Materials Science and Engineering C 54 (2015) 14–19
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Cross-linked chitosan improves the mechanical properties of calcium phosphate–chitosan cement Ashkan Aryaei a, Jason Liu b, Ahalapitiya.H. Jayatissa a, A. Champa Jayasuriya c,⁎ a b c
Department of Mechanical Engineering, University of Toledo, Toledo, OH 43606, USA School of Medicine, University of Toledo, OH 43614, USA Department of Orthopaedic Surgery, University of Toledo, Toledo, OH 43614, USA
a r t i c l e
i n f o
Article history: Received 9 April 2014 Received in revised form 26 February 2015 Accepted 21 April 2015 Available online 22 April 2015 Keywords: Calcium phosphate cement Chitosan Cross-linking Mechanical properties Porosity
a b s t r a c t Calcium phosphate (CaP) cements are highly applicable and valuable materials for filling bone defects by minimally invasive procedures. The chitosan (CS) biopolymer is also considered as one of the promising biomaterial candidates in bone tissue engineering. In the present study, some key features of CaP–CS were significantly improved by developing a novel CaP–CS composite. For this purpose, CS was the first cross-linked with tripolyphosphate (TPP) and then mixed with CaP matrix. A group of CaP–CS samples without cross-linking was also prepared. Samples were fabricated and tested based on the known standards. Additionally, the effect of different powder (P) to liquid (L) ratios was also investigated. Both cross-linked and uncross-linked CaP–CS samples showed excellent washout resistance. The most significant effects were observed on Young's modulus and compressive strength in wet condition as well as surface hardness. In dry conditions, the Young's modulus of cross-linked samples was slightly improved. Based on the presented results, cross-linking does not have a significant effect on porosity. As expected, by increasing the P/L ratio of a sample, ductility and injectability were decreased. However, in the most cases, mechanical properties were enhanced. The results have shown that cross-linking can improve the mechanical properties of CaP–CS and hence it can be used for bone tissue engineering applications. © 2015 Elsevier B.V. All rights reserved.
1. Introduction In tissue engineering, chitosan (CS), derived from chitin has been explored on several fronts as an organic substance to complex with calcium phosphate (CaP). Part of what makes chitosan a desirable to use substance is its biodegradability, which allows the body to break it down after healing [1,2], and biocompatibility, which includes minimal foreign body reaction while allowing human cells to grow onto it [3–6]. With such biocompatibility, CS is an ideal substance to complex into a scaffold with other additives [7]. One of the promising candidates for bone paste biomaterials is CaP, the main component of the inorganic portions of bone. CaP cements have gained clinical acceptance as valuable bone replacing biomaterials for almost three decades [8]. These scaffolds would form the basis of a moldable cement intended for bone regeneration applications while also allowing cells such as osteoblasts to grow normally onto them. Porosity, set time and mechanical properties such as, flexural strength, tensile strength, and hardness have been explored in vitro and in vivo. For the sake of clinical
⁎ Corresponding author at: University of Toledo, Department of Orthopaedic Surgery, 3065 Arlington Avenue, Dowling Hall # 2447, Toledo, OH 43614-5807, USA. E-mail address: [email protected] (A. Champa Jayasuriya).
http://dx.doi.org/10.1016/j.msec.2015.04.024 0928-4931/© 2015 Elsevier B.V. All rights reserved.
applications, traits such as washout resistance and injectability have also been heavily explored [9–13]. These properties have a significant effect on the feasibility of injectable cements in bone tissue engineering. CaP cements have drawbacks in terms of graft migration, brittleness, fatigue fracture, and handling difficulties etc. These properties need to be improved for clinical applications. For instance, washout resistance holds significance due to the intended environment of the CaP–CS complexes, which generally involves being inside the human body. The other important property is injectability, which involves the force required to move the CaP–CS cement through a syringe. Injectable CaP–CS cement can be administered to bone defects using a minimally invasive technique. CaP–CS scaffolds have been created with a variety of different methods to further explore possible methods for enhancements of their mechanical properties. The more recent methods have involved using nanoparticles or nanofibers [14–17]. Other methods involved precipitating beads of CaP–CS complexes out of a solution and lyophilization freeze drying [18]. Also the fabrication process has been heavily varied by the addition of different additives including organic acids, epoxides, and celluloses [19–21]. The addition of the additives has enhanced mechanical properties, injectability, and washout. On a chemical level, some of these additives have even induced crosslinking between chitosan.
