Purification and biocompatibility of fermented hyaluronic acid for its applications to biomaterials
- Sungchul Choi1, 2,
- Woncheol Choi2,
- Sekweon Kim2,
- Su-Yeon Lee3, 4,
- Insup Noh†3, 4Email author and
- Chan-Wha Kim†1Email author
© Choi et al.; licensee BioMed Central Ltd. 2014
Received: 15 January 2014
Accepted: 5 June 2014
Published: 13 June 2014
Hyaluronic acid (HA) is of importance due to its diverse applications in pharmaceuticals and medical devices such as dermal filler, adhesion barriers, carrier for cells and bioactive molecules as well as scaffold biomaterials for tissue engineering. Evaluations of purification and biocompatibility of HA are required for its applications to biomaterials.
After synthesizing HA by fermentation of streptococcus zooepidemicus for 25 hr, extensively purification of the fermented broth was performed to remove impurities using a filtration process for insoluble components and cells, and diverse adsorbents for soluble impurities. Its in vitro biocompatibility has been evaluated by measurement of cell counting and assay of cell live and dead. 60% yield of white HA powder was obtained, having 15–17 dL/g intrinsic viscosity with a molecular weight of approximately 1,000 kDa. While low molecular weight impurities and insoluble impurities were successfully removed using a ultrafiltration membrane with 50 KDa molecular weight cut, endotoxins, high molecular weight proteins and nucleic acids were removed from the broth by employing adsorbents such as alumina and activated carbons. Alumina showed the best results for the removal of endotoxins, all of the activated carbons were very effective in the removal of high molecular weight proteins and nucleic acids. The purified HA solution showed excellent cell compatibility with no cell damages as observed by both measurement of cell proliferation and observation of cell viability.
We obtained high molecular weight HA with excellent biocompatibility as judged by both measurement of cell proliferation and viability, indicating high possibility of its applications to biomaterials.
KeywordsHyaluronic acid Purification Activated carbon Adsorption Biomaterials
Hyaluronic acid (HA) is a linear anionic, non-sulfated, high molecular weight glycosaminoglycan with alternating D-glucuronic acid and N-acetyl-D-glucosamine. HA is distributed natively as a component of extracellular matrix in connective, epithelial and neutral tissues in body. HA has several important physico-chemical and biological properties including lubricity, visco-elasticity, water retention, biocompatibility, cell proliferation, morphogenesis, inflammation and wound repair as well as specific signal transduction and cellular interactions through cell surface receptors such as CD44, CD54 and CD168 [1–3]. HA can be degraded by both reactive oxygen intermediates and hyaluronases, which are synthesized by macrophages, fibroblasts and endothelial cells [4, 5].
Physical properties of the adsorbents employed in the adsorption process
Particle size (μm; d10, d50, d90)*
Methylene blue adsorption (g/100 g)
4, 15, 40
7, 30, 76
5.5 ~ 8
Norit SX Plus
6, 20, 80
Norit SX 1G
5, 25, 90
Norit C Gran
0.85 mm ~ 1.7 mm***
2 ~ 8
Norit GAC 1240+
0.43 mm ~ 2.0 mm***
Norit C Extra USP
4, 23, 100
Norit A Supra EUR
4, 20, 100
2.5, 6, 10
Fermentation and analysis of the properties of the fermenting broth
Streptococcus zooepidemicus was fermented in a bioreactor (5 L, Marubishi Inc.) at 37°C and 300 rpm for 25 hr with 1.0 vvm aeration to obtain hyaluronic acid, which have been reported well by numerous companies and researchers, including Akdamar HA and Volpi N [11, 12]. The complex media under an aerobic condition that was used as the fermenting broth consisted of yeast extract (15 ~ 25 g/L), potassium phosphate (2.5 g/L), glucose (60 ~ 80 g/L), magnesium sulfate (3 g/L), sodium chloride (5 g/L) and L-glutamate (7 g/L). The residual L-glucose in the fermenting broth was analyzed using 1 ml in a biochemical analyzer (MBI 7200 model, YSI Co., USA) and the optical density of the broth was measured with a Spectrophotometer (UV2450, Simadzu Co., Japan) at an wavelength of 600 nm after a 10 fold dilution. The broth’s viscosity was measured using a rotary viscometer (LV type, Brookfield Co.) and 400 mL of the solution.
