A novel route for the production of chitosan/poly(lactide-co-glycolide) graft copolymers for electrospinning This content has been downloaded from IOPscience. Please scroll down to see the full text. Download details: IP Address: 157.182.150.22 This content was downloaded on 04/10/2013 at 05:10 Please note that terms and conditions apply. 2010 Biomed. Mater. 5 065016 (http://iopscience.iop.org/1748-605X/5/6/065016) View the table of contents for this issue, or go to the journal homepage for more Home Search Collections Journals About Contact us My IOPscience
Transcript
A novel route for the production of chitosan/poly(lactide-co-glycolide) graft copolymers for
electrospinning
This content has been downloaded from IOPscience. Please scroll down to see the full text.
Download details:
IP Address: 157.182.150.22
This content was downloaded on 04/10/2013 at 05:10
Please note that terms and conditions apply.
2010 Biomed. Mater. 5 065016
(http://iopscience.iop.org/1748-605X/5/6/065016)
View the table of contents for this issue, or go to the journal homepage for more
Home Search Collections Journals About Contact us My IOPscience
A novel route for the production ofchitosan/poly(lactide-co-glycolide) graftcopolymers for electrospinningDeming Xie1,4, Huamei Huang2, Keith Blackwood3 and Sheila MacNeil3
1 Tissue Engineering Laboratory, Department of Biomedical Engineering, Jinan University, Guangzhou,510630, People’s Republic of China2 Morphological Experiments Center of Medical College, Jinan University, Guangzhou, 510630,People’s Republic of China3 Tissue Engineering Group, Department of Engineering Materials and Division of Biomedical Sciencesand Medicine, Kroto Research Institute, University of Sheffield North Campus, Broad Lane, SheffieldS3 7HQ, UK
Received 20 July 2010Accepted for publication 20 October 2010Published 15 November 2010Online at stacks.iop.org/BMM/5/065016
AbstractBoth chitosan and polylactide/polyglycolide have good biocompatibility and can be used toproduce tissue engineering scaffolds for cultured cells. However the synthetic scaffolds lackgroups that would facilitate their modification, whereas chitosan has extensive active amideand hydroxyl groups which would allow it to be subsequently modified for the attachment ofpeptides, proteins and drugs. Also chitosan is very hydrophilic, whereas PLGA is relativelyhydrophobic. Accordingly there are many situations where it would be ideal to have acopolymer of both, especially one that could be electrospun to provide a versatile range ofscaffolds for tissue engineering. Our aim was to develop a novel route of chitosan-g-PLGApreparation and evaluate the copolymers in terms of their chemical characterization, theirperformance on electrospinning and their ability to support the culture of fibroblasts as aninitial biological evaluation of these scaffolds. Chitosan was first modified with trimethylsilylchloride, and catalyzed by dimethylamino pyridine. PLGA-grafted chitosan copolymers wereprepared by reaction with end-carboxyl PLGA (PLGA–COOH). FT-IR and1H-NMRcharacterized the copolymer molecular structure as being substantially different to that of thechitosan or PLGA on their own. Elemental analysis showed an average 18 pyranose unitintervals when PLGA–COOH was grafted into the chitosan molecular chain. Differentialscanning calorimetry results showed that the copolymers had different thermal properties fromPLGA and chitosan respectively. Contact angle measurements demonstrated that copolymersbecame more hydrophilic than PLGA. The chitosan-g-PLGA copolymers were electrospun toproduce either nano- or microfibers as desired. A 3D fibrous scaffold of the copolymers gavegood fibroblast adhesion and proliferation which did not differ significantly from theperformance of the cells on the chitosan or PLGA electrospun scaffolds. In summary thiswork presents a methodology for making a hybrid material of natural and synthetic polymerswhich can be electrospun and reacts well as a substrate for cell culture.
(Some figures in this article are in colour only in the electronic version)
4 Author to whom any correspondence should be addressed.
1. Introduction
Chitosan is a special polysaccharide macromolecular materialwith properties and bioactivities which other synthetic
polymers cannot provide. In particular the active amideand hydroxyl groups of chitosan molecules make themattractive as they provide reactive groups which are suitablefor subsequent modification and they are very hydrophilic withexcellent biocompatibility, biodegradability and mechanicalfeatures.