A. Aryaei et al. / Materials Science and Engineering C 54 (2015) 14–19
Nonetheless, cross-linked CS containing CaP–CS scaffolds have not been fully explored. CS contains abundant amino and hydroxyl groups which enable it to form particles via both physical and chemical crosslinking. Ionic cross-linking of CS is a typical non-covalent interaction, which can be realized by association with negatively charged multivalent ions such as tripolyphosphate (TPP) [22]. This paper seeks to focus on the effects of cross-linking of CS in the CaP–CS scaffold and explore its effects on Young's modulus in both dry and wet conditions, porosity, and hardness. Moreover, different powder to liquid ratios (P/L) were examined. Also taken into account were washout resistance and injectability. 2. Materials and methods 2.1. Materials All components of CaP cement were purchased from Fisher Scientific. CS (85%, deacetylated), TPP and other chemical materials were purchased from Sigma-Aldrich. Statistical analysis was done using oneway ANOVA and p b 0.05 was considered as statistically significant. For each group, three samples were tested. 2.2. Scaffold fabrication The CaP blend was created with the intention of mimicking the inorganic component of bone. The blend comprised of 55% alpha tricalcium phosphate (α-TCP), 45% dicalcium phosphate anhydrous (DCPA), and 15% calcium carbonate (CaCO3) similar to that which was previously reported [23]. The CS solution was mixed at 2 wt.% with acetic acid or a ratio of 200 mg per 10 mL of acetic acid. The solution was stirred with a magnetic stirrer for 20 min and then set for 20 min allowing the air bubbles to disappear. A CaP powder blend was mixed with the CS solution at powder to liquid mass ratios (P/L) of 1.5, 1.7, and 2.0. The mixing was done manually until a consistent paste-like substance was formed. The paste-like substance was then put into cylindrical molds with dimensions of approximately d = ~ 6 mm and l = ~12 mm based on the ASTM-C39-05 standard. The molds were placed between two glass slides at 100% relative humidity and stored at 37 °C for 4 h. After 4 h, the samples were removed from the molds and placed into a saliva-like solution (SLS) for 20 h at 37 °C. SLS comprised 1.2 mM CaCl2, 0.72 mM KH2PO4, 30 mM KCl, and 50 mM HEPES buffer (4-(2hydroxyethyl)-1-piperazineethanesulfonic acid). This buffer was previously used to mimic plasma in teeth environment [9]. However, phosphate buffered saline (PBS) is another candidate to replace SLS to accurately mimic bone environment. After 20 h, the samples were removed from the SLS and they were ready for testing. The purpose of fabricating CaP–CS scaffolds using the molds is to measure the mechanical properties of CaP–CS cement. 2.3. Cross-linked scaffolds To create cross-linked samples, instead of placing the samples in the molding between glass slides at 100% humidity at 37 °C for 4 h, they were only placed in SLS for 2 h. After 2 h, the samples were moved into 80% (wt) TPP solution for 2 h at 37 °C to induce the chitosan within the scaffolds to cross-link at a molecular level. Samples had a high contact area to the TPP solution and based on the calculated porosity, it is assumed that the TPP solution penetrates into the entire sample. This time is fairly enough to form cross-linking between CS molecules [24]. Similar to uncross-linked scaffolds, the CaP–CS cross-linked scaffolds were wetted by immersing in the SLS solution for 20 h. 2.4. Washout resistance Ideally, a washout resistance test should be done after mixing and preparing paste-like CaP–CS but the test was done after 4 h which was
15
specified for cross-linking CS using TPP. Washout resistance was indicated by the sample maintaining integrity and not being dissolved by the SLS. Samples were carefully moved to a beaker with 20 mL of SLS and kept at room temperature for about 1 h. Then the washout properties were investigated. 2.5. Compressive strength and injectability testing Using the ADMET 2611 mechanical testing machine with an MTestQuatro controller, compressive strength and injectability tests were performed. For the compression test, the crosshead speed was set at 1 mm/min. This speed is suitable for CaP cements reported before [5]. Compressive moduli were tested for both wet and dry samples. Wet samples were the samples that were immediately taken out of the SLS. The dry samples were taken out of the SLS and incubated in a dry environment for 24 h to allow the fluid (mainly SLS) inside to evaporate out. This step was done for both cross-linked and uncross-linked groups. By attaching an appropriate custom made grip, an injectability test was also performed. A syringe was fitted to the machine and filled to about 5 mL of the CaP–CS paste. The force required for expulsion from the syringe was then tested. Since this test was done immediately after making CaP–CS paste, it was only done with uncross-linked samples. Injection speed was set at 6 mm/min. CaP cement pastes are usually considered as non-Newtonian fluid. However, the Hagen– Poiseuille relationship can be used to determine the viscosity [25]. 2.6. Porosity For porosity, samples were initially fully dry. They were then massed and had their exact dimensions measured. Each sample was then immersed into 25 mL of deionized H2O. After 2 h of immersion, the samples were removed and massed again. The mass difference was used to calculate the volume of H2O absorbed by each sample at 24 °C. The amount of H2O absorbed was placed in a ratio with the volume of absorbed the CaP–CS scaffold itself (VVwater ) to derive a dimensionless value Ch‐CaP scaffold
that indicated degree of porosity [13]. 2.7. Hardness For hardness test, dried samples were brought before a CM400AT Clark Microhardness tester. To properly prepare the samples, each was sandpapered at the round ends to create a flatter surface. The round ends were also marked with a marker to add contrast so as to better view with a microscope. The machine would generate small indentations at various forces. The applied force was 200 g force (gf). The indentations' dimensions, both horizontal and vertical diameter, would then be measured to generate a hardness value. Final hardness value can be calculated as follows: HV ¼
1:854F : d2
In this equation, HV is the value of Vicker's hardness, F is the adjusted load and d is the average of square's diagonals. 3. Results 3.1. Washout resistance After 4 h, almost all samples were easily removed from the mold and upon placement into the SLS solution, none of the CaP–CS scaffolds washed out. They maintained structural integrity and did not dissolve into the solution throughout the incubation time as shown in Fig. 1. Particularly, this figure shows the integrity of the samples with P/L = 1.7. For other samples with different P/L ratios, the same behavior was
16
A. Aryaei et al. / Materials Science and Engineering C 54 (2015) 14–19
Fig. 1. Washout resistance of samples after 1 h submerging in SLS.
observed. However, washout resistance for just CaP was low and the samples were crashed and disintegrated in SLS after a few minutes. This observation shows that having chitosan polymer in CaP enhances the integration of samples. 3.2. Compressive strength An increased P/L typically indicated stronger compressibility [26]. Cross-linking using the TPP solution enhanced mechanical properties, much more noticeably in the wet state of the CaP–CS scaffolds. Crosslinking CaP–CS paste using the TPP solution significantly increased the strength and Young's modulus. Fig. 2 shows the stress–displacement curve obtained from a compression test for samples in wet and dry conditions. As expected, by increasing the P/L ratio, the Young's modulus and compressive strength were increased [27]. The maximum compressive strengths for cross-linked CaP–CS samples in wet condition were 0.57, 0.73 and 1.28 MPa for P/L = 1.5, 1.7 and 2, respectively. Similarly, the values for uncross-linked CaP–CS samples in wet condition were 0.03, 0.06 and 0.27 MPa for P/L = 1.5, 1.7 and 2, respectively. It is worth noticing that a significant difference was observed between
cross-linked and uncross-linked samples. Adding a cross-linked polymer such as CS can drastically increase the Young's modulus and compressive strength. For the dry state, as shown in Fig. 3, cross-linking showed mixed results with regards to hardness. The Young's modulus of the CaP–CS scaffolds made with the above method typically were in the 106 Pa range. Fig. 3 shows the Young's modulus for samples in dry condition. Samples were completely crashed at the end of the test due to the low ductility. Similar to the wet condition, by increasing the P/L ratio, Young's modulus and compressive strength increased. For example, Young's modulus were 39.53, 45.96 and 76.31 MPa for P/L = 1.5, 1.7 and 2 in cross-linked samples, respectively. Overall, although the difference is not significant, similar to the wet condition, cross-linking enhanced the Young's modulus. 3.3. Injectability For injectability, only the uncross-linked samples were tested as shown in Fig. 4. This was because of the procedure that was followed to make cross-linked samples as explained above. At P/L = 2.0, the test failed. The syringe broke before any amount of the paste could be
Fig. 2. Stress–displacement curve for (A) cross-linked samples, and (B) uncross-linked samples. (C) Comparative results of the compression test in wet condition. * shows significant difference between the cross-linked and uncross-linked samples.
