Chitosan coatings with enhanced biostability in vivo Marat O. Gallyamov ,1,2 Ivan S. Chaschin,2 Matvey V. Bulat,1 Natalia P. Bakuleva,3 Gennadii A. Badun,4 Maria G. Chernysheva,4 Olga I. Kiselyova,1 Alexei R. Khokhlov1,2 1
Faculty of Physics, Lomonosov Moscow State University, Leninskie gory 1–2, Moscow 119991, Russian Federation Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences, Vavilova 28, Moscow 119991, Russian Federation 3 Bakulev Scientific Center for Cardiovascular Surgery of the Ministry of Health of the Russian Federation, Roublyevskoe Sh. 135, Moscow 121552, Russian Federation 4 Faculty of Chemistry, Lomonosov Moscow State University, Leninskie gory 1–2, Moscow 119991, Russian Federation 2
Received 29 June 2016; revised 29 November 2016; accepted 1 January 2017 Published online 00 Month 2016 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.b.33852 Abstract: In this article, we study the stability of chitosan coatings applied on glutaraldehyde-stabilized bovine pericardium when exposed to biodegradation in vivo in the course of model subcutaneous tests on rats. The coatings were deposited from carbonic acid solutions, that is, H2O saturated with CO2 at high pressure. Histological sections of treated pericardium samples demonstrated that the structure of pericardial connective tissues was not significantly altered by the coating application method. It was revealed that the dynamics of biodegradation depended on the total mass of chitosan applied as well as on the DDA of
chitosan used. As long as the amount of chitosan did not exceed a certain threshold limit, no detectable degradation occurred within the time of the tests (12 weeks for the rat model). For higher chitosan amounts, we detected a 20% reduction of the mass after the in vivo exposition. The presumed mechanism of C 2017 Wiley Periodicals, Inc. J Biomed such behavior is discussed. V Mater Res Part B: Appl Biomater 00B: 000–000, 2017.
Key Words: heart valve bioprosthesis, bovine pericardium, chitosan coating, carbonic acid, biodegradation
How to cite this article: Gallyamov MO, Chaschin IS, Bulat MV, Bakuleva NP, Badun GA, Chernysheva MG, Kiselyova OI, Khokhlov AR. 2017. Chitosan coatings with enhanced biostability in vivo. J Biomed Mater Res Part B 2017:00B:000–000.
INTRODUCTION
Chitosan, a biocompatible and biodegradable polycation, is extensively used in various biomedical applications, particularly to render objects, implanted into human body, biocompatibility, and antimicrobial properties.1 In the field of cardiovascular surgery implants, to date, a biotissue valve prosthesis is the most adequate alternative to a mechanical prosthesis, since its physical properties are most close to human ones and it does not require lifelong anticoagulant therapy. Most of heart valve biological prostheses consist of a metallic or polymer frame covered with xenogenic pericardium tissue. To render the prosthesis mechanical stability and to suppress the immune response from the recipient, the pericardium tissue is pretreated with glutaraldehyde (GA). The most serious drawback of such valve bioprostheses is their failure due to calcification after implantation.2 The reason of the calcification is not completely clear yet. One of the prevailing hypotheses is that remaining free groups of glutaraldehyde act as nucleation sites for calcification in the tissue. It has been demonstrated that the intensity of calcinosis can be essentially reduced by applying chitosan coatings on collagen-containing xenogeneic pericardium, which was proposed first by Chanda.3 It is believed, that the chitosan