Polylactide in contrast is strongly hydrophobic and lacksreactive groups. Accordingly the poor reactivity of polylactidecan limit its application in tissue engineering and it is oftenmodified by the introduction of synthetic, more rapidlydegrading polyglycolide (PGA) or the introduction of naturalpolymers such as collagen, gelatin, dextran, starch, alginate,hyaluronate, chondroitin sulfate, chitosan or RGD smallpeptides [1]. One can obtain novel biomaterials by a range ofmodification methods including copolymerization, blending,coating, grafting and absorption. For example, You et al [2]prepared starch/lactide copolymers by graft polymerizationand achieved copolymers with good biodegradability but poormechanical strength. Cai et al [3] and Ouchi et al [4] reportedthe production of blends and copolymers with polylactideand dextran respectively, reporting that the copolymers hadbetter mechanical properties and cellular affinity than theblends. Additionally in recent years there have been manyapproaches [5–15] to combine natural polymers with syntheticpolymers to achieve hybrid materials with the best of bothproperties.
A controllable method for introducing chitosan intoPLA/PLGA chains would give new biomaterials with thedesirable properties of increased hydrophilicity, and theintroduction of reactive groups for modification as desired.However, as polysaccharides are highly hydrophilic and donot melt and are insoluble in most organic solvents, theycannot react directly with lactides. Hence the aim of thisstudy was to design a novel route for producing a copolymer byinitially modifying chitosan with silylation reagents so it couldbe dissolved in organic solvent [16], and then grafted withend-carboxyl PLGA. We investigated the effect of chitosanmodification and PLGA–COOH chain length in the productionof copolymers which we then electrospun and examined fortheir ability to support cell culture.
We chose electrospinning as it is proving to be a veryversatile technique for producing scaffolds. The requirementfor specialized structural basement membrane collagens toassist in cell adhesion has been reported from this [17]and many other laboratories and with electrospinning fiberscan be produced covering a range of dimensions fromnano to micro, to some extent mimicking the collagenfibrils of the extracellular matrix [18]. The open spaceconformation of electrospun scaffolds is also favorable for cellinfiltration as confirmed in an increasing number of papersin which electrospun scaffolds are used for a wide rangeof tissue engineering applications [19–26]. Accordingly inthis study, we also evaluated the chitosan/PLGA copolymersfor their ability to support culture of human dermalfibroblasts compared to chitosan or PLGA scaffolds on theirown.
2. Materials and methods
2.1. Materials
The materials used were chitosan (Sigma, deacetylation>90%, viscosity >100 mPa s in 1% acetic acidsolution), end-carboxyl PLGA (LA/GA = 75/25, Mv= 5000, Shandong Medical Devices Institute, China),trimethylsilyl chloride (Sigma), hexamethyldisilizane(Sigma), 4-dimethylamine pyridine (DMAP, 99.5%, Sigma),1,1,1,2,2,2-hexafluoroisopropanol (HFIP, Sigma), N,N-dimethylfomamide (DMF, Sigma), DMEM medium and PBSsolution (Gibco).
2.2. Modification of chitosan
The preparation of trimethylsilylation of chitosan was asdescribed in [16]. Briefly, chitosan (0.500 g, 3.1 mmolpyranose) was pulverized and added to 40 mL of pyridine.The dispersion was heated at 100 ◦C for 24 h and allowedto cool to room temperature. Hexamethyldisilazane (5.00 g,31.0 mmol), chlorotrimethylsilane (TMS-Cl, 3.4 g,31.0 mmol) were added, and after being heated at gentle refluxfor 24 h with vigorous stirring, the mixture solution was pouredinto 300 mL of acetone. The precipitate was filtered, washedwith water, and dried in vacuo to give 0.5 g of a pale tanpowder.
2.3. Preparation of graft polymers
4-dimethylaminopyridine (DMAP, 0.38 g, 3.1 mmol),trimethylsilylation of chitosan (0.5 g), and designated ratioof end-carboxyl PLGA were mixed with 50 mL dehydratepyridine and stirred at reflux for reaction (1 mL triethylamineas the catalyst) according to different reaction conditions(reaction time was 6 , 8 and 10 h, temperature was set 80,100 and 120 ◦C, chitosan/PLGA mass ratio 1:1, 1:2 and 1:3respectively). After the termination of reaction, the mixtureappears as a clear solution; then it was dropwise added to200 mL deionized water. The precipitate was filtered andwashed with water twice more, and dried at 80 ◦C overnight.
