Embed Size (px)
Citation preview
This article was downloaded by: [Soonchunhyang University], [Byong-Taek Lee]On: 06 October 2014, At: 19:06Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of Biomaterials Science,Polymer EditionPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/tbsp20
Evaluation of the cytocompatibilityhemocompatibility in vivo bone tissueregenerating capability of different PCLblendsAndrew R. Padalhina, Nguyen Thuy Ba Linha, Young Ki Minb &Byong-Taek Leea
a Department of Regenerative Medicine, Institute of TissueRegeneration, College of Medicine, Soonchunhyang University,Cheonan 330-090, Koreab Department of Physiology, College of Medicine, SoonchunhyangUniversity, Cheona, KoreaPublished online: 22 Jan 2014.
To cite this article: Andrew R. Padalhin, Nguyen Thuy Ba Linh, Young Ki Min & Byong-Taek Lee(2014) Evaluation of the cytocompatibility hemocompatibility in vivo bone tissue regeneratingcapability of different PCL blends, Journal of Biomaterials Science, Polymer Edition, 25:5, 487-503,DOI: 10.1080/09205063.2013.878870
To link to this article: http://dx.doi.org/10.1080/09205063.2013.878870
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoever orhowsoever caused arising directly or indirectly in connection with, in relation to or arisingout of the use of the Content.
This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions
Dow
nloa
ded
by [
Soon
chun
hyan
g U
nive
rsity
], [
Byo
ng-T
aek
Lee
] at
19:
06 0
6 O
ctob
er 2
014
Evaluation of the cytocompatibility hemocompatibility in vivo bonetissue regenerating capability of different PCL blends
Andrew R. Padalhina, Nguyen Thuy Ba Linha, Young Ki Minb and Byong-Taek Leea*
aDepartment of Regenerative Medicine, Institute of Tissue Regeneration, College of Medicine,Soonchunhyang University, Cheonan 330-090, Korea; bDepartment of Physiology, College of
Medicine, Soonchunhyang University, Cheona, Korea
(Received 15 October 2013; accepted 20 December 2013)
In this study, the optimized formulations of polycaprolactone (PCL) combined withpoly(lactic-co-glycolic acid) (PLGA), gelatin (GEL), and biphasic calcium phosphate(BCP) were analyzed in terms of cytocompatibility with bone-related cells, hemo-compatibility, and in vivo bone-regenerating capacity to determine their potentials forbone tissue regeneration. Fiber morphology of PCL/GEL and PCL/BCP electrospunmats considerably differs from that of the PCL membrane. Based on the contactangle analyses, the addition of GEL and PLGA was shown to reduce the hydropho-bicity of these membranes. The assessment of in vitro cytocompatibility usingMC3T3-E1 cells indicated that all of the membranes were suitable for pre-osteoblastproliferation and adhesion, with PCL/BCP having a significantly higher reading afterseven days of incubation. The results of the in vitro hemocompatibility of the differ-ent fibrous scaffolds suggest that coagulation and platelet adhesion were higher forhydrophobic membranes (PCL and PCL/PLGA), while hemolysis can be associatedwith fiber morphology. The potential of the membranes for bone regeneration wasdetermined by analyzing the microCT data and tissue sections of samples implantedin 5 mm sized defects (one and two months). Although all of the membranes weresuitable for pre-osteoblast proliferation, in vivo bone regeneration after two monthswas found to be significantly higher in PCL/BCP (p < 0.001).
Keywords: PCL; electrospinning; in vitro cytocompatibility; hemocompatibility;in vivo bone regeneration
1. Introduction
Graft transplantation has been typically favored over artificial implants when dealingwith tissue regeneration, due to its inherent advantage of enabling the use of similartissue that has been matched with the recipient form donors. However, using tissuegrafts also poses significant disadvantages, namely donor availability, donor sitemorbidity, graft rejection, and to a lesser degree, donor–recipient pathogen transmis-sion.[1] Numerous studies have been conducted on both naturally occurring andsynthetic polymers as surface modifiers and components for different tissue scaffoldsthat lead to the improved biocompatibility of biomaterials used for regenerative medi-cine such as bone tissue engineering.[1–3]
Among the polymers used in the field of tissue engineering, aliphatic polyesters arewidely used due to their biodegradability and biocompatibility.[3,4] Various research
*Corresponding author. Email: [email protected]
© 2014 Taylor & Francis
Journal of Biomaterials Science, Polymer Edition, 2014Vol. 25, No. 5, 487–503, http://dx.doi.org/10.1080/09205063.2013.878870
Dow
nloa
ded
by [
Soon
chun
hyan
g U
nive
rsity
], [
Byo
ng-T
aek
Lee
] at
19:
06 0
6 O
ctob
er 2
014
have been conducted on different methods of fabricating tissue scaffolds usingpolycaprolactone (PCL). Aside from being a relatively cheaper biocompatible syntheticpolymer, PCL has been proven to be a versatile substance capable of forming stablemicrospheres, porous bodies, and electrospun membranes used as either drug deliverysystems or tissue scaffolds. Whether as a monolithic scaffold or as a compositecombined with other biocompatible materials, PCL can be considered as a good basematerial for developing scaffolds for tissue engineering.[5] Subsequent studies onelectrospun PCL-based fibrous scaffolds have provided a good background for applyingthese materials in tissue regeneration studies.[6–12] In this study, three optimizedbinary PCL mixtures have been tested to define their respective performances in termsof contributing to bone regeneration.