A. Aryaei et al. / Materials Science and Engineering C 54 (2015) 14–19
17
Fig. 3. Stress–displacement curve for (A) cross-linked samples, and (B) uncross-linked samples. (C) Comparative results of the compression test in dry condition.
expelled from the syringe due to the high viscosity of the fabricated paste. Generally, this followed the trend of increased P/L leading to decreased injectability [28], which would correspond to a higher injection force. For samples which we could do the injection test, the injected paste was coherent without any disintegration (Fig. 4). In addition, it showed great washout resistance after the injection. 3.4. Porosity The results of the porosity are shown in Fig. 5. Increased P/L ratios decreased the porosity while CS cross-linking seemed to have made a marginal effect of the porosity based on the results. Porosity variation between all samples was less than 10%. The mass of H2O absorbed was converted to a volume in mL based on its density at room temperature (24 °C) of 0.997 g/mL. These were the values used to generate the dimensionless porosity value. 3.5. Hardness Increased P/L ratios typically increase the hardness. CS cross-linking also significantly increased the hardness in comparison with uncrosslinked samples. For all P/L ratios, after cross-linking the CaP–CS sample, the hardness was at least doubled (Fig. 6). The surface hardness for uncross-linked CaP–CS samples were 11, 17.67 and 21 kgf/mm2 for P/L = 1.5, 1.7 and 2, respectively. However, after cross-linking, the
surface hardness changed to 33.1, 34.2 and 49.9 kgf/mm2 for P/L = 1.5, 1.7 and 2, respectively. 4. Discussion Both CS and CaP are common biomaterials in bone tissue engineering. Different papers have reported an enhancement in the mechanical properties of the mixture of CaP and CS composite biomaterials [12,13,29]. In addition, it has been shown that cross-linking CS has higher elastic modulus and hardness [24]. Our group previously showed that cross-linked CS with certain concentrations of TPP is not cytotoxic for cells [30] and CS mixed with other forms of materials such as carbon nanotubes and Zinc oxide nanoparticles is cytocompatible with osteoblasts [31,32]. The goal of this investigation was to explore the effect of cross-linking on mechanical properties of CaP–CS composite biomaterial. In the present study, a new method was applied to cross-link CS in CaP–CS matrix. Samples were made based on available standards. Fabricated samples, showed excellent cohesion (i.e., washout resistance), which was achieved by composition of the CS in CaP. We have shown that the Young's modulus of cross-linked CaP–CS samples was significantly higher than that of uncross-linked samples. The different components of CaP–CS also play an important role on the mechanical properties of final CaP–CS cement. As it was observed from Fig. 2, the ductility of samples was slightly decreased. This can be interpreted as
Fig. 4. (A) Injection test results for uncross-linked samples. (B) Washout resistance after injection.
18
A. Aryaei et al. / Materials Science and Engineering C 54 (2015) 14–19
Fig. 5. Comparative results of porosity for samples with different P/L ratios.
longer polymer chains in cross-linked samples. When the polymer chains cross-linked, the polymer chains lost the ability of free movement. Compressive strength values of injectable CaP reported in the literature range from a few MPa [33] up to 83 MPa [34] and these values are based on the method of testing and CaP composition. Cross-linking has been shown to enhance the mechanical properties of the samples. The Young's modulus of dried samples show higher values for both cross-linked and uncross-linked samples compared to the wet samples. However, the Young's modulus ranges of the samples still do not reach the levels of actual human bone [35]. Better fabrication techniques could make better use of the materials to make a sample that could have a potentially higher Young's modulus. In addition, increasing the cross-linking time and concentration can be beneficial. On the other hand, changing in biological properties and biocompatibility should be monitored and considered. Higher injectability could lead to less invasive procedures when uncross-linked CaP–CS cements are used in a clinical setting. Due to the specific procedure that was followed, we could not do the injection test for cross-linked samples and future investigations need to be done. Based on the presented results, P/L ratio has great influence on injection force. The injection test was conducted for P/L = 1.5 and 1.7 and the plastic syringe collapsed during the experiments for samples with P/L = 2. However, P/L = 2.0 could have possibly succeeded with a stronger designed syringe. Nonetheless, the large amount of force required for even P/L = 1.7 indicates a need for a method to generate a large magnitude of force quickly and consistently. To improve the injectability of CaP–CS, one method is reducing viscosity by adding
Fig. 6. Comparative results of a surface hardness test for samples with different P/L ratios. * shows significant difference between the cross-linked and uncross-linked samples.