Removal of cells
The bacterial strains and insoluble impurities in the highly viscous broth were removed by filtering with a depth filter (KD7, Begerow Co.) that had a 1.5 μm nominal retention rate after mixing with either 1, 3 or 5% diatomaceous earth (v/v). The efficacy of the broth filtration was confirmed by measuring the presence of bacteria with a Spectrophotometer (UV2450, Simadzu Co., Japan) at a wavelength of 600 nm.
Diafiltration was used for the removal of broth components, metabolic products and low molecular weight impurities from 5 L of broth. This process was conducted using a tangential flow filtration method by employing an Ultra Filter (Pellicon2, Millipore, USA) after removal of the bacterial strains and insoluble impurities. The sample was then diluted by a factor of 2 using deionized water. Diafiltration was performed by pumping the solution through the poly(ether sulphone) ultrafilter cassette with a dimension of 0.5 m2 and a molecular weight cut-off of either 30 or 50 kDa (PR sterile 40, Begerow Inc., Germany). After defining the number of the diafiltration purification as the number of alternating processes of both concentration and dilution, both the permeate and conductivity and the concentration of residual lactate were measured using a conductivity meter (Seveneasy; Mettler-Toledo, Swizerland) and the biochemical analyzer (MBI 7200; YSI Inc., USA), respectively. The concentration of residual lactates was measured by feeding 1 mL into the biochemical analyzer and measuring the electrical signals from the residual lactates, which were produced as a metabolic byproduct during the fermenting process. The electrical signals were measured by the biochemical analyzer during induction of lactates degradation by L-lactate oxidases. The volume of the permeate solution filtered from the membrane was measured in a dimension of L/h/m2 to check the buffer change time during diafiltration.
Purification by adsorbents
High molecular weight impurities were removed by adding different adsorbents to the diafiltrated broth. After addition of 2% adsorbent (w/v) into 1,000 mL of the diafiltrated broth, the solution was stirred with a stirrer (RZR2021, Heidolph Inc., Germany) for 10 hr. The adsorbents were then removed using a 0.45 μm filter (PR sterile 40, Begerow Inc., Germany).
Recovery of the hyaluronic acid in powder
Solidification of the HA sample was achieved by adding 3 L of acetone to the adsorption-processed solution. The HA powder was recovered and dried after filtration of the solidified HA with a 100 μm mesh (CISA, Spain). The HA yield was determined to be about 60%, which was measured by comparing the weight of the dry HA with that of the fermented weight.
Verification of hyaluronic acid powder was performed using the analytical methods described in both the European pharmacopoeia and Korean pharmacopoeia, which included measuring the residual levels of endotoxins, proteins, nucleic acids and small particles.
The kinetic chromogenic method (Kinetic-QCL Chromogenic LAL assay, Lonza, USA) was used to measure the amount of residual endotoxin. After syntheses of both standard and test solutions according to the procedures suggested by the vendor, the LAL kinetic chromogenic reagent was added and its optical density was measured at a wavelength of 405 nm using a microplate meter. Two mL standard solutions were obtained by diluting the standards to 0.005, 0.05, 0.5, 5 EU/mL with the LAL reagent water (Lonza Co., USA). The test solution was prepared by diluting the sample solution to the lowest possible level to minimize the effect of inhibition factors. The LAL reagent was prepared by adding up LAL reagent water to the LAL kinetic chromogenic reagent.
Examination of residual high molecular weight proteins in the hyaluronic acid powder was performed according to the protein tests recommended by the Korean pharmacopoeia. The test solution was prepared by dissolving 0.1 g HA powder in 20 mL deionized water. BSA was used as the reference solution and the BSA concentrations used were 1, 3, 5, 7 and 10%. Test and standard solutions were reacted with 2.5 mL cupri-tartaric solution for 10 min. After reacting with 0.5 mL of the phosphomolybdotunstic reagent for 30 min, the solution’s optical density was measured at a wavelength of 750 nm, which was used to calculate the concentrations of residual proteins. The phosphomolybdotunstic reagent was obtained by refluxing for 10 hr in a reflux condenser after the addition of 100 g sodium tunstate and 25 g sodium molybdate, and then 100 mL hydrogen chloride and 50 mL phosphonic acid. To remove excess bromine, the sample was boiled for 15 min after the addition of 150 g lithum sulfate, 50 mL water and bromine. Water was added to a final volume of 1,000 mL and then the sample was filtered. The cupri-tartaric solution was prepared by mixing 50 mL sodium carbonate with 0.1 mL of a solution that was previously prepared by dissolving 1.0 g copper sulfate and 2.0 g sodium tartarate in 100 mL deionized water. The 100 mL sodium carbonate solution was obtained by dissolving 4.0 g sodium carbonate in 0.2 M NaOH.