The purpose of this study is to compare these binary systems of PCL-based electro-spun membranes by taking previously optimized systems fabricated by incorporating aco-polymer, hydrolyzed protein (gelatin (GEL)) and biphasic calcium phosphate (BCP)based powder. Each additive combined with PCL has been determined to improve thebiocompatibility of the resulting optimized membranes in vitro; however, comparisonbetween these materials in terms of application for bone tissue regeneration is yet to beestablished through in vitro and in vivo experimentation. The first optimized membraneis a mixture of copolymers PCL and poly (lactic-co-glycolic acid) (PLGA). An initialstudy has been conducted for the application of PCL/PLGA blend as a scaffold for skin[7] and for general tissue engineering purposes [8] tested with keratinocytes and fibroustissue cells (L-929), respectively. The optimized mixture is composed of PCL and GELand has already been tested for tissue engineering and dermal reconstruction.[9,10]More recently it has also been improved through the loading of hydroxyapatite particles[11] for bone regeneration application. The final optimized membrane composed ofPCL and BCP powder was also selected due to its high in vitro performance based ona previous report using L929 cells, although it has not yet been tested in vivo.[12]
The in vitro cytocompatibility and bone tissue regeneration potential of theoptimized formulations of PCL/PLGA (80:20 ratio), PCL/GEL (50:50 ratio), andPCL/BCP (50% weight:volume) scaffolds were evaluated through proliferation assay,observation of attachment behavior through confocal microscopy, and implantation in5 mm diameter calvarial defects in rats. An additional parameter, hemocompatibility,was also tested to observe blood reaction upon contact with the tissue scaffolds. Thehemocompatibility in terms of hemolysis or break down of red blood cells upon mate-rial contact, blood cell attachment on the surface of scaffolds, and platelet adhesionhave been taken into consideration, noting that blood would be the primary fluid thatwill come in contact with implants. The healing process of damaged bone tissueinvolves the accumulation and stabilization of blood clots and dissolution of hematoma,which in turn provides the factors for initiating chemotaxis for cell-mediatedrepair.[13–17]
2. Materials and methods
2.1. Materials
Materials for fabricating the different electrospun membranes PCL (Mn 80,000), PLGA(85:15, Mw 50,000–75,000) and GEL (from porcine skin) were purchased from Sigma-Aldrich, USA. BCP powder measuring around 50–100 nm was synthesized throughmicrowave assisted process using Ca (OH)2 (Aldrich Chemical) and H3PO4 (DongwooFine Chemicals, Korea) as per reference.[18] Solvents used to dissolve the polymer
488 A.R. Padalhin et al.
Dow
nloa
ded
by [
Soon
chun
hyan
g U
nive
rsity
], [
Byo
ng-T
aek
Lee
] at
19:
06 0
6 O
ctob
er 2
014
components, tetrahydrofuran (THF, minimum 99%), and dimethylformamide (DMF,99%), were also purchased from Sigma–Aldrich, USA. The other two solvents, methylchloride and 2,2,2-trifluoroethanol (TFE, 99.0%), were procured from SK ChemicalsCo., Korea and Fluka, Buchs, Switzerland, respectively. Methylene chloride waspurchased from Dae Jung chemicals, Korea. 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl-tetrazolium bromide (MTT) was purchased from GIBCO Carlsbad, CA, USA.Dimethylsulfoxide (DMSO, 99.0%) was obtained from Samchun Pure Chemical Co.,Ltd (Pyeongtaek City, South Korea). Minimum essential medium (MEM) was obtainedfrom HyClone (Logan, UT, USA). Pre-osteoblast (MC3T3-E1) mice cells wereobtained from the American Type Culture Company, USA. All chemicals and solventswere of analytical reagent grade.
2.2. Fabrication of fibrous membrane scaffolds
The scaffolds were fabricated using different PCL blends, as optimized in previousstudies [6–12] and using a membrane composed of 12 wt.%/v PCL as the control. Theoptimized membranes were made by mixing PLGA, GEL, and BCP in separateindependent solutions with PCL. For creating PCL/PLGA and PCL/BCP blends, a solu-tion of 12 wt.%/v PCL was dissolved in a solvent containing 40% THF, 40% DMF,and 20% MC. PLGA solution was made by dissolving 10 wt.%/v of PLGA using thesame solvent. The PCL/PLGA blend was made by combining 1:4 ratio of PLGA andPCL, while the PCL/BCP blend was made by adding 50 wt.%/wt. of BCP in the 12wt.%/v PCL solution. The PCL/GEL blend was achieved by combining equal amountsof separate solutions of 12 wt.%/v of PCL and 12 wt.%/v GEL, both dissolved in TFE.
Each PCL blend was then electrospun using an electrospinning machine (Electros-pinning System, NanoNC, Korea) to create nanofibrous membranes. Each membranerequired different parameters due to the varying material compositions. Their respectiveelectrospinning parameters are enumerated as follows (voltage-kV, needle gauge – G,flow rate – ml/h, and tip-collector-distance – cm): PCL – 20 kV, 23 G, 0.5 ml/h, 20 cm;PCL/PLGA – 25 kV, 25 G, 0.5 ml/h, 20 cm; PCL/GEL – 25 kV, 21 G, 0.5 ml/h, 17 cm;and PCL/BCP – 20 kV, 21 G, 2 ml/h, 12 cm.
2.3. Characterization of fibrous membrane scaffolds
After fabrication, the dry samples were placed on a scanning electron microscope(SEM) mount and coated with platinum (Cressington 108 Auto). The fiber morphologyand average fiber diameter were observed and measured from images taken using SEMmicroscopy (SEM, JSM-7401F). Energy-dispersive X-ray spectroscopy (EDS) profilewas also taken to confirm the presence of elemental components associated with GELand BCP. The contact angle of each membrane scaffold was also tested using EasyDropContact Angle Measuring System (DSA, KRUSS GmbH).
2.4. Cell adhesion, proliferation and viability profiles
Celll adhesion, viability, and proliferation were observed using a pre-osteoblast(MC3T3-E1) commercial cell line. Circular samples of each membrane measuring 6mm in diameter were sterilized using ethanol and UV irradiation, and were conditionedwith MEM media containing 10% fetal bovine serum and 5% penicillin streptomycinfor 20 min prior to seeding with 1.5 × 104/ml of M3CT3-E1 cells.
Journal of Biomaterials Science, Polymer Edition 489
Dow
nloa
ded
by [
Soon
chun
hyan
g U
nive
rsity
], [
Byo
ng-T
aek
Lee
] at
19:
06 0
6 O
ctob
er 2
014
Observation of cell adhesion was carried out by confocal microscopy of seededmembrane samples stained with FITC and DAPI, while live and dead cell analysis wascarried out by staining seeded samples with Calcein AM and Ethidium homodimer, oneand five days after seeding. Micrographs of each sample were taken using a confocallaser scanning microscope (CLSM, Fluoview 1000, Olympus). Quantification of livingand dead cells was performed through computer aided analysis of at least 10 micro-graphs per sample. The images were analyzed using the accompanying FV10-ASW 3.0Viewer software and ImageJ. Cell proliferation on each scaffold was measured at 1, 3,5, and 7 days after seeding by conducting an MTT assay and using an ELISA reader(TECAN 150, Turner Biosystems) at 595 nm.
