the effect of ultrasonication on collagen fibril formation, the self-assembly kinetics
Y. Jiang et al. / Materials Science and Engineering C 59 (2016) 1038–1046
of collagen, fibrillar morphology, and physical and cell growthpromoting properties of collagen gels were analyzed and compared. The results of this study will facilitate the development of novel collagen biomaterials and help improve the biological and physical properties of collagen-based materials. 2. Materials and methods 2.1. Collagen preparation from fish skin Grass carp (Ctenopharyngodon idellus) skins were obtained from a supermarket in Wuhan (Hubei, China). The skins were removed manually, washed with chilled tap water, placed in polyethylene bags, and stored at − 25 °C until use. Collagen was prepared as described by Yang et al. [27]. 2.2. Ultrasonic treatment during collagen self-assembly Collagen was dissolved to yield a 3 g L− 1 solution in 0.5 mol L− 1 acetic acid and dialyzed against phosphate buffer (200 mM, pH 7.4) at 4 °C for 2 days. Subsequently, the collagen solution was incubated in the sample cell of an ultrasonic processor (94F0104, Kudos, Shanghai, China) at different powers at 35 kHz for various duration. To prevent an increase in temperature induced by the ultrasonic treatment, the temperature of the sample cell was controlled by a circulating water bath, maintained at 30 °C. After termination of the ultrasonication process, the collagen samples were further incubated in the sample cell at 30 °C such that the total incubation time was 60 min. The resultant collagen gels under different conditions are abbreviated as C0W0m, C100W5m, C140W5m, C180W5m, C180W15m, and C180W60m, the details of which are shown in Table 1. 2.3. Effect of ultrasonic treatment on the fibrillogenic kinetics of collagen The turbidity vs. time curves of collagen self-assembly under different conditions of ultrasonic treatment were obtained according to the method of Aukkanit and Garnjanagoonchorn [28]. The turbidity was monitored every 5 min by the absorbance at 310 nm using a UVspectrophotometer (UV-2000, Unico, Shanghai, China). In order to further study the effect of ultrasonication on collagen assembly, real-time changes in the molecular morphologies of C0W0m and C180W5m at 5, 15 and 60 min in the self-assembly process were observed by atomic force microscopy (AFM, SPM9700, Shimadzu, Japan). After drawing sample solutions from C0W0m and C180W5m, a series of dilutions ranging from 0- to 30-fold were immediately prepared by adding aliquots of the samples to an appropriate volume of deionized water in an ice–water bath. Aliquots of 20 μL diluted samples were deposited onto freshly cleaved sheets of mica and then air dried at 4 °C. Imaging was performed in tapping mode, typically using a square pyramidal silicon nitride tip with a nominal spring constant of ~ 0.58 N m− 1. All samples were imaged at room temperature and the force-mode images were obtained simultaneously. Table 1 Abbreviations and incubation conditions for different samples. Samples
Ultrasonic power (W)
Ultrasonic time at 30 °C (min)
Incubate time at 30 °C in the absence of ultrasonic (min)
C0W0m C100W5m C140W5m C180W5m C180W15m C180W60m
0 100 140 180 180 180
0 5 5 5 15 60
60 55 55 55 45 0
1039
2.4. Electron microscopy The microstructure of the collagen fibrils formed in the different conditions was observed by scanning electron microscopy (SEM; 5600LV, JEOL, Tokyo, Japan) at an acceleration voltage of 15 kV. After all samples were incubated at 30 °C for 60 min, the collagen gel was fixed with 2.5% (v/v) formalin in phosphate buffer for 12 h and dehydrated in ethanol with stepwise concentrations of 30, 50, 70, 90 and 100% (v/v) for 15 min each [29]. The samples were then dried with critical-point carbon dioxide using a Balzers CPD020 (Furstentum, Liechtenstein), mounted on a stub, and coated with gold/palladium alloy. For each sample, at least ten SEM images were randomly captured and at least ten different fibrils were selected from each image. Fibril diameters were measured from SEM images and the mean of 100 fibrils was reported in each case. The collagen fibrils with D-periodicity were also viewed by transmission electron microscopy (TEM; TEM 200CX, JEOL) using an accelerating voltage of 80 kV. TEM samples were prepared by placing the collagen suspensions on copper grids with 200 mesh size and removing excess water by placing a piece of filter paper at the edge of the grid. Then, the fibrils were negatively stained with 1% phosphotungstic acid at pH 7.4 for 15 s and the stained grids were air dried. D-periodicities were measured from TEM images and the mean of 