Nucleic acid analysis
Residual nucleic acid was analyzed in either a liquid or powder state. The amount of nucleic acid in the processing broth was obtained by measuring the optical density of a 1 mL solution at a wavelength of 260 nm using the Spectrophotometer (Simadzu Inc., Japan) and the amount of nucleic acid in the powder was determined by adding the powder to 30 mL 0.9% sodium chloride solution and then measuring absorbance at 260 nm.
Measurement of micro-particles in the processing broth
Micro-particles in the processing broth after treatment with diverse activated carbons were measured using a liquid particle counter (LS-200, Particle Measuring Systems Inc., USA). In this analysis, 20 mL test solution either with or without HA was added to the liquid particle counter as well as solution containing 2% of the activated carbons were fed in either granule or powder form.
Molecular weight of HA
The molecular weight of the dry HA was measured using a Multi Angle Laser Light Scattering (MALS) detector (DAWN, Wyatt Technology Co., US) and high performance liquid chromatography (HPLC) (1200 series, Agilent, Germany) (column: TSK G6000 PWXL, Tosoh, Japan). A 0.02 ~ 0.03% HA solution was filtered through a 0.45 μm filter (Acrovisk, PALL, USA) and then injected into the HPLC system with a glass syringe. The molecular weights of the HA compounds were measured using a multi-angle light scattering (MALS) detector.
40 mL 0.005%, 0.010%, 0.015% and 0.020% HA solutions were synthesized by dissolving the dry HA powder in buffer solution (0.01 M) with 0.15 M sodium chloride at pH 7. Relative viscosity was obtained by measuring the flow time of the HA solution with a capillary viscometer (Type 501 01, SI Analytics GmbH, Germany) and linear least-square regression of the Martin equation.
In vitroevaluation of HA samples
In vitro cell culture
MC3T3 cells, osteoblast precursor cell line, were in vitro cultured in α-MEM media (Sigma Aldrich, USA) containing 10% fetal bovine serum and penicillin-streptomycin (100 IU/mL) and gentamycin (2 L/mL) in an incubator with 5% CO2 at 37°C. Sterilized HA solution (Hyalrheuma Inj.; Hanmi Pharmaceutical Co., Pyeongtaek, Korea) was loaded on a 24 well polystyrene tissue culture plate (Nunc Co., USA). MC3T3 cells, 10,000 cells/well, were seeded on the surface of 200 μl HA solution and cultured under static conditions in an incubator with 5% CO2 at 37°C for 7 days.
Cell proliferation assay
Cell adhesion and proliferation on the HA solution was quantitatively measured by counting the number of cells with a cell counting kit-8 (CCK-8; Dojindo: Japan) by a microplate reader (Tecan: Australia). 100 μL solution of CCK-8 was inserted into the 1 mL cell culture medium and then the cell culture plate was incubated in 5% CO2 incubator at 37°C. After 4 hr, 100 μL α-MEM with CCK-8 was aliquoted into a 96 well plate and then the optical its density was measured at the wavelength of 450 nm .
In vitro cell viability by live & dead assay
For observation of cytotoxicity of the HA solution, cell viability on the HA solution was evaluated by Live & Dead assay after in vitro cell culture with MC3T3 cells for 7 days according to our previous report, by staining with fluororescence dye .
Results and discussion
Bacterial fermentation for HA production
Bacterial fermentation of streptococcus zooepidemicus in the L-glucose bioreactor continued until complete consumption of L-glucose in the culture broth. The fermentor was operated at 37°C and 300 rpm for 25 hr with 1.0 vvm aeration, which resulted in the production of 6.0 ~ 6.5 g HA/L. The optical density of the fermented solution ranged from 8.0 to 12.0 and the broth viscosity processed at 35°C and 21 spindles was as approximately 8000 cP. The molecular weight of the HA obtained was determined to be 900 ~ 1,100 kDa.