2.4. Desilylation to regenerate free hydroxyl and amino ofchitosan molecules
To 100 mL of a mixed solvent of methanol/acetic acid (3:1)was added 0.5 g of silylated chitosan/PLGA copolymers,and the mixture was stirred at room temperature for 24 h.The resulting undissolved substance was freeze-dried, and thesolid was dissolved in 30 mL of chloroform to wash out thePLGA residues. The precipitate was collected by filtration,washed with acetone and then with methanol, and dried to giveabout 0.35 g of chitosan/PLGA copolymers (named chitosan-g-PLGA) as a pale tan solid. The synthesis route is describedin figure 1. A combination of FTIR, 1H NMR and differentialscanning calorimetry (DSC) was then used to characterize thecopolymers.
2
Biomed. Mater. 5 (2010) 065016 D Xie et al
Figure 1. Schematic of copolymer preparation.
2.5. Characterization of graft copolymers
FT-IR spectra were obtained using EQUINOX-55 (FTIR,Bruker), wavelength range 4000–400 cm−1, resolution 2 cm−1.1H-NMR was conducted using a Varian Unity Invoa-500(Varian), with CF3COOD as the solvent, at 26 ◦C, 500 MHz.Thermal analysis (DSC) was under NETZSCH DSC-204,Al2O3 pot. Samples were heated from room temperature to120 ◦C with 10 ◦C min−1 heating rate and kept for 2 minand then cooled to 20 ◦C with cooling rate of 20 ◦C min−1
and kept for 5 min. Finally the temperature was increased by10 ◦C min−1 to 250 ◦C and the heating curve recorded. All thetests were in a N2 atmosphere at 20 mL min−1 gas flow rate.
Elemental analysis was conducted using anELEMENTAR Vario EL machine to determine the C,H, N content of samples. The hydrophilicity of the copolymerfilm was tested by measuring the contact angle (OpticalContact Angle Measurement System, OCA20), usingHFIP/DMF (95/5, m/m) as a solvent for the preparationof test films. A small solution droplet was carefully placedon the film surface and the contact angles determined usinga minimum of five measurements with less than 1◦ errorbetween each.
2.6. Electrospinning and characterization
Chitosan-g-PLGA solutions (5%, 7%, 9% and 12% m/m)were prepared by dissolving in HFIP/DMF (95/5, m/m).Chitosan was dissolved with TFA (trifluoroacetic acid) in8% (m/m). A 10% (m/m) PLGA (75/25) solution wasformed with HFIP only. The solution was transferred to a1 mL syringe attached to a blunt-tip stainless steel needle(20G). A steady flow of the solution from the needle spinneretwas achieved using a syringe pump (Aladdin 1000) at aflow rate from 0.6 to 1.5 mL h−1. A high-voltage powersupply (Brandenburg, Alphaseries III) was used to create anelectric field. The typical electrospinning distance betweenthe spinneret and the collector surface was between 10 and20 cm. Jet formation/stability was assisted by means ofan aluminum focusing ring placed 5 mm behind the tip ofthe needle. A voltage of 15 kV was used and electrospunfibers were collected on an aluminum foil coated rotated drumcollector.
2.7. In vitro degradation of chitosan-g-PLGA films
The chitosan-g-PLGA was dissolved in HFIP/DMF and caston a PTFE mold, with evaporation of the solvent formingthin films. The films were cut into disks of 10 mm diameterand 2 mm thickness using a biopsy punch. Samples weresterilized in 70% ethanol for 10 min and then washed threetimes with sterile PBS immediately prior to study. Sampleswere immersed in 2.0 mL of DMEM (no serum) solutionand incubated at 37 ◦C and under 5% CO2 atmosphere. TheDMEM medium was replaced the next day and then every4 days thereafter for up to 28 days. Samples were analyzed byweight at different time points to assess the degradation rate inDMEM medium. At specified time intervals, the films weretaken out of the media, washed with distilled water, dried andweighed. The degree of in vitro degradation was expressedas the percentage of the dried sample weight before and afterdegradation.