50 D-periodicities was reported in each case. 2.5. Viscoelasticity measurement of collagen gels Assays were performed in an oscillatory rheometer (AR 2000, TA Instruments, New Castle, DE, USA), using a parallel stainless steel plate with a diameter of 40 mm and a gap of 1 mm. Linear viscoelasticity range was determined at 1 Hz and 1% deformation was selected for frequency sweeps of all samples. In the rheological experiments, the collagen samples were incubated in the sample cell of the ultrasonicator with different ultrasonic powers at 30 °C for 5 min. Then, an appropriate volume (~ 3 mL) of the collagen samples was immediately transferred to the thermostatic inferior plate, the temperature of which was controlled by a Peltier module, and incubated for 55 min at 30 °C for gelation. Subsequently, the storage modulus (Gˊ) and loss modulus (G″) were recorded as functions of frequency ranging from 0.01 to 10 Hz. 2.6. Texture profile analysis (TPA) of collagen gels Collagen gels (2.0 cm diameter and 4.0 cm height) were compressed at 25 °C to 25% of their original height until rupture in a TA-XT2i Texture Analyzer (Stable Micro Systems, Surrey, UK) using a cylindrical probe (P/0.5) at a constant velocity of 1.0 mm s− 1. The texture of collagen gels was analyzed by a uniaxial compression test of two cycles (TPA mode) and the hardness of gels was obtained from the force vs. distance curve. In each determination, five gel samples were used and average values were calculated. 2.7. Structural analysis of collagen gels All the collagen gel samples were quickly frozen in liquid nitrogen (− 196 °C) for 20 min and then freeze-dried. Cylindrical sections, 5 mm in diameter, were removed from the freeze-dried collagen samples and coated with gold ions using an ion coater prior to observation. The structure of the collagen gel samples was examined by SEM (S-3000 N, Hitachi, Tokyo, Japan) using a 15 kV accelerating voltage. The parameters of collagen gel microstructure, i.e., pore length (L) and pore width (W) were quantized from those SEM images. For each sample, at least five SEM images were randomly captured and at least twenty different pores were selected
1040
Y. Jiang et al. / Materials Science and Engineering C 59 (2016) 1038–1046
from each image. The mean of the effective size of pores was calculated using the relationship: d = (L × W)0.5 [30].
3. Results and discussion 3.1. Kinetics of collagen self-assembly
2.8. Cell culture assays The ability to promote cell proliferation of collagen fibril gels, treated under different ultrasonic powers, was assayed by the method of Jia et al. [31]. NIH 3T3 fibroblasts (CCTCC, Wuhan, China) were maintained in Dulbecco's modified Eagle's medium (DMEM, BD Biosciences, San Jose, CA, USA) containing 10% (v/v) fetal bovine serum, and 1% (w/v) penicillin-streptomycin. Cells were grown in 5% CO2 at 37 °C. Collagen samples were dissolved to yield a 3 g L−1 solution in 0.5 mol L− 1 acetic acid and dialyzed against phosphate buffer (200 mmol L− 1, pH 7.4) at 4 °C for 2 days. After that, the collagen samples were incubated in the sample cell of the ultrasonicator with different powers at 35 kHz for 5 min. The collagen samples were then immediately dispensed into a 96-well tissue culture plate (50 μL per well) and incubated for 55 min at 30 °C to form collagen gels. Subsequently, the wells were sterilized with 75% ethanol for 3–4 h, followed by solvent exchange with phosphate-buffered saline (PBS) six times. The sterilized collagen samples were hydrated with DMEM for 12 h, and the DMEM was carefully removed prior to cell seeding. After treatment with 0.25% trypsin, the cell density was adjusted to 100,000 cells/mL and seeded on each well (200 μL/well). Those wells into which no tested collagen sample was placed were used as the control. The plate was incubated at 37 °C in a humidified 5% CO2 atmosphere and the cell morphology was periodically monitored by phase-contrast microscopy. After incubation for 3 days, cell layers were rinsed with PBS, and 50 μL of MTT (Amresco, Solon, OH, USA) solution (12 mmol L−1) was added to each well. After 4 h of incubation at 37 °C, the MTT solution was removed, and the produced insoluble formazan crystals were dissolved in 150 μL of dimethyl sulfoxide (Amresco) followed by an absorbance reading at 570 nm. The higher absorbance values indicate higher cell viability. 2.9. Statistical analysis All data were subjected to analysis of variance (ANOVA), and the differences between means were evaluated by Duncan's multiple-range test. The SPSS statistics program (SPSS 11.0 for Windows, SPSS Inc., Chicago, IL, USA) was used for data analysis.