HA purification of processing broth
Removal of bacteria and insoluble impurities
Removal of strains by filtration conditions
Nominal retention rate (μm)
Earth oil conc. (%)
OD (260 nm)
Flow rates (L/hr)
HA purification with diafiltration method
Next we removed impurities such as culture products and broth components from the bacteria-free broth through filtration. The extent of impurity removal was evaluated by measuring membrane conductivity and flow rates over the membrane pores, and the number of diafiltration cycles.
Measurement of conductivity and flow rates of broth
Removal of impurities by diafiltrations
The medium solutions that had been diafiltrated still contained significant amounts of impurities including colors (67.44 EU/mL) and endotoxins (77.73 μg/mL). Therefore, the filtrated broth was further purified using various adsorbents as described below.
Removal of endotoxin
The percentage of endotoxins and proteins present after the adsorption process
Endotoxin (initial: 67.44 EU/Mℓ)
Proteins (initial 77.73 μg/Mℓ)
Norit C Extra USP
Norit A Supra EUR
Norit SX Plus
Norit SX 1G
Norit C Gran
Norit GAC 1240+
Removal of proteins
High molecular weight proteins, which are metabolic byproducts in this study, were still present in the processing broth. While low molecular weight proteins were removed during the previous diafiltration process using a membrane with a molecular weight cut-off of 50 KDa, high molecular weight proteins needed to be removed using adsorbents. In contrast to endotoxin removal, alumina removed only 87.8% of the proteins, while the activated carbons such as Darco KB-B, Norit CN1, Norit A Supra EUR and Norit C Extra USP removed 96.3 to 97.4% of the proteins. Among the employed absorbents, Norit CN1 removed the most high molecular weight proteins (97.4%) (Table 3), but this increased rate of removal was not significantly different when compared to the other activated carbons such as Darco KB-B, Norit C Extra USP and Norit A Supra EUR. While Darco KB-B and Norit CN1 were acidic, the Norit C Extra USP and Norit A Supra EUR were neutral. In addition, Norit CN1 and Norit C Extra USP had lower methylene blue adsorption (29 and 22 g/100 g, respectively) and Darco KB-B and Norit A Supra EUR had relatively higher values (41 and 42 g/100 g, respectively). We excluded some activated carbon adsorbents from these experiments such as Norit SX Plus and 1G, Norit C Gran and Norit GAC 1240+ for the same reasons described in the endotoxin removal experiments, i.e. observation of colors remained.
Removal of nucleic acid
Removal of nucleic acids in the processing broth was also evaluated by employing various adsorbents. The results measured by the UV spectrometer at the wavelength of 260 nm showed that almost all the residual nucleic acids were removed regardless of the state of the activated carbons (powders vs. granule forms). The pH values of the activated carbons did not significantly affect removal of endotoxins, which ranged from 88.3 to 92.0%. The pH values of the Darco KB-B and Norit CN1 were acidic with a ≤ 15 moisture %, and the pH values of Norit C Extra USP and Norit A Supra EUR were neutral with a ≤ moisture 10%. Up to 99.7% of the residual nucleic acids were removed when those activated carbon adsorbents were used. In contrast, alumina removed only 85% of the initial residual nucleic acids. By comparing the removal of both high molecular weight proteins and nucleic acid by the activated carbons and alumina, a higher surface to mass ratio, BET, seemed to result in a higher removal of proteins and nucleic acids. Since the activated carbons had a relatively high BET, i.e. higher surface to mass ratios, ranging 1200 to 2050 m2/g, and alumina had a BET value of 103 m2/g, alumina seemed to be effective in removal of proteins and nucleic acids.
Removal of used adsorbents
Adsorbent particles and their size distributions, dependent upon the conditions of HA contents and stirring time during purification process by Norit C Gran
Norit C Gran. (%)
HA contents (%)
Particle(ea/mL, 1/100 dilution)
< 2 μm
2 ~ 5 μm
5 ~ 10 μm
10 ~ 15 μm
12 ~ 25 μm
Properties of the precipitated & dry HA polymers
White HA powders were obtained using sequential processing of precipitation of HA with 2 times acetone and then drying in clean bench. The final yield of HA was measured to be about 60% when comparing the weight of dry HA with that of the diafiltered dry weight. Its intrinsic viscosity of the HA samples was measured to be 15–17 dL/g with a molecular weight of approximately 1,000 kDa.
