the effect of plga doping of polycaprolactone films on the control of osteoblast adhesion and...

10
Biomaterials 27 (2006) 4409–4418 The effect of PLGA doping of polycaprolactone films on the control of osteoblast adhesion and proliferation in vitro Zheng Gui Tang, John Alan Hunt UK Centre for Tissue Engineering, Clinical Engineering, University of Liverpool, Liverpool, UK Received 18 January 2006; accepted 4 April 2006 Abstract Poly(e-caprolactone) (PCL) film was modified using specified amounts of poly(D,L-lactide-co-glycolide) (PLGA) to provide a means to control polymer degradation. The aim of the study was to determine the effects of doping PCL with PLGA on the materials degradation, morphology and cell adhesion, to determine the significant variables within the process that could provide further control of cell adhesion. PLGA-doped PCL films were aged in osteogenic medium at 37 1C for up to 28 days. The aged samples were analysed in terms of weight loss or weight gain, molecule deposition and surface morphology. Molecule deposition was determined using Fourier transform infrared spectroscopy in attenuated total reflectance mode (FTIR-ATR) and morphology was determined using scanning electron microscopy and interferometric microscopy. The loss of the PLGA doping during degradation enhanced the formation of nano- porous structures in the remaining PCL domains, which attracted the deposition of substances from the osteogenic medium, which favoured the attachment and growth of human osteoblasts. The growth of osteoblasts was influenced by the controlled release of acidic products through polymer blending. Two pairs of FTIR-ATR absorption bands at 1090 and 1110 cm 1 , and at 1180 and 1190 cm 1 were found to correlate to both PLGA and PCL, respectively. Changing the level of PLGA doping in PCL provided an approach to control the acidic products which can direct the growth of osteoblast cells. r 2006 Elsevier Ltd. All rights reserved. Keywords: PLGA-doped PCL films; Ageing; Cell culture medium; Nano-porous structures; Osteogenic surface smoothening 1. Introduction The targets for tissue-engineered therapies are the reconstruction of a network of matrix which has properties to deliver the appropriate signals to trigger the prolifera- tion and function of the correct cell types at the right moment. The matrix material could be best derived from biopolymers and molecules which are identical to the origin of the targeted tissue or organs. Scaffolds, fabricated from degradable polymers [1–3] have become a popular choice for matrices; however, they are of little biological resemblance to the natural host matrices. Pore size, porosity and interconnectivity are important characteristics in scaffolds. Conventional methodologies, such as salt leaching [4–7], freeze drying [8,9], and electrospinning [10–14] provide choices to produce scaf- folds with different external and internal structures. However, many scaffolds are far from ideal because of a lack of control and definition in their interconnectivity and usually only partial control or range limitations in pore size and porosity. Computer-aided solid free forming techniques [15–19] can overcome some practical limita- tions. These techniques have the added advantage of providing mould free fabrication with the ability to produce scaffolds with irregular profiles. However, some outstanding questions for the development of computer-aided solid free forming are: (1) How fine the line width of the polymer materials should be? (2) Is a pattern resolution down to the nano-structure optimum? and (3) What micro- and nano-structures should the scaffolds posses? ARTICLE IN PRESS www.elsevier.com/locate/biomaterials 0142-9612/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.biomaterials.2006.04.009 Corresponding author. Tel.: +44 1517065264; fax: +44 1517064915. E-mail address: [email protected] (J.A. Hunt).

Upload: zheng-gui-tang

Post on 26-Jun-2016

215 views

Category:

Documents


3 download

TRANSCRIPT

ARTICLE IN PRESS

0142-9612/$ - se

doi:10.1016/j.bi

�CorrespondE-mail addr

Biomaterials 27 (2006) 4409–4418

www.elsevier.com/locate/biomaterials

The effect of PLGA doping of polycaprolactone films on the control ofosteoblast adhesion and proliferation in vitro

Zheng Gui Tang, John Alan Hunt�

UK Centre for Tissue Engineering, Clinical Engineering, University of Liverpool, Liverpool, UK

Received 18 January 2006; accepted 4 April 2006

Abstract

Poly(e-caprolactone) (PCL) film was modified using specified amounts of poly(D,L-lactide-co-glycolide) (PLGA) to provide a means to

control polymer degradation. The aim of the study was to determine the effects of doping PCL with PLGA on the materials degradation,

morphology and cell adhesion, to determine the significant variables within the process that could provide further control of cell

adhesion. PLGA-doped PCL films were aged in osteogenic medium at 37 1C for up to 28 days. The aged samples were analysed in terms

of weight loss or weight gain, molecule deposition and surface morphology. Molecule deposition was determined using Fourier

transform infrared spectroscopy in attenuated total reflectance mode (FTIR-ATR) and morphology was determined using scanning

electron microscopy and interferometric microscopy. The loss of the PLGA doping during degradation enhanced the formation of nano-

porous structures in the remaining PCL domains, which attracted the deposition of substances from the osteogenic medium, which

favoured the attachment and growth of human osteoblasts. The growth of osteoblasts was influenced by the controlled release of acidic

products through polymer blending. Two pairs of FTIR-ATR absorption bands at 1090 and 1110 cm�1, and at 1180 and 1190 cm�1 were

found to correlate to both PLGA and PCL, respectively. Changing the level of PLGA doping in PCL provided an approach to control

the acidic products which can direct the growth of osteoblast cells.

r 2006 Elsevier Ltd. All rights reserved.