The kinetics of collagen fibrillogenesis was determined by turbidity measurement, which is the most commonly used method in the study of collagen self-assembly. The typical turbidity–time curves of collagen fibrillogenesis were sigmoidal with a lag phase, a growth phase, and a plateau phase. The lag phase is the first step without absorbance change, corresponding to the nucleation of collagen fibrils, and is also called the nucleation stage. The second step is a growth phase with rapid absorbance change, representing the growing cores of fibrils, and the third step is a maturity phase with a constant value of absorbance, indicating the formation of three-dimensional networks of fibrils [32]. The turbidity curves of collagen assembly with different powers of ultrasonic treatment are shown in Fig. 1a. The C0W0m sample displayed a typical sigmoidal curve with a lag phase of ~5 min, a growth phase of ~5–45 min, and maturity phase of ~60 min. However, the self-assembly curves of other collagen samples with ultrasonic treatment showed shorter lag and growth phases compared with C0W0m, and the lag and growth phases of C100W5m, C140W5m, and C180W5m decreased gradually in that order. These results suggest that ultrasonic treatment can increase the collagen fibrillogenic rate, and this effect is more obvious at higher ultrasonic powers. The turbidity curves of collagen assembly with different durations of ultrasonication at 180 W are shown in Fig. 1b. We found that the selfassembly curves for C180W5m, C180W15m, and C180W60m almost overlapped, which suggests that the duration of ultrasonication had no effect on the kinetics of collagen self-assembly. That is to say, the fibrillogenic rate of collagen can be accelerated by ultrasonication only at the beginning of collagen assembly. 3.2. AFM observation In order to further illustrate the impact of ultrasonication on the kinetics of collagen self-assembly, real-time changes in collagen morphology at different self-assembly stages were observed by AFM. As can be seen from the AFM images (Fig. 2), the morphology of the ultrasonicated collagen samples exhibited a visible difference from C0W0m at the nucleation (5 min) and growth (15 min) stages. In the nucleation stage, C0W0m was composed mainly of collagen monomers of ~ 1.5 nm in diameter and ~ 300 nm in length (Fig. 2a), but the ultrasonicated sample (C180W5m) was mainly composed of small
Fig. 1. Turbidity–time curves of collagen assembly with (a) different powers of ultrasonic treatment for 5 min, (b) different durations of ultrasonic treatment at 180 W.
Y. Jiang et al. / Materials Science and Engineering C 59 (2016) 1038–1046
1041
Fig. 2. AFM images of collagen morphology at different self-assembly stages obtained in tapping mode. (a), (b), and (c) are C0W0m samples that were incubated at 30 °C for 5, 15, and 60 min, respectively. (d), (e), and (f) are C180W5m samples that were incubated at 30 °C for 5, 15, and 60 min, respectively.
collagen aggregates, i.e. microfibers, of ~ 10–30 nm in diameter and ~ 600 nm in length (Fig. 2d). These results suggest that the ultrasonicated collagen had already completed nucleation by the beginning of self-assembly (after 5 min), which is consistent with the turbidity–time curve. In the growth stage of fibrils (15 min), the dominant components of C0W0m (Fig. 2b) and C180W5m (Fig. 2e) were microfibers, but the microfiber diameter of C180W5m (~ 20–60 nm) was greater than that of C0W0m (~ 15–35 nm). In this stage, the network structure of collagen fibrils had predominantly already formed, but the network structure of C180W5m was looser and more regular than that of C0W0m. In the maturity stage of fibrils (60 min), C0W0m and C180W5m were comprised of mature collagen fibers (Fig. 2c, f) and the network structure of collagen fibrils had completely formed. There was no apparent difference in the fibril morphology of these two samples, but the network structure of C0W0m was denser than that of C180W5m, which is similar to the growth stage of fibrils.