Biocompatibility of HA solution
We obtained white hyaluronic acid by the processes of fermentation and purification of the fermented broth with filtration and by the use of adsorbents. Removal of the fermented bacterial strains and insoluble impurities was achieved by employing both centrifuge and filtration using filters with a diameter of 0.6-1.5 μm and through the addition of 3 ~ 5% earth soils. Low molecular weight impurities were successfully removed by the ultrafilter with a molecular cut-off of 50 KDa after repeated processes. When we increased the recycling numbers, the permeances of the processing broth increased but the membrane conductivity and lactate concentration decreased, which indicated successful removal of impurities in the broth. Further purification with adsorbents removed nearly all the endotoxins, high molecular weight proteins and nucleic acids from the fermented broth. While alumina was the best adsorbent for the removal of endotoxins, activated carbons such as Norit KBB, CN1, C Extra USP, A Supra EUR were very effective in the removal of high molecular weight proteins and nucleic acids. The granular activated carbons, Norit C Gran, were not adequate for the purification of HA from fermenting broth due to the breakdown of the particles and subsequent entanglement with HA. These two effects result in the formation of a significant amount of large particles, which block the filtering membrane. The other activated carbon in granules, Norit GAC 1240+, induced metal corrosions, which is not adequate as an adsorbent. Among the employed activated carbons, the activated carbons in powders such as Norit C Extra USP and Norit A Supra EUR with a higher BET and neutral pH satisfied the criteria required for pharmaceuticals by EU. The results of HA purification using activated carbons indicated that the adsorbent morphology, species and type as well as processing modes such as either with or without stirring were critically important in obtaining high quality HA. The high quality HA could be employed as a biomaterial for pharmaceuticals, tissue engineering and medical devices as well as a carrier for cell therapy. The results of in vitro cell culture on the HA solution obtained by mass scale production demonstrated its excellent cell adhesion and proliferation as well as cell viability. These results indicated that the choices of adequate purification processes were important in obtaining biocompatible HA and the purified HA polymers were excellent candidate, respectively, for their applications to biomaterials.
Availability of supporting data
The data sets supporting the results of this article are included within the article.
Brnauer, Emmett & Teller
Cell counting kit-8
Multi Angle Laser Light Scattering
High performance liquid chromatography.
This work was supported by the Pioneer Research Center Program through the National Research Foundation of Korea funded by the Ministry of Education, Science and Technology (Grant No. 2012–0001052) and supported by a grant of the Korea Health technology R&D Project, Ministry of Health & Welfare, Republic of Korea (A120822).
- Burdick JA, Prestwich GD: Hyaluronic acid hydrogels for biomedical applications. Adv Mater. 2011, 23 (12): 41-56.View ArticleGoogle Scholar
- Choh SY, Cross D, Wang C: Facile synthesis and characterization of disulfide-cross-linked hyaluronic acid hydrogels for protein delivery and cell encapsulation. Biomacromolecules. 2011, 12 (4): 1126-1136.View ArticleGoogle Scholar
- Gojgini S, Tokatlian T, Segura T: Utilizing cell-matrix interactions to modulate gene transfer to stem cells inside hyaluronic acid hydrogels. Mol Pharm. 2011, 8 (5): 1582-1591.View ArticleGoogle Scholar
- Burdick JA, Chung C, Jia X, Randolph MA, Langer R: Controlled degradation and mechanical behavior of photopolymerized hyaluronic acid networks. Biomacromolecules. 2005, 6 (1): 386-391.View ArticleGoogle Scholar