suitable materials such as biocompatible and biodegradable oils [23]. The porosity results show mixed outcomes of cross-linking CS. The increased porosity of lower P/L ratios is most likely the result of having more space after the fluid has dried out. Maximizing porosity for these scaffolds could be useful as they allow more surface area for cells such as osteocytes to grow into, maximizing the biocompatibility of chitosan in the body. It has been shown that, adding CS does not have a significant effect on CaP–CS porosity [13]. In addition to that, we have shown that cross-linking does not have a meaningful effect on the porosity of the samples in different selected P/L ratios. Another interesting effect of cross-linking on mechanical properties of CaP–CS bone scaffolds is the enhancement of surface hardness. The average Vicker's microhardness of 100 μm-thick sections from a bone sample at a 25 g applied load was reported to be 50 kg/mm2 (490 MPa) [36]. In cross-linked samples, a hardness value close to bone tissue was achieved. Furthermore, the hardness is adjustable by changing P/L ratio and cross-linking duration time. The expected trends were followed where increased P/L leads to increased hardness and CS cross-linking leads to increased hardness as well as. 5. Conclusion In summary, this investigation demonstrated that the incorporation of CS into CaP bone cements caused a significant enhancement of the mechanical properties of the biocomposite. In addition to excellent washout resistance, both the compressive modulus and yield strength of the scaffolds were greatly improved, and a reinforced composite cement was obtained. Although cross-linking is not beneficial for the CaP–CS injectability, it has shown that the cross-linking of CS in CaP matrix can greatly improve hardness and Young's modulus especially in wet condition. In other words, with better and standardized fabrication techniques, the potential for clinical use of a cross-linking agent could lead to improved bone substitute and tissue engineering applications. However, other important factors such as biodegradability and biocompatibility should also be considered carefully. References [1] S.V. Madihally, H.W.T. Matthew, Biomaterials 20 (1999) 1133–1142. [2] F. van de Watering, J. van den Beucken, R.F. Lanao, J. Wolke, J. Jansen, Biodegradation of Calcium Phosphate Cement Composites, Degradation of Implant Materials, Springer, 2012. 139–172. [3] A. Di Martino, M. Sittinger, M.V. Risbud, Biomaterials 26 (2005) 5983–5990. [4] J.K. Francis Suh, H.W.T. Matthew, Biomaterials 21 (2000) 2589–2598. [5] J.L. Moreau, H.H.K. Xu, Biomaterials 30 (2009) 2675–2682. [6] P.J. VandeVord, H.W.T. Matthew, S.P. DeSilva, L. Mayton, B. Wu, P.H. Wooley, J. Biomed. Mater. Res. 59 (2002) 585–590. [7] L. Leroux, Z. Hatim, M. Freche, J. Lacout, Bone 25 (1999) 31S–34S. [8] L. Chow, W. Brown, J. Dent. Res. 63 (1984) 868–873. [9] S. Takagi, L.C. Chow, S. Hirayama, F.C. Eichmiller, Dent. Mater. 19 (2003) 797–804. [10] S. Takagi, L.C. Chow, S. Hirayama, A. Sugawara, J. Biomed. Mater. Res. B Appl. Biomater. 67B (2003) 689–696. [11] H.H.K. Xu, L.E. Carey, C.G. Simon Jr., S. Takagi, L.C. Chow, Dent. Mater. 23 (2007) 433–441. [12] H.H.K. Xu, C.G. Simon Jr., Biomaterials 26 (2005) 1337–1348. [13] Y. Zhang, M. Zhang, J. Biomed. Mater. Res. 55 (2001) 304–312. [14] Z. Babaei, M. Jahanshahi, S.M. Rabiee, Mater. Sci. Eng. C 33 (2012) 370–375. [15] F. Chen, Z.-C. Wang, C.-J. Lin, Mater. Lett. 57 (2002) 858–861. [16] Y. Zhang, J.R. Venugopal, A. El-Turki, S. Ramakrishna, B. Su, C.T. Lim, Biomaterials 29 (2008) 4314–4322. [17] W.W. Thein-Han, R.D.K. Misra, Acta Biomater. 5 (2009) 1182–1197. [18] B.M. Chesnutt, A.M. Viano, Y. Yuan, Y. Yang, T. Guda, M.R. Appleford, J.L. Ong, W.O. Haggard, J.D. Bumgardner, J. Biomed. Mater. Res. A 88A (2009) 491–502. [19] M.-S. Chiou, P.-Y. Ho, H.-Y. Li, Dyes Pigments 60 (2004) 69–84. [20] V. Thai, B.-T. Lee, J. Mater. Sci. Mater. Med. 21 (2010) 1867–1874. [21] M.P. Ginebra, M. Espanol, E.B. Montufar, R.A. Perez, G. Mestres, Acta Biomater. 6 (2010) 2863–2873. [22] H. Liu, C. Gao, Polym. Adv. Technol. 20 (2009) 613–619. [23] S. Heinemann, S. Rössler, M. Lemm, M. Ruhnow, B. Nies, Acta Biomater. 9 (2012) 6199–6207. [24] A. Aryaei, A.H. Jayatissa, A.C. Jayasuriya, J. Mech. Behav. Biomed. Mater. 5 (2012) 82–89. [25] M. Bohner, G. Baroud, Biomaterials 26 (2005) 1553–1563.
A. Aryaei et al. / Materials Science and Engineering C 54 (2015) 14–19 [26] E.F. Burguera, H.H. Xu, L. Sun, J. Biomed. Mater. Res. B Appl. Biomater. 84 (2008) 493–502. [27] I. Khairoun, M. Boltong, F. Driessens, J. Planell, J. Biomed. Mater. Res. 38 (1997) 356–360. [28] H.L.R. Alves, L.A. dos Santos, C.P. Bergmann, J. Mater. Sci. Mater. Med. 19 (2008) 2241–2246. [29] A. Cherng, S. Takagi, L. Chow, J. Biomed. Mater. Res. 35 (1997) 273–277. [30] A.C. Jayasuriya, A. Bhat, J. Biomed. Mater. Res. 93 (2010) 1280–1288. [31] A.C. Jayasuriya, A. Aryaei, A.H. Jayatissa, Mater. Sci. Eng. C 33 (2013) 3688–3696.
19
[32] A. Aryaei, A.H. Jayatissa, A.C. Jayasuriya, J. Biomed. Mater. Res. A 102 (2014) 2704–2712. [33] I. Rajzer, O. Castano, E. Engel, J. Planell, J. Mater. Sci. Mater. Med. 21 (2010) 2049–2056. [34] M. Bohner, Eur. Spine J. 10 (2001) S114–S121. [35] R.B. Ashman, J.Y. Rho, J. Biomech. 21 (1988) 177–181. [36] G. Boivin, Y. Bala, A. Doublier, D. Farlay, L. Ste-Marie, P. Meunier, P. Delmas, Bone 43 (2008) 532–538.