2.5. Hemocompatibility
To determine the blood reaction upon contact with each material, a hemocompatibilitytest was conducted on each individual rat previously implanted with a specific sample.A minimum amount of anti-coagulated blood, 1–2 ml, was aseptically drawn from eachindividual rat using a syringe with a 26 gauge needle and containing 1:9 ratio ofCitrate-dextrose solution (Sigma). Each optimized material was tested for hemolyticeffect, blood cell deposition, and platelet adhesion. The following procedures are basedon several studies on blood contacting biomaterials [14–17] with some slightmodifications.
Hemolysis due to material contact was evaluated by submerging 6.0 mm diametersamples in 1 ml of test solution consisting of 1:10 ratio of anti-coagulated wholeblood in sterile PBS. To eliminate hemolysis that might have resulted from thehandling of blood samples prior to testing, a negative control was also preparedconsisting of the test solution without any submerged material and a positive controlwas prepared consisting of a similar amount of anti-coagulated blood diluted indistilled water. After 30 and 60 min of exposure to test samples, the test solutionswere removed and centrifuged at 3000 rpm for 3 min to pellet out the cellularcomponents of the solution. To determine the amount of free hemoglobin releasedfrom hemolyzed erythrocytes, the supernatant was placed in a 96-well plate and readfor absorbance at 540 nm. Blood cell deposition on material surface was analyzed byapplying 200 μl of anti-coagulated blood on 6.0 mm samples which were then left tostand for 5, 10, and 15 min. After each sampling time, non-adherent blood cells werewashed off the membranes using PBS and adherent cells were quantified throughhemolysis.
Platelet adhesion was observed using platelet rich plasma (PRP). PRP was preparedby centrifugation of 10 ml anti-coagulated blood at 2500 rpm for 5 min and removingthe pelleted red blood cells, leaving behind the buffy coat in the middle and the uppersupernatant consisting of the plasma. The supernatant was again centrifuged at 1000rpm for 5 min to remove the remaining RBC component. Taking into consideration thatonly a very small amount of PRP can be prepared from a very small blood sample, theplasma was no longer removed and the volume of the solution was increased up to 5ml and the platelets were resuspended through vortex mixing.
A similar experimental design based on a previous study [19] was employed toquantify the amount of platelets adhered to the sample membranes. MTT assay wasconducted to quantify the number of platelets attached to the membranes after 15 and30 min of incubation.
490 A.R. Padalhin et al.
Dow
nloa
ded
by [
Soon
chun
hyan
g U
nive
rsity
], [
Byo
ng-T
aek
Lee
] at
19:
06 0
6 O
ctob
er 2
014
2.6. In vivo tests
The in vivo bone regeneration of the different PCL blends was evaluated by implanta-tion on rat calvarial defects. Circular samples of PCL, PCL/PLGA, PCL/GEL, andPCL/BCP measuring 6 mm in diameter were cut out using a paper puncher after which3–4 layers of each samples were stacked together to form a single implant material.Prior to implantation, each implant material was briefly washed with 70% ethanol andsterilized with ultraviolet radiation on both sides. Sprague Dawley rats were sedatedand circular defects measuring 5 mm in diameter were created on both sides of thefrontal plate of the skulls. The left side defect was implanted with a sample membranepre-soaked in phosphate buffered solution, while the right side defect was consideredas a control and no implant was placed. Implant samples were extracted after four andeight weeks of recovery. Micro CT data of the formalin fixed samples were recordedand tissue sections were stained with hematoxylin, eosin, and Masson’s trichrome forfurther analysis.
2.7. Statistical analysis
Each experiment was repeated at least three times on different days and data wereexpressed as the mean ± SD and analyzed through one-way ANOVA.
3. Results and discussion
3.1. Characterization of fibrous membrane scaffolds
The material characteristics of the fabricated samples closely resemble those of theoriginal descriptions from previous studies.[7–9,12] Both the fiber morphology andmeasurements of fiber diameters approximately match those of the former descriptions,with only a marginal difference. Figure 1 shows the SEM micrographs of fibrous mem-brane scaffolds composed of different PCL blends. Each PCL blend, with the exceptionof PCL/PLGA, has a distinct fiber morphology. For this experiment, the PCL mem-brane was established as a control membrane. The PCL (A) and the PCL/PLGA (B)membranes are basically composed of approximately similar diameter fibers rangingfrom 0.51 to 0.96 μm. On the other hand, membranes consisting of composite materi-als, PCL/GEL (C) and PCL/BCP (D), both have a highly altered fiber morphologycompared to that of the PCL membrane. The PCL/GEL membrane consists of fiberswith different sized diameters (0.25–1.03 μm). Incorporation of GEL into the electro-spun membrane was confirmed through EDS data, showing the presence of nitrogenwithin the fiber, which can only be derived from the hydrolyzed proteins. Of the mem-branes, the PCL/BCP composite membrane has the most altered fiber morphology.Aside from having a wider range of fiber diameter (1.3–7.8 μm), the fibers formed fromthe solution are highly irregular along their length. Some clumps of the PCL/BCPblend can be seen at random sites, which contribute to the rough overall uneven surfacemorphology of the fibers. In addition, the incorporation of the BCP powder was con-firmed through EDS showing calcium content within the fibers (Figure 1(D1)), whichcan only be derived from the BCP powder.
To determine the hydrophilic properties of each membrane scaffold, the sampleswere tested for water contact angle. The average of four measurements per sample wasused to determine the respective contact angles of each optimized membrane scaffold(Figure 2). Among the optimized scaffolds, the PCL/GEL (C) blend was the most
Journal of Biomaterials Science, Polymer Edition 491
Dow
nloa
ded
by [
Soon
chun
hyan
g U
nive
rsity
], [
Byo
ng-T
aek
Lee
] at
19:
06 0
6 O
ctob
er 2
014
hydrophilic, yielding the lowest contact angle, 80.7 ± 5.77°, while the controlmembrane composed of 12% PCL (A) was the most hydrophobic, having the highestreading of 97 ± 6.02°. The blends, PCL/BCP (D) and PCL/PLGA (B), were relativelymore hydrophilic than the control membrane, but less hydrophilic than the PCL/GELmembrane (91.75 ± 2.81° and 93.53 ± 2.91°, respectively). As expected, PCL/GEL had
Figure 2. Contact angle measurements of PCL (A), PCL/PLGA (B), PCL/GEL (C), and PCL/BCP (D) electrospun membranes.