- Johnson P, Ruffell B: CD44 and its role in inflammation and inflammatory diseases. Inflamm Allergy Drug Targets. 2009, 8 (3): 208-220.View ArticleGoogle Scholar
- Kang DY, Kim WS, Heo IS, Park YH, Lee S: Extraction of hyaluronic acid (HA) from rooster comb and characterization using flow field-flow fractionation (FlFFF) coupled with multiangle light scattering (MALS). J Sep Sci. 2010, 33 (22): 3530-3536.View ArticleGoogle Scholar
- Meyer K, Palmer J: The polysaccharide of the vitreous humor. J Biol Chem. 1934, 107: 629-634.Google Scholar
- Kendall FE, Heideberger M, Dawson MH: A serologically inactive polysaccharide elaborated by mucoid strains of group A hemolytic streptococcus. J Biol Chem. 1937, 118: 61-69.Google Scholar
- Tawada A, Masa T, Oonuki Y, Watanabe A, Matsuzaki Y, Asari A: Large-scale preparation, purification, and characterization of hyaluronan oligosaccharides from 4-mers to 52-mers. Glycobioloy. 2002, 12 (7): 421-426.View ArticleGoogle Scholar
- Chong BF, Blank LM, Mclaughlin R, Nielsen LK: Microbial hyaluronic acid production. Appl Microbiol Biotechnol. 2005, 66 (4): 341-351.View ArticleGoogle Scholar
- Akdamar HA, Sariozlu NY, Ozcan AA, Ersoz A, Denizli A, Say R: Separtation and purification of hyaluronic acid by glucuronic acid imprinted microbeads. Mater Sci and Eng C. 2009, 29 (4): 1404-1408.View ArticleGoogle Scholar
- Volpi N, Maccari F: Purification and characterization of hyaluronic acid from the mollusk bivalve mytilus galloprovincialis. Biochimie. 2003, 85 (6): 619-625.View ArticleGoogle Scholar
- Hong SS, Chen J, Zhang JG, Tao YC, Liu LY: Purification and structure identification of hyaluronic acid. Chin Chem Lett. 2004, 15 (7): 811-812.Google Scholar
- Rangaswamy V, Jain D: An efficient process for production and purification of hyaluronic acid from streptococcus equi subsp. Zooepidemicus. Biotechnol Lett. 2008, 30 (3): 493-496.View ArticleGoogle Scholar
- Kablik J, Monheitmd GD, Phd LY, Chang G, Gershkovich J: Comparative physical properties of hyaluronic acid dermal fillers. Dermatol Surg. 2009, 35 (Suppl 1): 302-312.View ArticleGoogle Scholar
- Pasquale MD, Gaspare MD: Effect of hypotonic 0.4% hyaluronic acid drops in dry eye patients: a cross-over study. Cornea. 2008, 27 (10): 1126-1130.View ArticleGoogle Scholar
- d’Ayala GG, Malinconico M, Laurienzo P: Marine derived polysaccharides for biomedical applications: chemical modification approaches. Molecules. 2008, 13 (9): 2069-2106.View ArticleGoogle Scholar
- Hahn SK, Jelacic S, Maier RV, Stayton PS, Hoffman AS: Anti-inflammatory drug delivery from hyaluronic acid hydrogels. J Biomater Sci Polym Ed. 2004, 15 (9): 1111-1119.View ArticleGoogle Scholar
- Awesh KY, Mishra P, Govind PA: An insight on hyaluronic acid in drug targeting and drug delivery. J Drug Target. 2008, 16 (2): 91-107.View ArticleGoogle Scholar
- Luo Y, Kirker KR, Prestwich GD: Cross-linked hyaluronic acid hydrogel films: new biomaterials for drug delivery. J Controlled Release. 2000, 69 (1): 169-184.View ArticleGoogle Scholar
- Leach JB, Bivens KA, Patrick CW, Schmidt CE: Photocrosslinked hyaluronic acid hydrogels: natural, biodegradable tissue engineering scaffolds. Biotechnol Bioen. 2003, 82 (5): 578-589.View ArticleGoogle Scholar
- Kim J, Kim IS, Cho TH, Lee KB, Hwang SJ, Tae G, Noh I, Lee SH, Park Y, Sun K: Bone regeneration using hyaluronic acid-based hydrogel with bone morphogenic protein −2 and human mesenchymal stem cell. Biomaterials. 2007, 28 (10): 1830-1837.View ArticleGoogle Scholar
- Kim J, Park Y, Tae G, Lee KB, Hwang CM, Hwang SJ, Kim IS, Noh I, Sun K: Characterization of low molecular weight hyaluronic acid based hydrogel and differential stem cell responses in the hydrogel microenvironments. J Biomed Mater Res A. 2009, 88A (4): 967-975.View ArticleGoogle Scholar
- Kim AR, Park HS, Kim SS, Noh I: Biological evaluation of cellulose hydrogel with temperature-responsive particles. Biomater Res. 2013, 17 (4): 181-186.Google Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.