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Cross-linked chitosan improves the mechanical properties of calcium phosphate–chitosan cement Ashkan Aryaei a, Jason Liu b, Ahalapitiya.H. Jayatissa a, A. Champa Jayasuriya c,⁎ a b c
Department of Mechanical Engineering, University of Toledo, Toledo, OH 43606, USA School of Medicine, University of Toledo, OH 43614, USA Department of Orthopaedic Surgery, University of Toledo, Toledo, OH 43614, USA
a r t i c l e
i n f o
Article history: Received 9 April 2014 Received in revised form 26 February 2015 Accepted 21 April 2015 Available online 22 April 2015 Keywords: Calcium phosphate cement Chitosan Cross-linking Mechanical properties Porosity
a b s t r a c t Calcium phosphate (CaP) cements are highly applicable and valuable materials for filling bone defects by minimally invasive procedures. The chitosan (CS) biopolymer is also considered as one of the promising biomaterial candidates in bone tissue engineering. In the present study, some key features of CaP–CS were significantly improved by developing a novel CaP–CS composite. For this purpose, CS was the first cross-linked with tripolyphosphate (TPP) and then mixed with CaP matrix. A group of CaP–CS samples without cross-linking was also prepared. Samples were fabricated and tested based on the known standards. Additionally, the effect of different powder (P) to liquid (L) ratios was also investigated. Both cross-linked and uncross-linked CaP–CS samples showed excellent washout resistance. The most significant effects were observed on Young's modulus and compressive strength in wet condition as well as surface hardness. In dry conditions, the Young's modulus of cross-linked samples was slightly improved. Based on the presented results, cross-linking does not have a significant effect on porosity. As expected, by increasing the P/L ratio of a sample, ductility and injectability were decreased. However, in the most cases, mechanical properties were enhanced. The results have shown that cross-linking can improve the mechanical properties of CaP–CS and hence it can be used for bone tissue engineering applications. © 2015 Elsevier B.V. All rights reserved.
1. Introduction In tissue engineering, chitosan (CS), derived from chitin has been explored on several fronts as an organic substance to complex with calcium phosphate (CaP). Part of what makes chitosan a desirable to use substance is its biodegradability, which allows the body to break it down after healing [1,2], and biocompatibility, which includes minimal foreign body reaction while allowing human cells to grow onto it [3–6]. With such biocompatibility, CS is an ideal substance to complex into a scaffold with other additives [7]. One of the promising candidates for bone paste biomaterials is CaP, the main component of the inorganic portions of bone. CaP cements have gained clinical acceptance as valuable bone replacing biomaterials for almost three decades [8]. These scaffolds would form the basis of a moldable cement intended for bone regeneration applications while also allowing cells such as osteoblasts to grow normally onto them. Porosity, set time and mechanical properties such as, flexural strength, tensile strength, and hardness have been explored in vitro and in vivo. For the sake of clinical
⁎ Corresponding author at: University of Toledo, Department of Orthopaedic Surgery, 3065 Arlington Avenue, Dowling Hall # 2447, Toledo, OH 43614-5807, USA. E-mail address: [email protected] (A. Champa Jayasuriya).
http://dx.doi.org/10.1016/j.msec.2015.04.024 0928-4931/© 2015 Elsevier B.V. All rights reserved.
applications, traits such as washout resistance and injectability have also been heavily explored [9–13]. These properties have a significant effect on the feasibility of injectable cements in bone tissue engineering. CaP cements have drawbacks in terms of graft migration, brittleness, fatigue fracture, and handling difficulties etc. These properties need to be improved for clinical applications. For instance, washout resistance holds significance due to the intended environment of the CaP–CS complexes, which generally involves being inside the human body. The other important property is injectability, which involves the force required to move the CaP–CS cement through a syringe. Injectable CaP–CS cement can be administered to bone defects using a minimally invasive technique. CaP–CS scaffolds have been created with a variety of different methods to further explore possible methods for enhancements of their mechanical properties. The more recent methods have involved using nanoparticles or nanofibers [14–17]. Other methods involved precipitating beads of CaP–CS complexes out of a solution and lyophilization freeze drying [18]. Also the fabrication process has been heavily varied by the addition of different additives including organic acids, epoxides, and celluloses [19–21]. The addition of the additives has enhanced mechanical properties, injectability, and washout. On a chemical level, some of these additives have even induced crosslinking between chitosan.