Figure 1. SEM micrographs showing respective fiber morphologies of PCL (A), PCL/PLGA(B), PCL/GEL (C), and PCL/BCP (D) electrospun membranes. EDS data showing presence ofnitrogen in electrospun GEL fibers (C1) interspersed with larger PCL fibers with lower nitrogenreading (C2) in PCL/GEL mat. Presence of calcium is also confirmed through EDS profile ofPCL/BCP electrospun membranes.
492 A.R. Padalhin et al.
Dow
nloa
ded
by [
Soon
chun
hyan
g U
nive
rsity
], [
Byo
ng-T
aek
Lee
] at
19:
06 0
6 O
ctob
er 2
014
the lowest contact angle measurement due to hydrophilic property and the ability toabsorb and retain the water of the GEL.[20] The addition of a more hydrophilicpolymer (PLGA) and BCP powder in the PCL electrospun fibers also reduces thecontact angle to a lesser extent by contributing to the limited hydrophilic property.
3.2. Cell adhesion, proliferation, and viability profiles
In vitro cytocompatibility is an important factor for developing materials suitable fortissue engineering. In this study, PCL optimized membranes combined with PLGA,GEL, and BCP have been tested using MC3T3-E1 cells to chiefly establish suitabilityof these membranes for bone tissue growth. Confocal microscopy of seeded cellsindicates minute differences between the adhesion behavior and viability of MC3T3-E1cells on different micro-fibrous membrane scaffolds. Cell proliferation on the seededscaffolds was determined using MTT assay. Quantitative data of cell viability were alsogathered from confocal micrographs using seeded cells stained using a live/dead cellkit. Figure 3 shows the cell proliferation profiles of MC3T3-E1 cells seeded on thePCL, PCL/PLGA, PLC/GEL, and PCL/BCP fibrous membrane scaffolds. Results showthat among these samples, PCL/BCP, followed by PCL/GEL, shows the greatest cellproliferation within seven days of culture. Statistical analysis indicates that thePCL/PLGA, PCL/GEL, and PCL/BCP electrospun membranes were significantly higherthree and seven days after cell seeding (PCL/PLGA and PCL/GEL: p < 0.05;PCL/BCP: p < 0.001). On the fifth day, there is a noticeable lag phase in which nosignificant difference is observed between the proliferations on all the samples. This ispossible considering that all the membranes were fabricated using the optimized formu-lations for tissue engineering.
Figure 4 shows the confocal micrographs of MC3T3-E1 cells seeded on PCL (A1,A2), PCL/PLGA (B1, B2), PCL/GEL (C1, C2), and PCL/BCP (D1, D2) membranescaffolds after 1–7 days. Twenty-four hours after seeding, the cell adhesion behavior onall four membranes did not show any obvious differences. However, at seven days ofincubation it can be clearly seen that the cell attachment on PCL/GEL and PCL/BCP
Figure 3. Cell proliferation of MC3T3-E1 cells seeded on PCL (A), PCL/PLGA (B), PCL/GEL(C), and PCL/BCP (D) electrospun membranes measured up to seven days.
Journal of Biomaterials Science, Polymer Edition 493
Dow
nloa
ded
by [
Soon
chun
hyan
g U
nive
rsity
], [
Byo
ng-T
aek
Lee
] at
19:
06 0
6 O
ctob
er 2
014
membranes was more extensive compared to that of the PCL and PCL/PLGAmembranes. In addition to the extensive attachments of the cells seeded on PCL/GELand PCL/BCP membranes, both membranes also contained more cells compared to theremaining samples.
Micrographs of seeded scaffolds stained for live/dead cell assay were also obtainedthrough confocal microscopy. A minimum of 10 images pictures per sample wereanalyzed and counted for the ratio of living and dead cells. As indicated in the kitmanual, all the cells in the samples are stained green and the dead cells are identifiedthrough the red stained nucleus. Cell viability for the samples is established by averag-ing the reading from each micrograph, which is calculated by subtracting the numberof dead cells from the total count of cells, dividing the difference by the total count ofcells, and then finally by multiplying the quotient by 100. Figure 5 shows the averageMC3T3-E1 cell viability count in each sample, one and seven days after seeding andconfocal micrographs for stained cells. Dead cells are red and viable cells are stainedgreen. The results of the viability count further supports the cell proliferation datasuggesting that at seven days, MC3T3-E1 cells have higher viability in the PCL/BCPelectrospun membrane. Based on the results of the in vitro tests, all of the electrospunmembranes were suitable for proliferating pre-osteoblast cells, with PCL/BCPmembranes having a the significantly higher number of cells by the end of the sevenday observation period. The initial observations showing propensity of cells to attachon the PCL/GEL membrane may be related to the aptitude of GEL to absorb fluids,allowing the perfusion of nutrients within the membrane matrix. The hydrophilicity andprotein rich surface of the PCL/GEL contribute to cell attachment and prolifera-tion.[9,10,20] Conversely, the PCL/BCP membrane also has the same effect, asdescribed in other previous studies, whereby osteoblasts cells prefer to adhere on roughsurfaces and the highly modified surface topography also provides an appropriate sub-strate for protein adsorption, consequently increasing the potential for cell adhesion[21,22] when in a serum supplemented culture media. In this study, the PCL/BCPmembrane possessed a highly altered, uneven surface suitable for osteoblastic cellattachment. This is further exemplified through observing the cell viability and
Figure 4. Confocal micrographs showing adhesion behavior of MC3T3-E1 cells on PCL(A1–A2), PCL/PLGA (B1–B2), PCL/GEL (C1–C2), and PCL/BCP (D1–D2) electrospunmembranes one and seven days after seeding.