Keywords: PLGA-doped PCL films; Ageing; Cell culture medium; Nano-porous structures; Osteogenic surface smoothening

1. Introduction

The targets for tissue-engineered therapies are thereconstruction of a network of matrix which has propertiesto deliver the appropriate signals to trigger the prolifera-tion and function of the correct cell types at the rightmoment. The matrix material could be best derived frombiopolymers and molecules which are identical to the originof the targeted tissue or organs. Scaffolds, fabricated fromdegradable polymers [1–3] have become a popular choicefor matrices; however, they are of little biologicalresemblance to the natural host matrices.

Pore size, porosity and interconnectivity are importantcharacteristics in scaffolds. Conventional methodologies,

e front matter r 2006 Elsevier Ltd. All rights reserved.

omaterials.2006.04.009

ing author. Tel.: +441517065264; fax: +44 1517064915.

ess: [email protected] (J.A. Hunt).

such as salt leaching [4–7], freeze drying [8,9], andelectrospinning [10–14] provide choices to produce scaf-folds with different external and internal structures.However, many scaffolds are far from ideal becauseof a lack of control and definition in their interconnectivityand usually only partial control or range limitations in poresize and porosity. Computer-aided solid free formingtechniques [15–19] can overcome some practical limita-tions. These techniques have the added advantage ofproviding mould free fabrication with the ability toproduce scaffolds with irregular profiles. However,some outstanding questions for the development ofcomputer-aided solid free forming are: (1) How fine theline width of the polymer materials should be? (2) Is apattern resolution down to the nano-structure optimum?and (3) What micro- and nano-structures should thescaffolds posses?

ARTICLE IN PRESSZ.G. Tang, J.A. Hunt / Biomaterials 27 (2006) 4409–44184410

In combining tissue engineering with nano-science andnano-technology, Tang [20] introduced the revised tangiblecharacteristics for an ideal scaffold in which controllingpore size, porosity, interconnectivity, and micro- and nano-structures should be considered. Castner and Ratner [21] intheir review addressed the importance of a precisionsurface for tissue engineering applications. Surfaces thatinteract precisely with biological systems will be complex;multicomponent, multilayer and oriented. Given theinherent complexity of the molecular structures that makeup individual biomolecules, fabrication and characterisa-tion of such surfaces will stretch the limits of surfacescience.

The importance of micro- and nano-structures wasrevealed in the authors’ recent publications; Tang et al.[22] produced poly(e-caprolactone) (PCL) films withdifferent micro-structures and molecular aggregations.Identifying that characteristics of the surfaces haveintimate relationships with surface wettability and celladhesion. The poly(D,L-lactide-co-glycolide) (65:35) (PLGA65/35)-doped PCL materials [23] have the potential tofabricate unique surface morphologies such as regularpatterns of micro-dots on the surface. Such patternedsurfaces were relatively hydrophilic which improved adhe-sion and proliferation of human osteoblast cells.

It was an interesting phenomenon that human osteoblastcells grew well on the regular patterned surfaces (Fig. 1)[23], where the micro-dots were aggregates of PLGA 65/35,which is a more degradable polymer compared to PCL. Itis not clear whether the micro-patterns with PLGA 65/35attract osteoblasts, ageing the sample films in cellculture medium might provide further information aboutthe interaction between the patterned materials andosteoblasts.

In this study, the hypothesis that complex cell culturemedia would affect the degradation of PCL-doped withPLGA to alter the morphology and protein layer in a

Fig. 1. The adhesion of osteoblasts onto the films (n ¼ 3) made from

poly(e-caprolactone) (PCL) and poly(lactide-co-glycolide) (PLGA) (65:35)

blends, and the 90/10 and 80/20 with regular patterned surfaces (Tang

et al., Biomaterials 26 (2005) 6618–6624).

dynamic fashion with increasing ageing period and this inturn would affect osteoblast adhesion was tested. Osteo-genic medium was used as the ageing solution, thedegradation of PLGA-doped PCL films was analysed interms of changes in weight and surface morphologies.Correlation of the surface morphology and molecularinformation with cell attachment and growth provided afurther understanding of the interaction between osteo-blasts and PLGA-doped PCL films.