2.8. Culture of fibroblasts on scaffolds
Normal human fibroblasts were isolated and cultured asreported previously [17]. Passage 3 fibroblasts were used andwere cultured in DMEM supplemented with 10% (v/v) fetalcalf serum, 2 mM glutamine, 0.625 mg mL−1 amphotericin B,100 IU mL−1 penicillin and 100 mg mL−1 streptomycin.
Scaffolds of diameter 2 cm were sterilized in 70% ethanolfor 10 min and washed three times in PBS and twice in medium.Scaffolds were then placed in 12-well plates and stainless steelrings with an internal diameter of 1 cm were applied withpressure to one side of the electrospun scaffolds. Cells wereseeded in a volume of 300 μL of medium and incubated forshort periods of time (up to 6 h) with 1 × 106 cells per scaffoldfor cell attachment or with a seeding density of 3 × 105 cellsper scaffold for up to 7 days to assess proliferation. Culturemedium was replenished twice a week. The extent of cellularadhesion and cell proliferation was determined by an MTTassay of cell viability as described in Blackwood et al [19].
For SEM of cells on scaffolds specimens were fixedand dehydrated according the methods of Yang et al [20].The samples were gold sputtered and then examined usingPhilips/FEI XL-20 SEM at an accelerating voltage of10–15 kV.
3
Biomed. Mater. 5 (2010) 065016 D Xie et al
Figure 2. FT-IR spectra of products (O: pure chitosan; I: TMS-chitosan; II: TMS-chitosan/PLGA copolymer; III: chitosan/PLGAcopolymer).
2.9. Statistics
Data were collected from triplicate samples and wereexpressed as the mean ± standard deviation (SD). Statisticalanalysis was performed with Student’s t-test and significancewas determined at p < 0.05.
3. Results
3.1. Synthesis of the PLGA-g-chitosan polymer
FT-IR spectra of chitosan, trimethylsilyl chitosan (TMS-chitosan) and chitosan-g-PLGA from 400 to 4000 cm−1 areshown in figure 2. These show the peak between 3200 and3500 cm−1 related to symmetric and asymmetric O−H andN−H respectively in the chitosan molecules in figure 2(O)and the peaks 2899–2958 cm−1 related to C−H vibrationabsorption of the −CH, −CH2, −CH3 groups especially fortrimethylsilyl. Peaks between 1150 and 1000 cm−1 wereattributed to pyranose rings.
Two new strong absorption bands appeared at 1254 and841 cm−1 respectively in the trimethylsilylation of chitosanspectrum which were absent from the pure chitosan spectrum
and may be assigned to the Si−CH3 and Si−O stretching bandfor the trimethylsilyl modified chitosan in figure 2(I) and (II).The ester C=O band at 1750 cm−1 shifts to a higher frequency(1754 cm−1) after the formation of graft polymers as seenin figure 2(II) and (III). We also see that bands of 1254 and841 cm−1 disappeared after desilylation, suggesting that theregeneration of −OH, −NH2 comes from chitosan molecules.
Figure 3 illustrates the 1H-NMR spectrum of deprotectionof the copolymer of chitosan/PLGA. The main resonancepeaks can be briefly attributed to such groups as follows: thestrong signals at δ = 0.72–1.03 correspond to −CH3 of PLGA;δ = 2.12–2.23 for −CH of PLGA; δ = 4.05–5.91 for hydrogensignals of pyranose ring of chitosan unit appearing as complexmultiplet; δ = 8.28 for the −NH proton resonance absorptionof which is linked to C2 in chitosan molecule; δ = 11.50 forCF3COOD solvent.
In figure 4, DSC analysis indicates that PLGA has aglass transition temperature at 62 ◦C. The modified chitosanafter silylation has no obvious glass transition within thetest temperature range, unlike pure chitosan; the latter hasa Tg at 126.7 ◦C. This can be attributed to the disruption ofthe crystallization of chitosan after silylation reaction. The
4
Biomed. Mater. 5 (2010) 065016 D Xie et al
Figure 3. 1H-NMR spectra of copolymer.
Figure 4. DSC curves of product.
melting temperature of the copolymers at 152.8 ◦C is lowerthan that of PLGA or PLLA (not shown, about 180 ◦C).