494 A.R. Padalhin et al.
Dow
nloa
ded
by [
Soon
chun
hyan
g U
nive
rsity
], [
Byo
ng-T
aek
Lee
] at
19:
06 0
6 O
ctob
er 2
014
morphology. The live dead cell assay showed more living cells attached to the PCLmembranes that were with GEL and BCP. This is also evident through the extensivecell attachment visible from the confocal micrographs. GEL has been known to be agood platform for culturing numerous types of cells,[20] while BCP typically serves asa biomimetic component for bone regeneration.[1,2,22,23]
3.3. Hemocompatibility tests
Developing hemocompatible tissue scaffolds is crucial for not only blood vessel graftsbut also for materials used for bone tissue regeneration. Blood is the principalphysiological fluid that comes into contact with bone tissue scaffolds. The formation ofhematoma is a naturally occurring process in bone fractures and allows the release ofchemotactic elements that signals the migration of cells towards the site of injury. Thus,the effective formation of a blood clot on the surface of tissue scaffolds substantiallycontributes to the inflammatory phase of the healing process.[24,25] Platelets and theircorresponding growth factors have been found to contribute to the healing process andregeneration of injured tissues.[26–29] During the inflammatory phase, breakdown ofthe clot and surrounding dead tissues occurs, which in turn releases signaling chemicalsthat guide the migrating repair cells.[30,31] To determine the blood tissue behaviorupon contact, hemolysis, coagulation, and platelet adhesion were tested for eachoptimized PCL blend. Figure 6 shows the free hemoglobin reading after 30 and 60 minof incubating each membrane scaffold with diluted whole blood. Based on the results,the PCL/BCP membrane generated a significantly higher number of hemolyzed cellsafter 30 and 60 min, while PCL/GEL had the least number of hemolyzed cells(P < 0.001). The rough surface of PCL/BCP fibers, being abrasive against red bloodcells, could have possibly contributed to the increased hemolysis. None of the electro-spun membranes exceeded the acceptable range of red blood cell breakdown (5%) forimplant materials based on ISO 10993-4:2002.
Figure 5. Confocal micrographs showing adhesion behavior of MC3T3-E1 cells on PCL(A1–A2), PCL/PLGA (B1–B2), PCL/GEL (C1–C2), and PCL/BCP (D1–D2) electrospunmembranes one and seven days after seeding.
Journal of Biomaterials Science, Polymer Edition 495
Dow
nloa
ded
by [
Soon
chun
hyan
g U
nive
rsity
], [
Byo
ng-T
aek
Lee
] at
19:
06 0
6 O
ctob
er 2
014
Figure 6 shows the results of blood coagulation tests after 30 and 60 minincubation. Blood cell deposition was significantly higher in the PCL/GEL membranefollowed by PCL/BCP, while it was the lowest in the PCL membrane, was the lowestacross all observation periods. The absorption of blood fluids, coupled with the dissolu-tion of the majority of the GEL fibers in PCL/GEL, resulted in increased deposition ofthe blood cell component on the membrane surface. None of the membranes hadgreater than 1% blood cell deposited on their surface.
Finally, membranes were tested for platelet adhesion using PRP from whole bloodsamples. The number of platelets adhered to the surface of each membrane wasestimated using MTT assay (Figure 6). Based on the results, PCL and PCL/PLGA werethe most conducive surfaces for platelet attachment. In the case of PCL/GEL, GEL’sabsorbing capability and dissolution of GEL component again contributes to the
Figure 6. Hemocompatibility data based on hemolysis (A), blood cell deposition (B) andplatelet adhesion (C) of PCL, PCL/PLGA, PCL/GEL, and PCL/BCP electrospun membranes.
496 A.R. Padalhin et al.
Dow
nloa
ded
by [
Soon
chun
hyan
g U
nive
rsity
], [
Byo
ng-T
aek
Lee
] at
19:
06 0
6 O
ctob
er 2
014
attachment of the platelets. Protein in the blood can be adsorbed on calcium phosphatesurfaces to some extent, as suggested in previous studies,[32–34] which could thenresult in platelet adhesion. Platelets adhered on the membrane surfaces were alsoobserved through SEM microscopy (Figure 7), compared to other membranes,and the platelets adhered on PCL/GEL are highly obscured by the melted GEL(Figure 7(C)).
Based on the hemocompatibility tests, PCL and PCL/PLGA were the membranesmost suitable for forming clots and accumulating blood cells. Although not fullyinvestigated in this paper, numerous studies regarding polymer surfaces have supportedthat reduction of contact angle and reduced protein adsorption on material surfacessignificantly reduces platelet adhesion.[21,35–38] In this study, PCL and PCL/PLGAare characteristically more hydrophobic than PCL/GEL and PCL BCP, making thesesurfaces more suitable for protein adsorption which could contribute to increasedplatelet deposition.[36,38–40] In the case of PCL/GEL, capability of GEL to absorbfluids, hydrophilic and dissolution property contributes to reduced platelet attachment.Upon contact of hydrophilic surfaces to whole blood serum, plasma proteins that alsoinclude anti-platelet adhesion molecules, are adsorbed resulting to reduced plateletadhesion.[41] In addition to protein adsorption, the dissolution and swelling of theGEL component of the PCL/GEL membrane would enable entrapment and retention ofblood cells. However, the GEL in PCL/GEL membrane was not cross-linked thus, it issusceptible to break down and eventual displacement in aqueous conditions resulting tothe reduced attachment and deposition of both cells and platelets. Protein in the blood
Figure 7. SEM micrographs showing platelet adhesion on the surfaces of PCL (A), PCL/PLGA(B), PCL/GEL (C), and PCL/BCP (D) electrospun membranes after 30 min of incubation in PRP(-platelets).
Journal of Biomaterials Science, Polymer Edition 497
Dow
nloa
ded
by [
Soon
chun
hyan
g U
nive
rsity
], [
Byo
ng-T
aek
Lee
] at
19:
06 0
6 O
ctob
er 2
014
can also be adsorbed on calcium phosphate surfaces to some extent as suggested byprevious studies [32–34] which could then result to platelet adhesion but to a lesserdegree compared to bare polymer surfaces.