coating masks free aldehyde groups through covalent bonding them to amino and hydroxyl groups in chitosan.4 We have shown that by means of dissolving chitosan in pressurized carbonic acid solutions (H2O saturated with CO2 at high pressure) one can modify the pericardial tissue on the surface and in bulk, which leads to alterations in the structure and properties of the collagen matrix. As a result, the rate of calcinosis was significantly reduced.5 Pressurized carbonic acid solutions possess antimicrobial activity at the stage of treatment and subsequent decompression, therefore no further sterilization of the modified material is required. Moreover, absolute biocompatibility of carbonic acid as a solvent for chitosan due to its complete and spontaneous decomposition into water and carbon dioxide after decompression is also a valuable factor justifying this approach.6 Mechanical properties of chitosan-coated pericardium were studied in our earlier work5 and were found to be within the range of values previously described in the literature, which confirms the applicability of this approach for making valve prosthetic material. Still, pressurized carbonic acid is not a conventional solvent for chitosan. In spite of the fact that chitosan was extensively used in different biomedical applications,
Correspondence to: M. O. Gallyamov, e-mail:
[email protected] Contract grant sponsor: Russian Foundation for Basic Research; contract grant numbers: 16-33-60006, 16-03-00702-a
C 2017 WILEY PERIODICALS, INC. V
1
including tissue engineering,1,7 there were no examples of using pressurized carbonic acid as a solvent for chitosan to prepare chitosan-modified tissues for bioprosthetic substitutes before Ref. 5. Macromolecular conformation of chitosan chains in the pressurized carbonic acid solutions is not typical and seems to be not really the optimal one from the viewpoint of good film-forming properties.6 Achievable concentrations are not high and thus the typical amount of chitosan routinely deposited from such solutions is comparatively small, requiring highly sensitive methods for its reliable detection. Therefore, the general applicability of pressurized carbonic acid as a proper/beneficial solvent for chitosan for biomedical applications in general (and particularly for obtaining improved bioprosthesis) still requires experimental verification on some simple model systems, as a starting point. Further, resistance to enzymatic degradation is one of important factors, which should be considered when developing an improved biomaterial for heart valve construction. Because chitosan is known to be subjected to bioresorption in vivo, the question of durability of such chitosan coatings on implants is of major concern. Previous extensive research has shown several pathways of chitosan in vivo biodegradation in mammals.8,9 In mice, chitosan with degree of deacetylation of 50% readily degraded and cleared, thus bioaccumulation can be considered of no concern.10 Enzymatic hydrolysis is catalyzed by lysozyme, chitinase, chitosanase and hexaminodase, while nonspecific hydrolysis pathways are mediated by oxidative agents produced by immune system.8,11 The digestion of chitosans with various degrees of deacetylation (DDA) was studied by Aiba, who demonstrated that the rate of enzymatic hydrolysis strongly depends not only on DDA, but on the sequence of deacetylated groups as well.12 Thus, the issue of stability and durability of rather thin chitosan coatings on the surface of heart valves bio prostheses still requires detailed investigation. In the present article, we study the stability of chitosan coating, applied onto the xenogenic tissue from carbonic acid solutions under high pressure on the small animal model. In experiments held on the rat model, we determine the amount of chitosan, remaining on collagen matrices of tissue heart valve prosthesis after in vivo degradation within given time intervals. We apply advanced tritium-labelling technique in order to ensure high sensitivity and reliability in the detection of the residual chitosan amount. The other issue we address here is the influence of the treatment by pressurized solvent on the structure of pericardial tissue. MATERIAL AND METHODS
Deposition of coatings In the present work, we used three chitosan samples, with similar degree of deacetylation (DDA), but varying in molecular mass. Samples “Chitosan 50” and “Chitosan 100” having the average molecular mass of 50 and 100 kDa, and DDA of 95 and 97%, respectively, were produced by PRIMEX, Iceland. The third sample, “Chitosan 210,” with molecular mass around 210 kDa and DDA of 84% was from Sigma-Aldrich (catalogue number #448869). High purity CO2 (>99.997%; Linde Gas Rus, Russia) and freshly purified Milli-Q water (Milli-Q Synthesis) were used
2
GALLYAMOV ET AL.