3.2. Elemental analysis of PLGA-g-chitosan
Table 1 lists the elemental analysis results of the preparedcopolymer products with different reaction conditionsincluding temperature, time and mass ratio of reactants
according to the orthogonal array test. Elements and levels:t (6 h, 8 h, 10 h), T (80 ◦C, 100 ◦C, 120 ◦C) and mchitosan/mPLGA
(1:1, 1:2, 1:3).The C/N element contents ratio of the TMS-protected
chitosan molecules was used to calculate the substitute degree(DS) of TMS (trimethylsilyl) groups in the chitosan molecule(calculated DS = 2.4, according to the reference reported[16]). The number of pyranose units per graft PLGA chainwas calculated by the N element content and PLGA molecularweight (MV = 5000). Generally, the copolymer averagedone PLGA chain for each 18 pyranose unit interval in thechitosan backbone. By orthogonal test analysis, R values(Rtime = 0.045, Rtemperature = 0.015, Rmassratio = 0.093) werecalculated respectively and the mass ratio of reactants had themost important influence on the graft polymerization processcompared with the other two. Based on the yield of copolymer,the best reaction condition was under 120 ◦C for 6 h withmchitosan/mPLGA = 1:3. Higher graft density of PLGA chainsonto chitosan molecules was also found; the number of intervalpyranose units per graft PLGA chain reached 13.
3.3. Contact angles of chitosan-g-PLGA polymers
The contact angle values of different polymers were shownin table 2. With the introduction of hydrophilic chitosan,the chitosan-g-PLGA contact angle is reduced from 67.5◦ forendocarboxyl PLGA to 58◦ for the TMS-chitosan-g-PLGA
Table 1. Elemental analysis of CHN in copolymers.
Reaction conditionsElement contents
(%) The number of interval pyranose units per
t (h) T (◦C) mchitosan/mPLGA C H N graft PLGA chain (calculated)
Figure 5. SEM images of electrospun fibers using the chitosan-g-PLGA solution at a polymer concentration of: (A) 5 wt%; (B) 7 wt%; (C)9 wt%; (D) 12 wt%.
Table 2. Effect of adding chitosan on the contact angle ofPLGA-g-chitosan.
PolymerPoly(L-lactide)
End-carboxylPLGAfilm
TMS-protectionchitosan/PLGAcopolymerfilm∗
Deprotectionchitosan/PLGAcopolymer film
Contact 77 67.5 58a 47angle 53b 44(deg) 49c 37
∗ The mass ratio of TMS-protected chitosan to end-carboxyl PLGAis 1:3 (a), 1:2 (b) and 1:1 (c) respectively at 120 ◦C for 10 h.
copolymer film, while the chitosan-g-PLGA was only 47◦ afterthe TMS groups were removed due to the high hydrophobicityof these groups. The table also shows that the more chitosanadded to the copolymers, the lower the contact angle value andthe more hydrophilic the copolymers became.
3.4. Electrospinning of chitosan-g-PLGA copolymers fibers
Figure 5 shows SEM images of electrospun fibers (15 kVvoltage, 10 cm distance and 1 mL h−1 injection rate) ofvarious concentrations of chitosan-g-PLGA solutions. At theconcentration of 5 wt% the polymer solution formed verythin fibers (150–300 nm) with many bead-like structures of5–15 μm (figure 5(A)). At 7 wt% of polymer the fiberswere 300–500 nm with fewer beads (3–8 μm evident)(figure 5(B)). At 9 wt%, figure 5(C), the chitosan-g-PLGAfibers had a smooth and uniform surface. Fibers were 300–800 nm in diameter and free of any beads. At 12 wt%,figure 5(D), fibers were 1–3.5 μm in diameter. Clearly theconcentration of chitosan-g-PLGA solution played a majorrole in fiber diameter and fiber morphology.
3.5. In vitro degradation of chitosan-g-PLGA fibrousscaffolds
The SEM micrographs in figure 6 show the morphologies ofthe chitosan-g-PLGA (1:3) fibrous meshes after immersingthem into DMEM medium at 37 ◦C, 5% CO2 atmosphere for7 days. As shown in figures 6(A) and (B), during theincubation in medium, the fibers swelled markedly such thatthere were few gaps visible between fibers and the fibersappeared partially adhered to one another.