3.4. In vivo bone regeneration
Figure 8 shows the reconstructed 3D images of skull sections implanted with thedifferent membrane samples. Within a month (Figure 8(D1)), PCL/BCP electrospunmembrane showed increased regeneration along the margin of the 5 mm diameterdefect. This becomes more pronounced two months after implantation, with PCL/BCP(Figure 8(D2)) achieving greater bone regeneration compared with PCL, PCL/PLGA,and PCL/GEL (Figure 8(A2–C2)). 3D analysis of the micro CT data (Figure 9) alsoconfirms a significantly higher percent bone volume when PCL/BCP was implanted forboth observation periods. 3D analysis of the micro CT data supports the 3D recon-structed models of the rat calvarias. Little significant difference was observed betweenall the optimized membranes one month after implantation, with PCL/BCP having thehighest reading of percent bone volume (6.12 ± 1.95 mm3). Calculated 3D analysis indi-cates that bone volume and percent bone volume are significantly higher (p < 0.001) inPCL/BCP (12.12 ± 1.75 mm3) than in PCL (6.85 ± 0.23 mm3), and slightly higher thanthe PCL/PLGA (9.27 ± 0.85 mm3) and PCL/GEL (9.63 ± 0.45 mm3) membranes, oneand two months after implantation.
Bone regeneration on the defect sites was visualized using tissue sections stainedwith hematoxylin and eosin (H and E) and Masson’s trichrome. Figure 10 shows the Hand E stained tissue sections one month post implantation. While fibrous tissueformation is prevalent on all implanted samples, it is more pronounced in the PCL andPCL/PLGA samples (Figure 10(A–B)). All tissue sections also show bone regenerationalong the margin of the defect. Tissue sections of samples extracted two months afterimplantation (Figure 11) show markedly improved bone regeneration along the marginof the defects; however, fibrous tissue is persistent within the majority of the implanted
Figure 8. Three-dimensional reconstruction of micro CT data of rat calvaria implanted withPCL (A), PCL/PLGA (B), PCL/GEL (C), and PCL/BCP (D) electrospun membranes showingregeneration after 1 (A1–D1) and 2 (A2–D2) months.
498 A.R. Padalhin et al.
Dow
nloa
ded
by [
Soon
chun
hyan
g U
nive
rsity
], [
Byo
ng-T
aek
Lee
] at
19:
06 0
6 O
ctob
er 2
014
samples. Defects implanted with PCL/GEL membranes (Figure 11(C)) revealed moreextensive vascularization among all the other implanted samples. Although defectsimplanted with PCL/BCP and PCL/PLGA membranes (Figure 11(B) and (D)) stillpossessed extensive fibrous tissue, these implants also showed dense collagendeposition (Figure 11(B1) and (D1)) distant from that of the defect margin. Masson’strichome staining was also applied on a tissue section to further characterize tissueformation on implanted membranes. The majority of the samples did have blue stainedfibrous collagen (Figure 11(A1–D1)). However, compared to other samples, PCL/PLGAand PCL/BCP showed deposition of dense collagen. Also, compared to PCL/PLGA,PCL/BCP contained larger amounts of dense collagen (stained blue), whichis also identified to be new bone, and dense collagen with red pigmentation(Figure 11(D1)).
Bone tissue repair is initiated through the migration and differentiation of stem cellsand other surrounding tissue.[30,31,42,43] Deposition of dense collagen during callusformation in defect areas is carried out by osteoblast cells, thus scaffolds intended forbone tissue regeneration should primarily be osteoconductive, in addition to beingosteoinductive, and be capable of osteointegration.[13,30,31] Interestingly, even thoughPCL yielded the highest blood cell deposition and relatively lower hemolysis, it simplypromoted persistence of fibrous tissue up to two months after implantation. Develop-ment of a stable hematoma is crucial for healing in bone fractures. However, prolongedpresence of inflammatory agents was found to negatively affect bone regenera-tion.[25,44] Although not fully observed in the current research, this condition couldbe associated with the PCL membrane’s susceptibility for activating platelets and con-tributing to the coagulation cascade, forming a highly stable hematoma. Further studyis recommended to investigate this possibility.
Although extensive vascularization was prevalent within the defect area, this didnot contribute significantly to the overall regeneration of bone tissue on the 5 mmdefects when PCL/GEL membranes were used. PCL/BCP still generated significantlyhigher regeneration among all optimized membranes, possibly due to its potential forreduced blood cell adhesion and osteoconductivity. The combination of BCP and PCLpolymer combines the appropriate characteristics of materials for bone tissueregeneration. Hyrophobic polymers have been found to have an affinity for proteinadsorption, while calcium phosphate derived powders have long been identified as
Figure 9. Calculated tissue volume, bone volume and corresponding percent bone volumeindicate significant difference (p > 0.001) in bone tissue formation after two months.
Journal of Biomaterials Science, Polymer Edition 499
Dow
nloa
ded
by [
Soon
chun
hyan
g U
nive
rsity
], [
Byo
ng-T
aek
Lee
] at
19:
06 0
6 O
ctob
er 2
014
competent materials for bone healing due to their biodegradation, osteoconductivity,and potentially osteoinductive properties under in vivo conditions. PCL has alreadybeen proven to be a suitable material for tissue engineering, while combining it withother polymers further enhances its functionality.[45,46] The addition of BCP, or othercalcium phosphate derived components, significantly improves the bone tissueregenerating capability of electrospun PCL.
Figure 10. Tissue sections of one month implant samples of PCL (A), PCL/PLGA (B), PCL/GEL (C), and PCL/BCP (D) electrospun membranes stained with hematoxilyn and eosin 5 mmrat calvaria defect showing mostly fibrous tissue growth for the majority of the samples and ini-tial bone regeneration along the defect margin.
500 A.R. Padalhin et al.
Dow
nloa
ded
by [
Soon
chun
hyan
g U
nive
rsity
], [
Byo
ng-T
aek
Lee
] at
19:
06 0
6 O
ctob
er 2
014
4. Conclusion
This study has successfully compared the bone regeneration capability of severaloptimized binary formulations of PCL. Even though all membranes show competitiveresults regarding in vitro cytocompatibility and hemocompatibility, the incorporation ofa calcium phosphate containing compound in the composite PCL/BCP membrane hasshown that this is the most effective material for bone regeneration based on the resultsof the in vivo implantation. This reveals that although all of the membranes werefabricated using optimized formulations and conditions, the incorporation of anosteoconductive component resulting in a composite material was more conducive forbone tissue growth. Further studies are suggested to investigate the possibleenhancement of the PCL/BCP for bone tissue engineering.