A. Aryaei et al. / Materials Science and Engineering C 54 (2015) 14–19
Nonetheless, cross-linked CS containing CaP–CS scaffolds have not been fully explored. CS contains abundant amino and hydroxyl groups which enable it to form particles via both physical and chemical crosslinking. Ionic cross-linking of CS is a typical non-covalent interaction, which can be realized by association with negatively charged multivalent ions such as tripolyphosphate (TPP) [22]. This paper seeks to focus on the effects of cross-linking of CS in the CaP–CS scaffold and explore its effects on Young's modulus in both dry and wet conditions, porosity, and hardness. Moreover, different powder to liquid ratios (P/L) were examined. Also taken into account were washout resistance and injectability. 2. Materials and methods 2.1. Materials All components of CaP cement were purchased from Fisher Scientific. CS (85%, deacetylated), TPP and other chemical materials were purchased from Sigma-Aldrich. Statistical analysis was done using oneway ANOVA and p b 0.05 was considered as statistically significant. For each group, three samples were tested. 2.2. Scaffold fabrication The CaP blend was created with the intention of mimicking the inorganic component of bone. The blend comprised of 55% alpha tricalcium phosphate (α-TCP), 45% dicalcium phosphate anhydrous (DCPA), and 15% calcium carbonate (CaCO3) similar to that which was previously reported [23]. The CS solution was mixed at 2 wt.% with acetic acid or a ratio of 200 mg per 10 mL of acetic acid. The solution was stirred with a magnetic stirrer for 20 min and then set for 20 min allowing the air bubbles to disappear. A CaP powder blend was mixed with the CS solution at powder to liquid mass ratios (P/L) of 1.5, 1.7, and 2.0. The mixing was done manually until a consistent paste-like substance was formed. The paste-like substance was then put into cylindrical molds with dimensions of approximately d = ~ 6 mm and l = ~12 mm based on the ASTM-C39-05 standard. The molds were placed between two glass slides at 100% relative humidity and stored at 37 °C for 4 h. After 4 h, the samples were removed from the molds and placed into a saliva-like solution (SLS) for 20 h at 37 °C. SLS comprised 1.2 mM CaCl2, 0.72 mM KH2PO4, 30 mM KCl, and 50 mM HEPES buffer (4-(2hydroxyethyl)-1-piperazineethanesulfonic acid). This buffer was previously used to mimic plasma in teeth environment [9]. However, phosphate buffered saline (PBS) is another candidate to replace SLS to accurately mimic bone environment. After 20 h, the samples were removed from the SLS and they were ready for testing. The purpose of fabricating CaP–CS scaffolds using the molds is to measure the mechanical properties of CaP–CS cement. 2.3. Cross-linked scaffolds To create cross-linked samples, instead of placing the samples in the molding between glass slides at 100% humidity at 37 °C for 4 h, they were only placed in SLS for 2 h. After 2 h, the samples were moved into 80% (wt) TPP solution for 2 h at 37 °C to induce the chitosan within the scaffolds to cross-link at a molecular level. Samples had a high contact area to the TPP solution and based on the calculated porosity, it is assumed that the TPP solution penetrates into the entire sample. This time is fairly enough to form cross-linking between CS molecules [24]. Similar to uncross-linked scaffolds, the CaP–CS cross-linked scaffolds were wetted by immersing in the SLS solution for 20 h. 2.4. Washout resistance Ideally, a washout resistance test should be done after mixing and preparing paste-like CaP–CS but the test was done after 4 h which was
15
specified for cross-linking CS using TPP. Washout resistance was indicated by the sample maintaining integrity and not being dissolved by the SLS. Samples were carefully moved to a beaker with 20 mL of SLS and kept at room temperature for about 1 h. Then the washout properties were investigated. 2.5. Compressive strength and injectability testing Using the ADMET 2611 mechanical testing machine with an MTestQuatro controller, compressive strength and injectability tests were performed. For the compression test, the crosshead speed was set at 1 mm/min. This speed is suitable for CaP cements reported before [5]. Compressive moduli were tested for both wet and dry samples. Wet samples were the samples that were immediately taken out of the SLS. The dry samples were taken out of the SLS and incubated in a dry environment for 24 h to allow the fluid (mainly SLS) inside to evaporate out. This step was done for both cross-linked and uncross-linked groups. By attaching an appropriate custom made grip, an injectability test was also performed. A syringe was fitted to the machine and filled to about 5 mL of the CaP–CS paste. The force required for expulsion from the syringe was then tested. Since this test was done immediately after making CaP–CS paste, it was only done with uncross-linked samples. Injection speed was set at 6 mm/min. CaP cement pastes are usually considered as non-Newtonian fluid. However, the Hagen– Poiseuille relationship can be used to determine the viscosity [25]. 2.6. Porosity For porosity, samples were initially fully dry. They were then massed and had their exact dimensions measured. Each sample was then immersed into 25 mL of deionized H2O. After 2 h of immersion, the samples were removed and massed again. The mass difference was used to calculate the volume of H2O absorbed by each sample at 24 °C. The amount of H2O absorbed was placed in a ratio with the volume of absorbed the CaP–CS scaffold itself (VVwater ) to derive a dimensionless value Ch‐CaP scaffold
that indicated degree of porosity [13]. 2.7. Hardness For hardness test, dried samples were brought before a CM400AT Clark Microhardness tester. To properly prepare the samples, each was sandpapered at the round ends to create a flatter surface. The round ends were also marked with a marker to add contrast so as to better view with a microscope. The machine would generate small indentations at various forces. The applied force was 200 g force (gf). The indentations' dimensions, both horizontal and vertical diameter, would then be measured to generate a hardness value. Final hardness value can be calculated as follows: HV ¼
1:854F : d2
In this equation, HV is the value of Vicker's hardness, F is the adjusted load and d is the average of square's diagonals. 3. Results 3.1. Washout resistance After 4 h, almost all samples were easily removed from the mold and upon placement into the SLS solution, none of the CaP–CS scaffolds washed out. They maintained structural integrity and did not dissolve into the solution throughout the incubation time as shown in Fig. 1. Particularly, this figure shows the integrity of the samples with P/L = 1.7. For other samples with different P/L ratios, the same behavior was
16
A. Aryaei et al. / Materials Science and Engineering C 54 (2015) 14–19
Fig. 1. Washout resistance of samples after 1 h submerging in SLS.
observed. However, washout resistance for just CaP was low and the samples were crashed and disintegrated in SLS after a few minutes. This observation shows that having chitosan polymer in CaP enhances the integration of samples. 3.2. Compressive strength An increased P/L typically indicated stronger compressibility [26]. Cross-linking using the TPP solution enhanced mechanical properties, much more noticeably in the wet state of the CaP–CS scaffolds. Crosslinking CaP–CS paste using the TPP solution significantly increased the strength and Young's modulus. Fig. 2 shows the stress–displacement curve obtained from a compression test for samples in wet and dry conditions. As expected, by increasing the P/L ratio, the Young's modulus and compressive strength were increased [27]. The maximum compressive strengths for cross-linked CaP–CS samples in wet condition were 0.57, 0.73 and 1.28 MPa for P/L = 1.5, 1.7 and 2, respectively. Similarly, the values for uncross-linked CaP–CS samples in wet condition were 0.03, 0.06 and 0.27 MPa for P/L = 1.5, 1.7 and 2, respectively. It is worth noticing that a significant difference was observed between
cross-linked and uncross-linked samples. Adding a cross-linked polymer such as CS can drastically increase the Young's modulus and compressive strength. For the dry state, as shown in Fig. 3, cross-linking showed mixed results with regards to hardness. The Young's modulus of the CaP–CS scaffolds made with the above method typically were in the 106 Pa range. Fig. 3 shows the Young's modulus for samples in dry condition. Samples were completely crashed at the end of the test due to the low ductility. Similar to the wet condition, by increasing the P/L ratio, Young's modulus and compressive strength increased. For example, Young's modulus were 39.53, 45.96 and 76.31 MPa for P/L = 1.5, 1.7 and 2 in cross-linked samples, respectively. Overall, although the difference is not significant, similar to the wet condition, cross-linking enhanced the Young's modulus. 3.3. Injectability For injectability, only the uncross-linked samples were tested as shown in Fig. 4. This was because of the procedure that was followed to make cross-linked samples as explained above. At P/L = 2.0, the test failed. The syringe broke before any amount of the paste could be
Fig. 2. Stress–displacement curve for (A) cross-linked samples, and (B) uncross-linked samples. (C) Comparative results of the compression test in wet condition. * shows significant difference between the cross-linked and uncross-linked samples.