Figure 7 shows the weight loss versus time of the chitosan-g-PLGA copolymers in DMEM media over 3 weeks. Thelower the content of PLGA, the faster the rate of degradation.After 4 weeks, the weight loss of chitosan-g-PLGA (1:1) wasaround 33.6%, whereas chitosan-g-PLGA (1:2) was 22.1% andthat of chitosan-g-PLGA (1:3) was only 19.0%. Specially, thechitosan-g-PLGA (1:1) is very different from the other twoin earlier degradation; at the first week, the mass loss ratioreached 13.6%. This difference in the rate of degradation wasmaintained throughout 4 weeks.
3.6. Culture of cells on chitosan-g-PLGA electrospunscaffolds
The interaction of normal dermal fibroblasts with the scaffoldswas examined. Acute cell adhesion over 6 h was assessedusing an MTT ESTA assay, as shown in figure 8 (left).This shows that the cells adhered similarly and rapidly to allthree scaffolds with no significant difference between them.Figure 8 (right) shows the total cell viability assessed after 1,2, 3, 4 and 7 days of culture on the three scaffolds. Againthere was no significant difference between cell performanceson all three scaffolds, whether these were chitosan, PLGA orchitosan-g-PLGA.
Figure 9 shows the appearance of the cells on the chitosan-g-PLGA scaffold. Figure 9(A) at higher power shows the
6
Biomed. Mater. 5 (2010) 065016 D Xie et al
Figure 6. SEM images of electrospun fibrous chitosan-g-PLGA (1:3) scaffolds before and after 7 days incubation in DMEM media at 37 ◦Cin a CO2 incubator. (A) Surface of a dry scaffold; (B) surface of the scaffold after 7 days in media. Scale bar = 50 μm.
Figure 7. Rate of degradation of chitosan-g-PLGA scaffolds inDMEM medium. Results shown are means of n = 3 replicates.
appearance of cells on the scaffolds after 7 days culture.Fibroblasts had a normal morphology, long and flat and strap-like. Figure 9(D) shows the cross section of the same scaffoldafter culture with cells for 7 days by which time it is evidentthat the fibers are surrounded by cells and the cellular-producedmatrix. (Culture of these cells on PLGA scaffolds looked verysimilar as recently shown from our laboratory, Blackwoodet al [19].)
Figure 8. Interaction of fibroblasts with electrospun scaffolds. Left: the viability of cells cultured on scaffolds for 0.5–6 h was assessed.Right: cells were cultured on scaffolds for a period of 1–7 days prior to the assessment of total cellular viability. Results shown are means ±SD of n = 3 triplicates.
4. Discussion
The aim of this study was to investigate a novel methodfor the preparation of chitosan-g-PLGA copolymers forelectrospinning, evaluating these polymers in terms of theirappearance, biodegradation and ability to support fibroblastculture when electrospun into scaffolds.
PLA, PGA and PLGA have potential advantages as theirdegradation rate can be changed to match the rate of neo-tissue regeneration with sufficient mechanical strength to keepthe scaffolds until the new tissue forms. But they still havesome disadvantages, such as more hydrophobicity and the lackof cell-recognition signals. Their interactions with the hostenvironment still have much potential for improvement. Thejustification for introducing chitosan into PLGA is to achievea range of materials with the extra versatility of improvedbiocompatibility, hydrophilicity, mechanical stability andmodified biodegradability while introducing reactive sites(−OH, –NH2) to promote cell interaction or subsequentmodification of the scaffolds as desired.
There are many tissue engineering applications whereit would be desirable to preload a scaffold with peptides,proteins, cytokines or drugs either for use as a cell-freebiodegradable scaffold to implant into wounds and encouragecellular ingrowth, or to combine with cultured cells inthe laboratory as tissue engineered scaffolds. How to improve
7
Biomed. Mater. 5 (2010) 065016 D Xie et al
Figure 9. Appearance of chitosan-g-PLGA scaffolds after culture with fibroblasts. (A) Higher power of the surface of the scaffold afterculture with fibroblasts for 7 days. (B) Higher power cross section of the scaffold after culture with cells for 7 days. Scale bar as shown.
the biomaterial/cell interface property for eliciting the seedingcellular adhesion and maintaining differentiated phenotypicexpression has become one of the major challenges in the fieldof tissue engineering.