According to the results of turbidity measurement and AFM analysis, ultrasonic treatment at the nucleation stage of collagen fibrillogenesis can accelerate collagen self-assembly. The nucleation phase of collagen fibrillogenesis is at the beginning of collagen assembly (~ 5 min for C0W0m). At this phase, ultrasonication can remove gas attached to the surface of collagen molecules, thereby facilitating collagen– collagen interactions. On the other hand, ultrasonication can produce a strong mechanical oscillation through cavitation. Within the cavitation bubbles and the immediate surrounding area, violent shock waves are produced, which can increase collagen movement and the probability of molecular interaction [33]. In addition, it has been known that ultrasonic processing can promote the ions in the solution to disperse more evenly, which may also be one of the reasons for acceleration of collagen self-assembly. It has been reported that divalent phosphate ions, which is the main components of phosphate buffer, binding on collagen molecules appear to form salt bridges within
1042
Y. Jiang et al. / Materials Science and Engineering C 59 (2016) 1038–1046
regions of high excess positive charge in collagen molecular, which can promote the collagen assembling process [34]. Therefore, ultrasonication during the nucleation phase caused an increase in the collagen-assembly rate, and increasing the ultrasonic power increases the growth rate of collagen fibrillogenesis. However, collagen assembly is a spontaneous and orderly process of molecular arrangement after formation of the collagen fibril nucleus [7]. Thus, extending the time of ultrasonication after the nucleation stage of collagen fibrillogenesis has no discernible effect on the
kinetics of collagen self-assembly. Furthermore, we found that ultrasonication for extended times led to delamination of the assembly product and heterogeneous collagen gels. 3.3. Fibril morphologies SEM images of self-assembled collagen fibrils are shown in Fig. 3, in which it can be seen that all of the collagen fibrils were interwoven into
Fig. 3. SEM images of self-assembled collagen fibrils of (A) C0W0m, (B) C100W5m, (C) C140W5m, and (D) C180W5m. The fibril-diameter distribution of (a) C0W0m, (b) C100W5m, (c) C140W5m, and (d) C180W5m. Histograms of diameter distribution measured from SEM images. The x-axis is the diameter in nm, and the y-axis is the frequency. Average diameters and standard deviations are in nm.
Y. Jiang et al. / Materials Science and Engineering C 59 (2016) 1038–1046
a network that is the typical structure of collagen fibrils after fibrillogenesis. However, the network structure of collagen fibrils treated by ultrasonication at the nucleation stage was looser in comparison with C0W0m, which is consistent with the AFM results. The fibril diameter distribution shown in Fig. 3a–d indicates that the mean fibril diameters of the different collagen samples were 102.9 ± 25.3 nm (C0W0m), 92.8 ± 24 nm (C100W5m), 81.8 ± 18.7 nm (C140W5m), and 79.7 ± 14.9 nm (C180W5m) (p b 0.05), and showed a gradual downward trend. We observed the D-periodicity structure of collagen fibrils by TEM (Fig. 4), which showed that all the collagen fibrils had an obvious periodic banding pattern (D-periodicity) along each fibril, which is a characteristic fingerprint of collagen fibrils [17]. The mean lengths of Dperiodicity measured from TEM images were 66.1 ± 2.6 nm (C0W0m), 65.0 ± 2.1 nm (C100W5m), 62.2 ± 3.2 nm (C140W5m), and 62.5 ± 2.0 nm (C180W5m). Compared with C0W0m, the length of D-periodicity of C140W5m and C180W5m significantly decreased (p b 0.05). As can be seen from the results of SEM and TEM, the ultrasonic treatment did not affect the formation of collagen fibers and D-periodicity. However, a marked reduction in fibril diameter and D-periodicity length was observed in the ultrasonicated collagen samples when compared with C0W0m, and these decreases were directly related to increases in ultrasonic power. Moreover, the fibril-diameter distribution of ultrasonicated collagen was more homogeneous, which is confirmed by the results in Fig. 3a–d and the standard deviation values of the mean fibril diameters. These results indicate that ultrasonication at the nucleation stage of collagen assembly may influence the fibril-formation process of collagen. The assembly of collagen into fibrils is an orderly behavior driven by the loss of solvent molecules from the surface of the protein. It was reported that the morphology and distribution of mature collagen fibrils are decided by those of ‘early fibrils’, which appear to be intermediates
1043