A. Aryaei et al. / Materials Science and Engineering C 54 (2015) 14–19
17
Fig. 3. Stress–displacement curve for (A) cross-linked samples, and (B) uncross-linked samples. (C) Comparative results of the compression test in dry condition.
expelled from the syringe due to the high viscosity of the fabricated paste. Generally, this followed the trend of increased P/L leading to decreased injectability [28], which would correspond to a higher injection force. For samples which we could do the injection test, the injected paste was coherent without any disintegration (Fig. 4). In addition, it showed great washout resistance after the injection. 3.4. Porosity The results of the porosity are shown in Fig. 5. Increased P/L ratios decreased the porosity while CS cross-linking seemed to have made a marginal effect of the porosity based on the results. Porosity variation between all samples was less than 10%. The mass of H2O absorbed was converted to a volume in mL based on its density at room temperature (24 °C) of 0.997 g/mL. These were the values used to generate the dimensionless porosity value. 3.5. Hardness Increased P/L ratios typically increase the hardness. CS cross-linking also significantly increased the hardness in comparison with uncrosslinked samples. For all P/L ratios, after cross-linking the CaP–CS sample, the hardness was at least doubled (Fig. 6). The surface hardness for uncross-linked CaP–CS samples were 11, 17.67 and 21 kgf/mm2 for P/L = 1.5, 1.7 and 2, respectively. However, after cross-linking, the
surface hardness changed to 33.1, 34.2 and 49.9 kgf/mm2 for P/L = 1.5, 1.7 and 2, respectively. 4. Discussion Both CS and CaP are common biomaterials in bone tissue engineering. Different papers have reported an enhancement in the mechanical properties of the mixture of CaP and CS composite biomaterials [12,13,29]. In addition, it has been shown that cross-linking CS has higher elastic modulus and hardness [24]. Our group previously showed that cross-linked CS with certain concentrations of TPP is not cytotoxic for cells [30] and CS mixed with other forms of materials such as carbon nanotubes and Zinc oxide nanoparticles is cytocompatible with osteoblasts [31,32]. The goal of this investigation was to explore the effect of cross-linking on mechanical properties of CaP–CS composite biomaterial. In the present study, a new method was applied to cross-link CS in CaP–CS matrix. Samples were made based on available standards. Fabricated samples, showed excellent cohesion (i.e., washout resistance), which was achieved by composition of the CS in CaP. We have shown that the Young's modulus of cross-linked CaP–CS samples was significantly higher than that of uncross-linked samples. The different components of CaP–CS also play an important role on the mechanical properties of final CaP–CS cement. As it was observed from Fig. 2, the ductility of samples was slightly decreased. This can be interpreted as
Fig. 4. (A) Injection test results for uncross-linked samples. (B) Washout resistance after injection.
18
A. Aryaei et al. / Materials Science and Engineering C 54 (2015) 14–19
Fig. 5. Comparative results of porosity for samples with different P/L ratios.
longer polymer chains in cross-linked samples. When the polymer chains cross-linked, the polymer chains lost the ability of free movement. Compressive strength values of injectable CaP reported in the literature range from a few MPa [33] up to 83 MPa [34] and these values are based on the method of testing and CaP composition. Cross-linking has been shown to enhance the mechanical properties of the samples. The Young's modulus of dried samples show higher values for both cross-linked and uncross-linked samples compared to the wet samples. However, the Young's modulus ranges of the samples still do not reach the levels of actual human bone [35]. Better fabrication techniques could make better use of the materials to make a sample that could have a potentially higher Young's modulus. In addition, increasing the cross-linking time and concentration can be beneficial. On the other hand, changing in biological properties and biocompatibility should be monitored and considered. Higher injectability could lead to less invasive procedures when uncross-linked CaP–CS cements are used in a clinical setting. Due to the specific procedure that was followed, we could not do the injection test for cross-linked samples and future investigations need to be done. Based on the presented results, P/L ratio has great influence on injection force. The injection test was conducted for P/L = 1.5 and 1.7 and the plastic syringe collapsed during the experiments for samples with P/L = 2. However, P/L = 2.0 could have possibly succeeded with a stronger designed syringe. Nonetheless, the large amount of force required for even P/L = 1.7 indicates a need for a method to generate a large magnitude of force quickly and consistently. To improve the injectability of CaP–CS, one method is reducing viscosity by adding
Fig. 6. Comparative results of a surface hardness test for samples with different P/L ratios. * shows significant difference between the cross-linked and uncross-linked samples.