for preparation of carbonic acid solutions. High pressure setup used is described elsewhere.5 In the present work, pretreated collagen-rich matrices of bovine pericardium were used. The pretreatment was carried out in accordance with technological regulations approved for surgery practice in A.N. Bakulev Scientific Center for Cardiovascular Surgery. The samples were cut out of bovine pericardium tissue in form of square pieces with the lateral size of 1 3 1 cm2. The samples were stabilized in 0.625% aqueous solution of GA in HEPES buffer and washed in 1% sodium dodecyl sulfate aqueous solution and then once again in 0.625% GA. The thickness of the samples was (0.65 6 0.02) mm and the weight of a dried sample was about (10.6 6 1.4) mg. The stabilized samples were then rinsed in sterile saline solution and placed in the container of the high pressure setup, where chitosan dissolved in carbonic acid solutions was applied onto the matrices. The dissolving of chitosan in carbonic acid solutions and the application of the coating onto samples were carried out according to the procedure previously described in Ref. 5. The determination of the amount of chitosan applied onto matrices by means of tritium labelling was performed similarly to the previously described procedure.5 In this study, an aqueous suspension of chitosan with the specific radioactivity of 250 kBq/mg (asp) was obtained and lyophilized. The obtained chitosan flakes were then dissolved in carbonic acid and applied as a coating onto collagen tissue following the procedure described previously5. Tritium thermal activation technique was used in order to introduce tritium into chitosan molecules.13 To determine the amount of chitosan bound to collagen tissue a 10–17 mg sample of coated collagen matrix was mixed with 0.15 mL of concentrated nitric acid in a glass vial, closed with a polyethylene cap, and heated up until the sample was completely dissolved. The heating was continued for another 5 min, and then the vial was cooled down to 48C. The resulting solution was then mixed with 18 mL of Ultima Gold scintillation cocktail (PerkinElmer) and the emitted intensity (I) of tritium beta radiation was measured by means of the liquid scintillation spectrometer RackBeta 1215 (LKB, Finland) with the registration efficiency (e) close to 35%. The amount of chitosan was calculated as m5 eaIsp . In vitro testing of chitosan coatings degradation For model in vitro tests chitosan films were applied onto freshly cleaved mica surface from carbonic acid solution under the same conditions as it was done for pericardium matrices. The same samples of chitosan were used. The concentrations of chitosan solutions were 5 mg mL21, respectively. After incubation a small piece of mica was cut from each mica sample, and control AFM imaging was performed. The rest of the mica sample was subjected to chitosan degradation procedure either in lysozyme solution, or in hydrogen peroxide assay. Degradation in lysozyme was performed in freshly prepared lysozyme solution in 0.1 M PBS buffer, the concentration of lysozyme being 1–5 mg mL21. The chicken egg lysozyme
CHITOSAN COATINGS WITH ENHANCED BIOSTABILITY IN VIVO
ORIGINAL RESEARCH REPORT
TABLE I. Solubility of Chitosan Samples in Carbonic Acid Solutions, Mean Mass of Adsorbed Chitosan on the Collagen Matrices from Carbonic Acid Solutions with Average Statistical Deviation of the Mass as Detected by Tritium Labelling Method
Chitosan Sample Chitosan 100 Chitosan 50 Chitosan 210
Chitosan Solubility in Carbonic Acid (g L21)
Mass of Chitosan Adsorbed on the Collagen Matrices (mg)
The Ratio of Chitosan/Matrix (%)
361 862 10 6 2
0.032 6 0.002 0.047 6 0.001 0.109 6 0.008
0.30 6 0.05 0.45 6 0.06 1.03 6 0.15