in the formation of mature fibrils [7,12]. In general, the collagen molecules gathered randomly into microfibers at the nucleation stage of assembly, resulting in a nonuniform distribution of microfibril diameters. When the collagen microfibrils assembled further into fibrils, the fibril diameters were also heterogeneous. However, ultrasonication at the nucleation phase of self-assembly resulted in an increase in the collision rate of collagen as well as an increase in collagen self-assembly by the cavitation effect, which may lead to greater homogeneity of microfibers formed during the nucleation phase compared with untreated collagen. As a result, when the number of total collagen molecules is constant, increasing the number of collagen microfibers caused a reduction in the mean fibril diameter as well as diameter heterogeneity. As shown in the TEM images, the length of D-periodicity of C140W5m and C180W5m markedly decreased compared with that of C0W0m, which indicated that the molecular arrangement in those fibrils also changed. It has been shown that the structure of collagen fibrils is determined by the number and shape of the nuclei formed during the lag period, which can be dramatically influenced by the assembly conditions [29]. According to this view, changes in the length of D-periodicity of fibrils formed under ultrasonic conditions may be attributable to the influence of ultrasound on the number and shape of the nuclei; however, more data are needed to confirm this hypothesis. 3.4. Viscoelasticity of collagen gels The viscoelasticity of the collagen fibril gels was analyzed by using an oscillatory rheometer and the frequency–dependence curves of G′ and G″ moduli of different collagen gels are shown in Fig. 5. The G′ modulus describes the elasticity and the G″ modulus reflects the dissipated energy as a characteristic viscosity. As can be seen in Fig.5, the G′ moduli of all samples remained virtually unchanged over the range of measured frequencies and were much larger than the corresponding G″, which is consistent with the typical rheological properties of the gel structure
Fig. 4. TEM images of self-assembled collagen fibrils of (a) C0W0m, (b) C100W5m, (c) C140W5m, and (d) C180W5m.
1044
Y. Jiang et al. / Materials Science and Engineering C 59 (2016) 1038–1046
Fig. 5. Rheological spectra of different collagen fibril gels: (a) storage modulus (G′); (b) loss modulus (G″).
[35]. Compared with C0W0m, the G′ and G″ moduli of other samples treated with ultrasonication tended to decrease with increases in ultrasonic power, which indicates that the elasticity and viscosity of the collagen gels decreased after ultrasonication during self-assembly. Hydrogels generally form a random network of hydrophilic and hydrophobic polymers, wherein a large amount of water is retained [36–40]. Collagen gels formed by self-assembly is also a kind of hydrogel and its viscoelasticity depends on the structure and distribution of collagen fibrils. According to our SEM observations, the fibrils in the ultrasonicated collagen gels were looser and more tenuous compared with C0W0m, which is likely the main reason for the lower G′ and G″ moduli of those gels. 3.5. Textural properties of collagen gels The hardness of different collagen gels was analyzed by a texture analyzer and the results are shown in Fig. 6. The hardness of different gels was 97.6 ± 3.2 N (C0W0m), 68.9 ± 4.3 N (C100W5m), 65.8 ± 4.1 N
(C140W5m), and 62.1 ± 3.1 N (C180W5m). A gradual downward trend along with increases in ultrasonic power was observed, which indicates that a “softer” gel can be obtained after ultrasonication during the nucleation phase of collagen fibrillogenesis. The texture of gels is closely related to their network structural properties, such as the density and size and distribution uniformity of gel pores. The SEM images of different collagen gels are shown in Fig. 7, which shows that the C0W0m gel was compact and had homogeneously sized pores. However, more heterogeneous pore structures with larger pore size (Fig. 6) can be observed in C100W5m, C140W5m and C180W5m, with increases in ultrasonic power. Studies have shown that the water-holding capacity of protein gels increases after ultrasonication because a greater number of hydrophilic moieties of amino acids become exposed to the surrounding water [41], resulting in a more heterogeneous pore structure. These changes in pore structure are the primary cause of differences in collagen gel texture. Collagen gels are widely used in biomedical materials such as scaffolds in tissue engineering, delivery matrices for cells or genes in gene therapy, and injectable carriers for drug delivery [42]. The requirements for the textural properties of collagen gels in different application fields are not identical. For example, collagen gels with sufficient strength and elasticity for use as scaffolds for mechanical support may not be suitable for applications that require a soft, low-viscosity gel such as injectable drug carriers. In this study, ultrasonic treatment during collagen self-assembly was found to cause a reduction in strength and viscosity of collagen gel, making it a potentially useful tool for controlling the physical properties of collagen gels to meet the diverse needs of practical application.