suitable materials such as biocompatible and biodegradable oils [23]. The porosity results show mixed outcomes of cross-linking CS. The increased porosity of lower P/L ratios is most likely the result of having more space after the fluid has dried out. Maximizing porosity for these scaffolds could be useful as they allow more surface area for cells such as osteocytes to grow into, maximizing the biocompatibility of chitosan in the body. It has been shown that, adding CS does not have a significant effect on CaP–CS porosity [13]. In addition to that, we have shown that cross-linking does not have a meaningful effect on the porosity of the samples in different selected P/L ratios. Another interesting effect of cross-linking on mechanical properties of CaP–CS bone scaffolds is the enhancement of surface hardness. The average Vicker's microhardness of 100 μm-thick sections from a bone sample at a 25 g applied load was reported to be 50 kg/mm2 (490 MPa) [36]. In cross-linked samples, a hardness value close to bone tissue was achieved. Furthermore, the hardness is adjustable by changing P/L ratio and cross-linking duration time. The expected trends were followed where increased P/L leads to increased hardness and CS cross-linking leads to increased hardness as well as. 5. Conclusion In summary, this investigation demonstrated that the incorporation of CS into CaP bone cements caused a significant enhancement of the mechanical properties of the biocomposite. In addition to excellent washout resistance, both the compressive modulus and yield strength of the scaffolds were greatly improved, and a reinforced composite cement was obtained. Although cross-linking is not beneficial for the CaP–CS injectability, it has shown that the cross-linking of CS in CaP matrix can greatly improve hardness and Young's modulus especially in wet condition. In other words, with better and standardized fabrication techniques, the potential for clinical use of a cross-linking agent could lead to improved bone substitute and tissue engineering applications. However, other important factors such as biodegradability and biocompatibility should also be considered carefully. References [1] S.V. Madihally, H.W.T. Matthew, Biomaterials 20 (1999) 1133–1142. [2] F. van de Watering, J. van den Beucken, R.F. Lanao, J. Wolke, J. Jansen, Biodegradation of Calcium Phosphate Cement Composites, Degradation of Implant Materials, Springer, 2012. 139–172. [3] A. Di Martino, M. Sittinger, M.V. Risbud, Biomaterials 26 (2005) 5983–5990. [4] J.K. Francis Suh, H.W.T. Matthew, Biomaterials 21 (2000) 2589–2598. [5] J.L. Moreau, H.H.K. Xu, Biomaterials 30 (2009) 2675–2682. [6] P.J. VandeVord, H.W.T. Matthew, S.P. DeSilva, L. Mayton, B. Wu, P.H. Wooley, J. Biomed. Mater. Res. 59 (2002) 585–590. [7] L. Leroux, Z. Hatim, M. Freche, J. Lacout, Bone 25 (1999) 31S–34S. [8] L. Chow, W. Brown, J. Dent. Res. 63 (1984) 868–873. [9] S. Takagi, L.C. Chow, S. Hirayama, F.C. Eichmiller, Dent. Mater. 19 (2003) 797–804. [10] S. Takagi, L.C. Chow, S. Hirayama, A. Sugawara, J. Biomed. Mater. Res. B Appl. Biomater. 67B (2003) 689–696. [11] H.H.K. Xu, L.E. Carey, C.G. Simon Jr., S. Takagi, L.C. Chow, Dent. Mater. 23 (2007) 433–441. [12] H.H.K. Xu, C.G. Simon Jr., Biomaterials 26 (2005) 1337–1348. [13] Y. Zhang, M. Zhang, J. Biomed. Mater. Res. 55 (2001) 304–312. [14] Z. Babaei, M. Jahanshahi, S.M. Rabiee, Mater. Sci. Eng. C 33 (2012) 370–375. [15] F. Chen, Z.-C. Wang, C.-J. Lin, Mater. Lett. 57 (2002) 858–861. [16] Y. Zhang, J.R. Venugopal, A. El-Turki, S. Ramakrishna, B. Su, C.T. Lim, Biomaterials 29 (2008) 4314–4322. [17] W.W. Thein-Han, R.D.K. Misra, Acta Biomater. 5 (2009) 1182–1197. [18] B.M. Chesnutt, A.M. Viano, Y. Yuan, Y. Yang, T. Guda, M.R. Appleford, J.L. Ong, W.O. Haggard, J.D. Bumgardner, J. Biomed. Mater. Res. A 88A (2009) 491–502. [19] M.-S. Chiou, P.-Y. Ho, H.-Y. Li, Dyes Pigments 60 (2004) 69–84. [20] V. Thai, B.-T. Lee, J. Mater. Sci. Mater. Med. 21 (2010) 1867–1874. [21] M.P. Ginebra, M. Espanol, E.B. Montufar, R.A. Perez, G. Mestres, Acta Biomater. 6 (2010) 2863–2873. [22] H. Liu, C. Gao, Polym. Adv. Technol. 20 (2009) 613–619. [23] S. Heinemann, S. Rössler, M. Lemm, M. Ruhnow, B. Nies, Acta Biomater. 9 (2012) 6199–6207. [24] A. Aryaei, A.H. Jayatissa, A.C. Jayasuriya, J. Mech. Behav. Biomed. Mater. 5 (2012) 82–89. [25] M. Bohner, G. Baroud, Biomaterials 26 (2005) 1553–1563.
A. Aryaei et al. / Materials Science and Engineering C 54 (2015) 14–19 [26] E.F. Burguera, H.H. Xu, L. Sun, J. Biomed. Mater. Res. B Appl. Biomater. 84 (2008) 493–502. [27] I. Khairoun, M. Boltong, F. Driessens, J. Planell, J. Biomed. Mater. Res. 38 (1997) 356–360. [28] H.L.R. Alves, L.A. dos Santos, C.P. Bergmann, J. Mater. Sci. Mater. Med. 19 (2008) 2241–2246. [29] A. Cherng, S. Takagi, L. Chow, J. Biomed. Mater. Res. 35 (1997) 273–277. [30] A.C. Jayasuriya, A. Bhat, J. Biomed. Mater. Res. 93 (2010) 1280–1288. [31] A.C. Jayasuriya, A. Aryaei, A.H. Jayatissa, Mater. Sci. Eng. C 33 (2013) 3688–3696.
19
[32] A. Aryaei, A.H. Jayatissa, A.C. Jayasuriya, J. Biomed. Mater. Res. A 102 (2014) 2704–2712. [33] I. Rajzer, O. Castano, E. Engel, J. Planell, J. Mater. Sci. Mater. Med. 21 (2010) 2049–2056. [34] M. Bohner, Eur. Spine J. 10 (2001) S114–S121. [35] R.B. Ashman, J.Y. Rho, J. Biomech. 21 (1988) 177–181. [36] G. Boivin, Y. Bala, A. Doublier, D. Farlay, L. Ste-Marie, P. Meunier, P. Delmas, Bone 43 (2008) 532–538.