was purchased from AppliChem PanReac, catalogue number A3711,000. The activity of lysozyme was indicated as 22,800 U mg21. The activity of lysozyme was additionally tested using EnzCheck Lysozyme Assay Kit (E-22013) (Molecular Probes, Inc). The resulting activity was roughly 40,000 U mg21. Chitosan-coated mica plates were immersed into the lysozyme solution described above and incubated at 378C for 72 h. The samples were then rinsed with MQ water and dried in airflow. AFM imaging was performed straight after. For control samples, pure mica plates were immersed into the same solution. Oxidative-reductive degradation (ORD) of chitosan coatings with H2O2 in the presence of Fe31 was performed in a 130 mM solution of H2O2 containing also 1 mM of FeCl3. The samples of chitosan-coated mica were incubated for 72 h at 378C. The samples were then rinsed with MQ water and dried in airflow and imaged with AFM. For control samples, pure mica plates were immersed into the same solution. AFM images were performed using MultiMode IIIa (former Digital Instruments, USA) scanning probe microscope operating in tapping mode in air. In vivo experiment on rat model In vivo chitosan degradation tests were performed on rats, which, due to their fast metabolism, can serve as a convenient model for in vivo express-testing with relatively short characteristic times. Collagen-rich pericardium samples of about 1 3 1 cm2 lateral size coated with chitosan (applied from carbonic acid solutions) were implanted subcutaneously into male Wistar rats, aged 1 month, weighting 90–115 g (KrolInfo, Russia), under anesthesia (20 mg kg21 of zolazepam & tiletamine (Zoletil) with 3 mg kg21 of xylazine). After the implantation the animals were kept in normal conditions and received food ad libitum. In 4 months after the implantation, the animals were sacrificed by injection of an overdose of sodium thiopental. The collagen samples were explanted from the animals, rinsed in buffered saline in order to remove remnants of animal blood. No signs of any inflammation response were encountered. The tests for immunogenity were carried out in accordance with ISO 10993-10-2011, ISO 10993-6–2011 protocols. The mass of remaining chitosan was then determined as described above. The protocol for euthanasia of animals was developed in accordance with FELASA Guidelines and Recommendations for euthanasia of experimental animals, the Animal Welfare Act, the NIH Guide for Care and Use of Laboratory Animals and was approved by A.N. Bakulev Scientific Center for
Cardiovascular Surgery in compliance with internal guidelines for care and use of laboratory animals. For each sample type, 10 rats were used. Histological sections Samples were fixed for histological analysis in 10% solution of neutral formalin for 7 days, with its subsequent gradual exchange for alcohols with increasing concentrations, and then the samples were embedded in paraffin. The prepared 4–6 lm microtomic sections were stained with hematoxylin & eosin as well as with picric acid and fuchsin (van Gieson’s stain). The investigation and analysis of histologic preparations were carried out using Olympus BX51 optical microscope. RESULTS
Evaluation of mass of chitosan adsorbed on xenogenic pericardium matrices from carbonic acid solutions by means of tritium labelling Chitosan macromolecules were applied onto pericardium collagen matrices by means of direct adsorption from their solution in carbonic acid. We determined the typical amount of chitosan that had adsorbed onto the xenogenic tissue of heart valve bioprosthesis from saturated solutions in carbonic acid by means of tritium labelling of chitosan. The total amount of the adsorbed material was rather small, thus beyond the reliability limit of typical gravimetrical study. Therefore, tritium labelling technique seems to be the most relevant method for reliable detection of such small variations of mass. The obtained data are summarized in Table I. Histologic analysis of chitosan-treated pericardium samples To visualize the structure of the pericardium matrices treated with chitosan in carbonic acid and in attempt to reveal the chitosan coating on them, histological sections were prepared. The images of histological sections of pericardium matrices are represented in Figure 1. Histological analysis with typical differential stains for collagen (hematoxylin & eosin and Van Gieson’s) reveals that coating with chitosan in carbonic acid does not essentially affect the structure of collagen fibers (unlike some other solvents for chitosan, e.g., the conventional one, acetic acid, which may noticeably disturb the organization of collagen fibrils14). Indeed, the appearance and arrangement of fibrils is quite typical. Yet, some collagen fibrils are distributed randomly and packless (Figure 1a). Van Gieson staining reveals reduced fuchsinophilicity of the structures (Figure 1b). Presumably, the observed reduced intensity of the differential
JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS | MONTH 2017 VOL 00B, ISSUE 00
3
Stability of chitosan coating on pericardium tissue in rat model In this series of experiments we analyzed the amount of chitosan remaining on bovine pericardium matrices after in vivo experiments on rat model. Collagen matrices with tritium-labeled chitosan coating of known mass were subcutaneously implanted into rats and left for 4 months. After 4 months the matrices were explanted and the amount of tritium-labelled chitosan remaining on matrices was determined. It was revealed that in two out of three investigated chitosan samples, for which the mass ratio of adsorbed chitosan to the matrix weight was 0.2 and 0.3% («Chitosan 50», «Chitosan 100», respectively), the amount of adsorbed chitosan did not change after in vivo presence. The data is summarized in Table II. Alternatively, for “chitosan 210” sample, which demonstrated the highest level of adsorption from carbonic acid solution compared to other samples, the mass of the adsorbed chitosan reduced by 15% after the in vivo experiment. Thus, we can conclude that using tritium labeling we still reliably reveal a detectable amount of chitosan, remaining on collagen matrices after the presence in animal (rat) body, which models the presence of valve prosthesis in human body.
FIGURE 1. Histological sections of pericardium matrices treated with chitosan in carbonic acid stained with hematoxylin & eosin (a) and with Van Gieson stain (b). Optical magnification 4003. Image size 500 3 376 lm2. Scale bars correspond to 50 lm.
staining is the manifestation of the influence of chitosan coating, which may affect binding of the stains molecules to the collagen matrix. Still, the demonstrated fact, that the organization of collagen fibrils is not significantly disturbed by the exposure in solutions of chitosan in carbonic acid, correlates well with our previous observations that mechanical properties of the tissue were not deteriorated by the treatment.5
In vitro testing of chitosan degradation In AFM images of all three samples one can clearly see a film of chitosan, densely covering all the surface of mica substrates. The thickness of the film was measured by means of applying very high force in AFM while scanning a smaller area, followed by subsequent imaging a larger area under normal imaging conditions. The measured thickness varied from 20 to 120 nm, depending on the location on the sample. No “islands” of uncovered mica were detected. After oxidative-reductive degradation (ORD), no detectable chitosan films were revealed in either of the samples, and images did not show any reliable difference from control samples of pure mica immersed in ORD solution. After incubation in lysozyme solution, the degradation of chitosan coatings was not essential (Figure 2). All the surface of mica was still covered with chitosan, the overall amount of the adsorbed material, being in correlation with
TABLE II. Data on Stability of Chitosan Coverage on Bovine Pericardium Tissue After In Vivo Experiment: Mass of Chitosan on the Matrix Before and After the In Vivo Incubation as Detected by Tritium Labelling Method, the Mass Ratio of Chitosan/ Matrix Before and After In Vivo Experiment and the Portion of Chitosan Remaining in the Matrix After the In Vivo Experiment Mass of the Chitosan on the Matrix
Sample Chitosan 100 Chitosan 50 Chitosan 210
The Mass Ratio of Chitosan/Matrix
Before In Vivo Experiment (mg)
After In Vivo Experiment (mg)
Before In Vivo Experiment (%)
After In Vivo Experiment (%)
The Portion of Chitosan Remaining in the Matrix After the In Vivo Experiment (%)
0.032 6 0.002 0.047 6 0.001 0.109 6 0.008
0.033 6 0.006 0.046 6 0.014 0.085 6 0.011
0.30 6 0.05 0.45 6 0.06 1.03 6 0.15
0.31 6 0.07 0.44 6 0.14 0.80 6 0.15
102 6 21 98 6 30 78 6 12
The Difference in Chitosan Amount Before and After In Vivo Experiments, ANOVA, p valuesa 0.85 0.91 0.00014
a Values of p of