3.6. Cell culture assays
Fig. 6. Hardness and mean pore size of different collagen fibril gels.
The in vitro ability to promote cell proliferation of the different collagen gel samples was investigated using NIH 3 T3 fibroblasts and an MTT assay. As shown in Fig. 8, all collagen gels exhibited a significant ability to promote cell growth compared with the nocollagen control (p b 0.05), which is consistent with the results of another study [43]. Moreover, C180W5m and C140W5m displayed a greater ability to promote cell growth than C0W0m (p b 0.05), although this was not true for C100W5m (p N 0.05).
Y. Jiang et al. / Materials Science and Engineering C 59 (2016) 1038–1046
1045
Fig. 7. SEM images of collagen gels of (a) C0W0m, (b) C100W5m, (c) C140W5m, and (d) C180W5m.
NIH 3 T3 fibroblasts are ~ 50–80 μm long and ~ 5–15 μm in diameter at the widest point (images not shown). Structural observation of collagen gels by SEM showed that the mean pore size of C180W5m and C140W5m was ~ 5.10–5.93 μm, which was more suitable for the migration of NIH3T3 cells than that of C0W0m (mean pore size, ~ 4.22 μm). It has been reported that the ability of collagen gels to regulate cell growth is influenced by physical parameters of its network structure such as pore size [44]. Generally, if the pores in the gel are too small, cell migration is limited, resulting in the formation of cellular capsules around the edges of the scaffold. In this study, the higher cell viability observed in the C180W5m and C140W5m samples may be at least partly attributable to the greater heterogeneity of their pore size compared with that of C0W0m. 4. Conclusion
Fig. 8. Cell growth after 3 days of culture on C0W0m, C100W5m, C140W5m, C180W5m, and control (no collagen). Bars (mean ± standard deviation, n = 8) with different letters have mean values that are significantly different (p b 0.05).
In this study, changes in the fibrillogenic kinetics, fibrillar morphology, and texture and biocompatibility of collagen gels after ultrasonic treatment under different conditions were measured. The present results indicate that the collagen fibrillogenic rate can be increased by ultrasonic processing at the nucleation stage of assembly and the diameters and D-periodicity lengths of the resulting fibrils were smaller
1046
Y. Jiang et al. / Materials Science and Engineering C 59 (2016) 1038–1046
and more homogeneous than those of the untreated samples. Moreover, compared with untreated collagen fibril gels, ultrasonically treated collagen gels were softer in texture and had a larger pore size, along with increased ability to promote cell proliferation. Table 1 Abbreviations and incubation conditions for different samples. Acknowledgments This study was financially supported by the National Natural Science Foundation of China (No. 21376183), Project of Application and Foundation Study of Wuhan, China (2013020501010177) and by Innovation Team Program of Hubei province, China (No. T201208). References [1] C.H. Lee, A. Singla, Y. Lee, Biomedical applications of collagen, Int. J. Pharm. 