AcknowledgmentsThis work was supported by Mid-career Research Program through NFR grant funded by theMEST (NO 2009-0092808), Republic of Korea and partially supported by SoonchunhyangUniversity Research Fund. The author would also like to thank Rose Ann Franco and Thi-HeipNguyen for their help in the sample preparation and Kim Shin Woo and Kim Hyoung Suk forfacilitating the in vivo experiments.
References[1] Salgado AJ, Coutinho OP, Reis RL. Bone tissue engineering: state of the art and future
trends. Macromol. Biosci. 2004;4:743–765.
Figure 11. Micrographs of tissue sections of implanted samples of PCL (A, A1, A2), PCL/PLGA (B, B1, B2), PCL/GEL (C, C1, C2), and PCL/BCP (D, D1, D2) electrospun membranesstained with hematoxilyn and eosin, and Masson’s trichome, two months after implantation incritical sized rat calvaria defect (blood vessel – ▴, collagen – ♦, new bone – ).
Journal of Biomaterials Science, Polymer Edition 501
Dow
nloa
ded
by [
Soon
chun
hyan
g U
nive
rsity
], [
Byo
ng-T
aek
Lee
] at
19:
06 0
6 O
ctob
er 2
014
[2] Schroeder JE, Mosheiff R. Tissue engineering approaches for bone repair: concepts andevidence. Injury. 2011;42:609–613.
[3] Tian H, Tang Z, Zhuang X, Chen X. Biodegradable synthetic polymers: preparation,functionalization and biomedical application. Prog. Polym. Sci. 2012;37:237–280.
[4] Zong C, Xue D, Yuan W, Wang W, Shen D, Tong X, Shi D, Liu L, Zheng Q, Gao C, WangJ. Reconstruction of rat calvarial defects with human mesenchymal stem cells and osteoblast-like cells in poly-lactic-co-glycolic acid scaffolds. Eur Cell Mater. 2010;20:109–120.
[5] Woodruff MA, Hutmacher DW. The return of a forgotten polymer – polycaprolactone in the21st century. Prog. Polym. Sci. 2010;35:1217–1256.
[6] Roohani-Esfahani SI, Nouri-Khorasani S, Lu Z, Appleyard R, Zreiqat H. The influencehydroxyapatite nanoparticle shape and size on the properties of biphasic calcium phosphatescaffolds coated with hydroxyapatite–PCL composites. Biomaterials. 2010;31:5498–5509.
[7] Franco RA, Nguyen TH, Lee BT. Preparation and characterization of electrospun PCL/PLGA membranes and chitosan/gelatin hydrogels for skin bioengineering applications. J.Mater. Sci. Mater. Med. 2011;22:2207–2218.
[8] Hiep NT, Lee BT. Electro-spinning of PLGA/PCL blends for tissue engineering and theirbiocompatibility. J. Mater. Sci. Mater. Med. 2010;21:1969–1978.
[9] Chong EJ, Phan TT, Lim IJ, Zhang YZ, Bay BH, Ramakrishna S, Lim CT. Evaluation ofelectrospun PCL/gelatin nanofibrous scaffold for wound healing and layered dermalreconstitution. Acta Biomater. 2007;3:321–330.
[10] Zhao P, Jiang H, Pan H, Zhu K, Chen W. Biodegradable fibrous scaffolds composed ofgelatin coated poly(ε-caprolactone) prepared by coaxial electrospinning. J. Biomed. Mater.Res. A. 2007;83A:372–382.
[11] Ba Linh NT, Min YK, Lee BT. Hybrid hydroxyapatite nanoparticles-loaded PCL/GE blendfibers for bone tissue engineering. J. Biomater. Sci., Polym. Ed. 2013;24:520–538.
[12] Nguyen TP, Lee BT. Fabrication and characterization of BCP nano particle loaded PCLfiber and their biocompatibility. Korean J. Mater. Res. 2010;20:392–400.
[13] Albrektsson T, Johansson C. Osteoinduction, osteoconduction and osseointegration. EurSpine J. 2001;10:S96–101.
[14] Motlagh D, Allen J, Hoshi R, Yang J, Lui K, Ameer G. Hemocompatibility evaluation ofpoly(diol citrate) in vitro for vascular tissue engineering. J. Biomed. Mater. Res. A.2007;82A:907–916.
[15] Chenglong L, Dazhi Y, Guoqiang L, Min Q. Corrosion resistance and hemocompatibilityof multilayered Ti/TiN-coated surgical AISI 316L stainless steel. Mater. Lett.2005;59:3813–3819.
[16] Zhao ML, Li DJ, Yuan L, Yue YC, Liu H, Sun X. Differences in cytocompatibility andhemocompatibility between carbon nanotubes and nitrogen-doped carbon nanotubes.Carbon. 2011;49:3125–3133.
[17] Li J, Zheng W, Zheng Y, Lou X. Cell responses and hemocompatibility of g-HA/PLAcomposites. Sci. China Life Sci. 2011;54:366–371.
[18] Lee BT, Youn MH, Paul RK, Lee KH, Song HY. In situ synthesis of spherical BCPnanopowders by microwave assisted process. Mater. Chem. Phys. 2007;104:249–253.
[19] Maekawa Y, Yagi K, Nonomura A, Kuraoku R, Nishiura E, Uchibori E, Takeuchi K. Atetrazolium-based colorimetric assay for metabolic activity of stored blood platelets.Thromb. Res. 2003;109:307–314.
[20] Gómez-Guillén MC, Giménez B, López-Caballero ME, Montero MP. Functional andbioactive properties of collagen and gelatin from alternative sources: a review. FoodHydrocoll. 2011;25:1813–1827.
[21] Wilson CJ, Clegg RE, Leavesley DI, Pearcy MJ. Mediation of biomaterial–cell interactionsby adsorbed proteins: a review. Tissue Eng. 2005;11:1–18.