This paper describes a novel route for the chitosan-g-PLGA preparation. The characterization results using FT-IR,1H NMR and DSC confirmed a successful synthesis of thecopolymer, and in terms of elemental analysis showed that thiswas predictable and that one could control the graft reactionresults by varying the reaction conditions. From the dataobtained, it appeared that there were 13–25 pyranose unitsper PLGA chain achieved in this engraftment. With changesin the reaction conditions, the copolymer composition can becontrolled, varying the density of graft PLGA chains and themolecular weight of grafted PLGA chains, and the physico-chemical properties of the copolymers can be predicted. Themore chitosan that is introduced, the more hydrophilic thecopolymer and the more rapidly it degrades thus one can titratethe mechanical properties and rate of degradation desired byvarying the ratio of chitosan to PLGA.
Initial in vitro degradation data showed as expectedthat the higher the concentration of PLGA the lower thedegradation rate. In figure 7, copolymers with a higher PLGAcontent had lower degradation rates. This can be attributed toPLGA molecular chains’ higher hydrophobicity.
In addition the addition of PLGA acted to slow downthe rate of degradation of the scaffolds. Also while thedegradation products of PLGA can reduce the pH of thesurroundings with possible damage to cells and tissues, thisis not the case for chitosan which may help neutralize thisacidity and ameliorate some of these problems when usedin composite scaffolds. Possibly the neutralization effect ofincluding chitosan may be responsible for slowing down thedegradation rate of copolymers due to the avoidance of so-called self-catalyzation, preventing the copolymer structurefrom accelerated disintegration and rapid loss of mechanicalstrength. A delayed rate of degradation could also be useful forthe drug controlled release and tissue engineering scaffolds.However, the nature of the degradation in terms of breakdownproducts and pH changes requires further study, as it was notinvestigated directly in this study.
The number of papers on electrospinning of scaffoldsattests to the versatility and flexibility of this technique
for producing scaffolds for tissue engineering and otherapplications. As expected the percentage of chitosan presentaffected the thickness of the fibers and the nature of theelectrospun mat produced. As predicted the higher theconcentration of chitosan, the thicker the fibers—the keyvariable was the concentration of polymer solution. At lowpercentages of chitosan-g-PLGA, nanofibers formed but oftenwith many beads of polymer on them. Increasing polymersolutions to 9 wt% and higher led to much thicker microfibersof 1–3.5 μm diameter. Thus depending on the concentrationof the polymer one can achieve nanofibers or microfibers asdesired. The thickness of the electrospun mats that could beproduced was interesting in that these generally were very thin,but after immersion in media for 7 days there was extensiveswelling of the fibers due to, no doubt, the very hydrophilicnature of the chitosan.
Interestingly there were no significant differences in theinitial adhesion of cells to electrospun scaffolds made ofchitosan alone, PLGA alone or the copolymer. Similarly therewas no significant difference between the growths of cellsover 7 days on the three different scaffolds, so despite grosschanges in the fiber dimensions and the overall morphologyof the scaffolds all three provided an effective scaffold forfibroblast adhesion and growth.
Acknowledgments
This work was supported by Biomaterials and TissueEngineering Innovation Fund (the 3rd stage of 211 Projectin Jinan University) and China Scholarship Council and TheNational Natural Science Fund (31070862). The authors alsothank Dr Carla Pegoraro for help with SEM.