221 (2001) 1–22. [2] E.A.A. Neel, L. Bozec, J.C. Knowles, O. Syed, V. Mudera, R. Day, J.K. Hyun, Collagen — emerging collagen based therapies hit the patient, Adv. Drug Deliv. Rev. 65 (2013) 429–456. [3] S. Zeng, C. Zhang, H. Ling, P. Yang, P. Hong, Z. Jiang, Isolation and characterization of acid-solubilised collagen from the skin of Nile tilapia (Oreochromis niloticus), Food Chem. 116 (2009) 879–883. [4] B. Brodsky, J.A.M. Ramshaw, The collagen triple-helix structure, Matrix Biol. 15 (1997) 545–554. [5] S. Xu, A. Liu, Q. Chen, M. Lv, M. Yonese, H. Liu, Self-assembly nano-structure of type I collagen adsorbed on Gemini surfactant LB monolayers, Colloid. Surface. B 70 (2009) 124–131. [6] D.A. Cisneros, C. Hung, C.M. Franz, D.J. Muller, Observing growth steps of collagen self-assembly by time-lapse high-resolution atomic force microscopy, J. Struct. Biol. 154 (2006) 232–245. [7] K.E. Kadler, D.F. Holmes, J.A. Trotter, J.A. Chapman, Collagen fibril formation, Biochem. J. 316 (1996) 1–11. [8] J.K. Rainey, C.K. Wen, M.C. Goh, Hierarchical assembly and the onset of banding in fibrous long spacing collagen revealed by atomic force microscopy, Matrix Biol. 21 (2002) 647–660. [9] N.N. Fathima, A. Dhathathreyan, T. Ramasami, J. Krägel, R. Miller, Degree of crosslinking of collagen at interfaces: adhesion and shear rheological indicators, Int. J. Biol. Macromol. 48 (2011) 67–73. [10] S.W. Chang, B.P. Flynn, J.W. Ruberti, M.J. Buehler, Molecular mechanism of force induced stabilization of collagen against enzymatic breakdown, Biomaterials 33 (2012) 3852–3859. [11] K.L. Yung, C.L. Deng, Comparison of physical–chemical properties of type I collagen from different species, Food Chem. 99 (2006) 244–251. [12] M. Yan, B. Li, X. Zhao, S. Qin, Effect of concentration, pH and ionic strength on the kinetic self-assembly of acid-soluble collagen from walleye pollock (Theragra chalcogramma) skin, Food Hydrocolloid. 29 (2012) 199–204. [13] Y. Nomura, S. Toki, Y. Ishii, K. Shirai, Improvement of the material property of shark type I collagen by composing with pig type I collagen, J. Agric. Food Chem. 48 (2000) 6332–6336. [14] I. Bae, K. Osatomi, A. Yoshida, A. Yamaguchi, K. Tachibana, T. Oda, et al., Characteristics of a self-assembled fibrillar gel prepared from red stingray collagen, Fish. Sci. 75 (2009) 765–770. [15] Y. Li, A. Asadi, M.R. Margo, D.P. Elliot, pH effects on collagen fibrillogenesis in vitro: electrostatic interactions and phosphate binding, Mat. Sci. Eng. C-Mater. 29 (2009) 1643–1649. [16] S. Chen, N. Hirota, M. Okuda, M. Takeguchi, H. Kobayashi, N. Hanagata, T. Ikoma, Microstructures and rheological properties of tilapia fish-scale collagen hydrogels with aligned fibrils fabricated under magnetic fields, Acta Biomater. 7 (2011) 644–652. [17] X. Cheng, U.A. Gurkan, C.J. Dehen, M.P. Tate, H.W. Hillhouse, G.J. Simpson, O. Akkus, An electrochemical fabrication process for the assembly of anisotropically oriented collagen bundles, Biomaterials 29 (2008) 3278–3288. [18] D. Knorr, M. Zenker, V. Heinz, D.U. Lee, Applications and ultrasonics in food potential of processing, Trends Food Sci. Tech. 15 (2004) 261–266.
[19] M.J.W. Povey, Ultrasonics of food, Contemp. Phy. 39 (1998) 467–478. [20] T.J. Mason, L. Paniwnyk, J.P. Lorimer, The uses of ultrasound in food technology, Ultrason. Sonochem. 3 (1996) S253–S260. [21] P.C.S.F. Tischer, M.R. Sierakowski, H. Westfahl Jr., C.A. Tischer, Nanostructural reorganization of bacterial cellulose by ultrasonic treatment, Biomacromolecules 11 (2010) 1217–1224. [22] M.A. Bahattab, Polystyrene latex synthesized in presence of ultrasonic initiation, J. Appl. Polym. Sci. 121 (2011) 2535–2542. [23] A.C. Soria, M. Villamiel, Effect of ultrasound on the technological properties and bioactivity of food: a review, Trends Food Sci. Tech. 21 (2010) 323–331. [24] M. Locke, E. Nussbaum, Continuous and pulsed ultrasound do not increase heat shock protein 72 content, Ultrasound Med. Biol. 27 (2001) 1413–1419. [25] O.Y.N.G. Christine, Y.F.N.G. Gabriel, K.N.S. Edwina, C.P.L. Mason, Therapeutic ultrasound improves strength of achilles tendon repair in rats, Ultrasound Med. Biol. 29 (2003) 1501–1506. [26] H.K. Kim, Y.H. Kim, H.J. Park, N.H. Lee, Application of ultrasonic treatment to extraction of collagen from the skins of sea bass Lateolabrax japonicus, Fish. Sci. 79 (2013) 849–856. [27] H. Yang, H. Wang, Y. Zhao, H. Wang, H. Zhang, Effect of heat treatment on the enzymatic stability of grass carp skin collagen and its ability to form fibrilsin vitro, J. Sci. Food Agric. 