[22] Garrido CA, Lobo SE, Turibio FM, LeGeros RZ. Biphasic calcium phosphate bioceramicsfor orthopaedic reconstructions: clinical outcomes. Int. J. Biomater. 2011;2011:129727.
[23] Anselme K. Osteoblast adhesion on biomaterials. Biomaterials. 2000;21:667–681.[24] Ozaki A, Tsunoda M, Kinoshita S, Saura R. Role of fracture hematoma and periosteum
during fracture healing in rats: interaction of fracture hematoma and the periosteum in theinitial step of the healing process. J. Orthop. Sci. 2000;5:64–70.
502 A.R. Padalhin et al.
Dow
nloa
ded
by [
Soon
chun
hyan
g U
nive
rsity
], [
Byo
ng-T
aek
Lee
] at
19:
06 0
6 O
ctob
er 2
014
[25] Kolar P, Schmidt-Bleek K, Schell H, Gaber T, Toben D, Schmidmaier G, Perka C, ButtgereitF, Duda GN. The early fracture hematoma and its potential role in fracture healing. TissueEng. Part B: Rev. 2010;16:427–434.
[26] Arpornmaeklong P, Kochel M, Depprich R, Kübler NR, Würzler KK. Influence of platelet-rich plasma (PRP) on osteogenic differentiation of rat bone marrow stromal cells. Anin vitro study. Int. J. Oral Maxillofac. Surg. 2004;33:60–70.
[27] Dominiak M, Łysiak-Drwal K, Solski L, Żywicka B, Rybak Z, Gedrange T. Evaluation ofhealing processes of intraosseous defects with and without guided bone regeneration andplatelet rich plasma. An animal study. Ann. Anat. 2012;194:549–555.
[28] Jeong Park YJ, Moo Lee YM, Nae Park SN, Yoon Sheen SY, Pyoung Chung CP, Lee SJ.Platelet derived growth factor releasing chitosan sponge for periodontal bone regeneration.Biomaterials. 2000;21:153–159.
[29] Marsell R, Einhorn TA. The biology of fracture healing. Injury. 2011;42:551–555.[30] McKibbin B. The Biology of fracture healing in long bones. J. Bone Joint Surg.
1978;60:150–162.[31] Wang K, Zhou C, Hong Y, Zhang X. A review of protein adsorption on bioceramics. Inter-
face focus. 2012;2:259–277.[32] Wang J, Zhang H, Zhu X, Fan H, Fan Y, Zhang X. Dynamic competitive adsorption of
bone-related proteins on calcium phosphate ceramic particles with different phase composi-tion and microstructure. J. Biomed. Mater. Res. B: Appl. Biomater. 2013;101B:1069–1077.
[33] Zhu XD, Zhang HJ, Fan HS, Li W, Zhang XD. Effect of phase composition and microstruc-ture of calcium phosphate ceramic particles on protein adsorption. Acta Biomater.2010;6:1536–1541.
[34] Chen H, Yuan L, Song W, Wu Z, Li D. Biocompatible polymer materials: role of protein–surface interactions. Prog. Polym. Sci. 2008;33:1059–1087.
[35] Comelles J, Estévez M, Martínez E, Samitier J. The role of surface energy of technicalpolymers in serum protein adsorption and MG-63 cells adhesion. Nanomedicine. 2010;6:44–51.
[36] Marsh RJ, Jones RAL, Sferrazza M. Adsorption and displacement of a globular protein onhydrophilic and hydrophobic surfaces. Colloids Surf., B. 2002;23:31–42.
[37] Liu P, Chen Q, Yuan B, Chen M, Wu S, Lin S, Shen J. Facile surface modification ofsilicone rubber with zwitterionic polymers for improving blood compatibility. Mater. Sci.Eng., C. 2013;33:3865–3874.
[38] Li Q, Wang Z, Zhang S, Zheng W, Zhao Q, Zhang J, Wang L, Wang S, Kong D. Function-alization of the surface of electrospun poly(epsilon-caprolactone) mats using zwitterionicpoly(carboxybetaine methacrylate) and cell-specific peptide for endothelial progenitor cellscapture. Mater. Sci. Eng., C. 2013;33:1646–1653.
[39] Zhu A, Lu P, Wu H. Immobilization of poly(ɛ-caprolactone)–poly(ethylene oxide)–poly(ɛ-caprolactone) triblock copolymer on poly(lactide-co-glycolide) surface and dual biofunc-tional effects. Appl. Surf. Sci. 2007;253:3247–3253.
[40] Lee JH, Lee HB. Platelet adhesion onto wettability gradient surfaces in the absence andpresence of plasma proteins. J. Biomed. Mater. Res. 1998;41:304–311.
[41] Doblaré M, Garcı́a JM, Gómez MJ. Modelling bone tissue fracture and healing: a review.Eng. Fract. Mech. 2004;71:1809–1840.
[42] Schmidt-Bleek K, Schell H, Schulz N, Hoff P, Perka C, Buttgereit F, Volk HD, Lienau J,Duda GN. Inflammatory phase of bone healing initiates the regenerative healing cascade.Cell Tissue Res. 2012;347:567–573.
[43] Opal SM. Phylogenetic and functional relationships between coagulation and the innateimmune response. Crit. Care Med. 2000;28:S77–S80.
[44] Baker SC, Rohman G, Southgate J, Cameron NR. The relationship between the mechanicalproperties and cell behaviour on PLGA and PCL scaffolds for bladder tissue engineering.Biomaterials. 2009;30:1321–1328.
[45] Ghasemi-Mobarakeh L, Prabhakaran MP, Morshed M, Nasr-Esfahani MH, Ramakrishna S.Bio-functionalized PCL nanofibrous scaffolds for nerve tissue engineering. Mater. Sci. Eng.,C. 2010;30:1129–1136.
[46] Sousa I, Mendes A, Bártolo PJ. PCL scaffolds with collagen bioactivator for applications intissue engineering. Procedia Eng. 2013;59:279–284.
Journal of Biomaterials Science, Polymer Edition 503
Dow
nloa
ded
by [
Soon
chun
hyan
g U
nive
rsity
], [
Byo
ng-T
aek
Lee
] at
19:
06 0
6 O
ctob
er 2
014