References
[1] Yang S et al 2001 The design of scaffolds for use in tissueengineering: part 1. Traditional factors Tissue Eng. 7 679–89
[2] You Y and Shen X 2002 Synthesis and degradability ofacetylation starch/DL-lactide graft copolymers Ion Exch.Absorpt. 18 58–63
[3] Cai Q et al 2003 Synthesis and characterization ofbiodegradable polylactide-grafted dextran and itsapplication as compatilizer Biomaterials 24 3555–62
[4] Ouchi T, Kontani T and Ohya Y 2003 Modification ofpolylactide upon physical properties by solution-cast blendsfrom polylactide and polylactide-grafted dextran Polymer44 3927–33
[5] Liu Y, Tian F and Hu K A 2004 Synthesis and characterizationof a brush-like copolymer of polylactide grafted ontochitosan Carbohydr. Res. 339 845–51
[6] Wu Y et al 2005 Synthesis and characterization of a novelamphiphilic chitosan-polylactide graft copolymerCarbohydr. Polym. 59 165–1
[7] Manisara P, Pitt S and Ratana R 2005 Preparation andcharacterization of hexanoyl chitosan/polylactide blendfilms Carbohydr. Polym. 60 343–50
[8] Pan T, Zhang G L and Ma J B 2006 Synthesis andcharacterization of β-cyclodextran/poly(DL-lactide) graftcopolymers Acta Polymerica Sin. 12 330–4
[9] Wu T-M and Wu C-Y 2006 Biodegradable poly(lacticacid)/chitosan-modified montmorillonite nanocomposites:preparation and characterization Polym. Degrad. Stab.91 2198–204
[10] Luckachan G E and Pillai C K S 2006 Chitosan/oligo L-lactidegraft copolymers: effect of hydrophobic side chains on thephysico-chemical properties and biodegradabilityCarbohydr. Polym. 64 254–66
[11] Duan B, Yuan X and Zhu Y 2006 A nanofibrous compositemembrane of PLGA–chitosan/PVA prepared byelectrospinning Eur. Polym. J. 42 2013–22
[12] Hao F and Dong C-M 2007 Synthesis and characterization ofphthaloyl-chitosan-g-poly(L-lactide) using an organiccatalyst Carbohydr. Polym. 70 258–64
[13] Jayakumar R and Tamura H 2008 Synthesis, characterizationand thermal properties of chitin-g-poly(ε-caprolactone)copolymers by using chitin gel Int. J. Biol. Macromol.43 32–6
[14] Liao Y Z and Xin M H 2007 Preparation and characterizationO-lauroyl chitosan/poly(L-lactic acid) blend membranes bysolution-casting approach Chin. Chem. Lett. 18 213–6
[15] Kohei T, Takeshi S and Hiromitsu Y 2008 Establishingchitosan coated PLGA nanosphere platform loaded withwide variety of nucleic acid by complexation with cationiccompound for gene delivery Int. J. Pharm. 354 210–6
[16] Keisuke K, Masaaki H, Satoru K, Hironori Y and Jin Y 2004Trimethylsilylation of chitosan and some properties of theproduct Carbohydr. Polym. 56 333–7
[17] Ralston D R, Layton C, Dalley A J, Boyce S G, Freedlander Eand MacNeil S 1999 The requirement for basementmembrane antigens in the production of humanepidermal/dermal composites in vitro Br. J. Dermatol.140 605–15
[18] Neves N M, Campos R, Pedro A, Cunha J, Macedo F and ReisR L 2007 Patterning of polymer nanofiber meshes byelectrospinning for biomedical applications Int. J.Nanomed. 2 433–8
[19] Blackwood K A et al 2008 Development of biodegradableelectrospun scaffolds for dermal replacement Biomaterials29 3091–104
[20] Yang F, Murugan R, Wang S and Ramakrishna S 2005Electrospinning of nano/micro scale poly(L-lactic acid)aligned fibers and their potential in neural tissueengineering Biomaterials 26 2603–10
[21] Christopherson G T, Song H and Mao H-Q 2009 Theinfluence of fiber diameter of electrospun substrates onneural stem cell differentiation and proliferationBiomaterials 30 556–64
[22] Woo K M et al 2007 Nanofibrous scaffolding promotesosteoblast differentiation and biomineralizationBiomaterials 28 335–43
[23] Gupta D, Venugopal J, Mitra S, Giri Dev V Rand Ramakrishna S 2009 Nanostructured biocompositesubstrates by electrospinning and electrospraying for themineralization of osteoblasts Biomaterials 30 2085–94
[24] Klinkhammer K, Seiler N, Grafahrend D, Gerardo-Nava J,Mey J, Brook G A, Moller M, Dalton P D and Klee D 2009Deposition of electrospun fibers on reactive substrates for invitro investigations Tissue Eng. C 15 77–85
[25] Ramakrishna S, Fujihara K, Teo W-E, Lim T C and Ma Z2005 Electrospinning Process, An Introduction toElectrospinning and Nanofibers (Singapore: WorldScientific) pp 96–8
[26] Yang X, Shah J and Wang H 2009 Nanofiber enabledlayer-by-layer approach toward three-dimensional tissueformation Tissue Eng. A 15 945–56