95 (2015) 329–336. [28] N. Aukkanit, W. Garnjanagoonchorn, Temperature effects on type I pepsinsolubilised collagen extraction from silver-line grunt skin and its in vitro fibril self-assembly, J. Sci. Food Agric. 90 (2010) 2627. [29] L. Sang, X. Wang, Z. Chen, J. Lu, Z. Gu, X. Li, Assembly of collagen fibrillar networks in the presence of alginate, Carbohyd. Polym. 82 (2010) 1264–1270. [30] S. Zmora, R. Glicklis, S. Cohen, Tailoring the pore architecture in 3-D alginate scaffolds by controlling the freezing regime during fabrication, Biomaterials 23 (2002) 4087–4094. [31] Y. Jia, H. Wang, H. Wang, Y. Li, M. Wang, J. Zhou, Biochemical properties of skin collagens isolated from black carp (Mylopharyngodon piceus), Food Sci. Biotechnol. 21 (2012) 1585–1592. [32] P. Noitup, M.T. Morrissey, W. Garnjanagoonchorn, In vitro self-assembly of silverline grunt type I collagen: effects of collagen concentrations, pH and temperatures on collagen self-assembly, J. Food Biochem. 30 (2006) 547–555. [33] W. Chen, H. Yu, Y. Liu, P. Chen, M. Zhang, Y. Hai, Individualization of cellulose nanofibers from wood using high-intensity ultrasonication combined with chemical pretreatments, Carbohyd. Polym. 83 (2011) 1804–1811. [34] E.L. Mertz, S. Leikin, Interactions of inorganic phosphate and sulfate anions with collagen, Biochemistry 43 (2004) 14901. [35] D. Velegol, F. Lanni, Cell traction forces on soft biomaterials. I. Microtechnology of type I collagen gels, Biophys. J. 81 (2001) 1786–1792. [36] M.E. Francis-Sedlak, S. Uriel, J.C. Larson, H.P. Greisler, D.C. Venerus, E.M. Brey, Characterization of type I collagen gels modified by glycation, Biomaterials 30 (2009) 1851–1856. [37] C.A. Miles, N.C. Avery, V.V. Rodin, A.J. Bailey, The increase in denaturation temperature following cross-linking of collagen is caused by dehydration of fibres, J. Mol. Biol. 346 (2005) 551–556. [38] C.A. Miles, T.J. Sims, N.P. Camacho, A.J. Bailey, The role of the alpha 2 chain in the stabilization of the collagen type I heterotrimer: a study of the type I homotrimer in oim mouse tissues, J. Mol. Biol. 321 (2002) 797–805. [39] P. Angele, J. Abke, R. Kujat, H. Faltermeier, D. Schumanna, M. Nerlicha, et al., Influence of different collagen species on physico-chemical properties of crosslinked collagen matrix, Biomaterials 25 (2004) 2831–2841. [40] H.O. Ho, Y.T. Tsai, R.N. Chen, M.T. Sheu, Viscoelastic characterizations of acellular dermal matrix (ADM) preparations for use as injectable implants, J. Biomed. Mater. Res. 70A (2004) 83–96. [41] G. Kresic, V. Lelas, A.R. Jambrak, Z. Herceg, S.R. Brncic, Influence of novel food processing technologies on the rheological and thermophysical properties of whey proteins, J. Food Eng. 87 (2008) 64–73. [42] S. Kyle, A. Aggeli, E. Ingham, M.J. McPherson, Production of self-assembling biomaterials for tissue engineering, Trends Biotechnol. 27 (2009) 423–433. [43] G.V.N. Rathna, Gelatin hydrogels: enhanced biocompatibility, drug release and cell viability, J. Mater. Sci.: Mater. Med. 19 (2008) 2351–2358. [44] P. Friedl, E.B. Brocker, The biology of cell locomotion within three-dimensional extracellular matrix, Cell. Mol. Life Sci. 57 (2000) 41–64.