2. Materials and methods

2.1. Film preparation

PCL (MW 65k, Sigma-Aldrich) and PCL/PLGA (65:35, Sigma

Aldrich) chloroform solutions (6% w/v) were cast into petri dishes and

the solvent slowly evaporated. The films were removed and vacuum dried

for a further 48 h. The samples were grouped according to the surface that

had been exposed to air during casting. Circular discs of 10mm in

diameter were cut and washed in deionised water. They were then

dehydrated by progression through alcohol solutions (70, 80, 90, 95, 100,

100, 100v/v %) and vacuum dried for 48 h.

2.2. Ageing of films in osteogenic media

Circular discs of the films, 5 in each group, were weighed and incubated

in 20ml osteogenic media at 37 1C for 3, 7, 14 and 28 days. The osteogenic

media was Dulbecco’s modified minimum essential medium (DMEM)

(Invitrogen) supplemented with 10% foetal calf serum (FCS), 2mM

L-glutamine, 50U/ml penicillin, 50mg/ml streptomycin, 100mM L-ascorbic

acid 2 phosphate, 10 nM dexamethasone and 5mM b glycerophosphate.

The aged discs were taken out at the specific time points and washed in

deionised water. The dehydrated discs were weighed after 48 h of drying

under vacuum.

2.3. Attenuated total reflection Fourier transform infrared

spectroscopy (ATR-FTIR)

ATR-FTIR spectra of the sample were generated by a Thermo Nicolet

NexusTM FTIR (Cambridge, UK) controlled by OMNIC software

Version 6.1a. Briefly, the specimen was mounted onto a SMART

OMNI-Sampler connecting with FTIR Nexus. The machine was operated

under the experiment ATR-GE and spectra were obtained by accumulat-

ing 32 scans in the range 600–4000 cm�1 with a resolution of 4 cm�1.

2.4. Scanning electron microscopy (SEM) and interferometric

microscopy

The surface micro-structures and profiles were observed using Field

Emission Scanning Electron Microscopy (FE-SEM) (LEO 1550, Cam-

bridge, UK) and interferometric microscopy (WYKO NT3300 surface

profilometer). Briefly, the discs for FE-SEM were coated with chromium

(2min and about 50 nm thick) under 125mA. The coated sample was

placed in the vacuum chamber of the FE-SEM and viewed at a voltage of

5 kV. However, the discs for interferometric microscopy were used as

prepared without any coating and treatment. All the surfaces inspected

were the surfaces facing air in the petri-dish casting. The surfaces facing

the petri-dish glass side were not considered in this research.

3. Results

The percentage of weight loss of PLGA-doped PCL incell culture medium at different PLGA doping levels and

ARTICLE IN PRESSZ.G. Tang, J.A. Hunt / Biomaterials 27 (2006) 4409–4418 4411

different time points (Fig. 2), demonstrated that atdifferent PLGA doping levels, the PCL/PLGA (100/0)and PCL/PLGA (90/10) were the samples loosing the mostweight in the group in 7 days, up to 0.8% for PCL/PLGA(100/0) and 1.49% for PCL/PLGA (90/10), respectively;whereas the PCL/PLGA (80/20) and PCL/PLGA (70/30)were the samples loosing the most weight in the group in 28days, up to 1.09% for PCL/PLGA (80/20) and 0.99% forPCL/PLGA (70/30), respectively. Interestingly, the PCL/PLGA (70/30) gained weight in 14 days, up to 0.38%weight gain. Considering the different time points sepa-rately the highest weight loss occurred at 7 days with the

Fig. 2. Percentage of weight loss of PCL/PLGA (100/0, 90/10, 80/20 and 70/3

PLGA doping points and (b) at different time points.

9.80E+01

9.90E+01

1.00E+02

1.01E+02

1.02E+02

1.60E+03 1.62E+03 1.64E+03 1.66E+03 1.68E+03 1.70E+03

Wavenumber (cm-1)

Ref

lect

ance

(%)

9.80E+01

9.70E+01

9.90E+01

1.00E+02

1.01E+02

1.02E+02

1.60E+03 1.62E+03 1.64E+03 1.66E+03 1.68E+03 1.70E+03

Ref

lect

ance

(%)

0D 3D 7D 14D 28D

PCL/PLGA (100:0)

Wavenumber (cm-1)

PCL/PLGA (80:20)

0D 3D 7D 14D 28D

Fig. 3. FTIR spectra at the characteristic region (1600–1700 cm�1) f

PCL/PLGA (90/10) whereas the highest weight gainappeared at 14 days with the PCL/PLGA (70/30). ThePCL/PLGA (80/20) and PCL/PLGA (70/30) compara-tively lose more weight in the group in 28 days. The weightloss and gain of all samples appeared wave like with moreweight loss in 7 days, followed more weight gain in 14 days,and kept weight loss up to 28 days. The whole ageingprocess was a process of weight loss countered with weightgain. Doping PLGA into PCL changed the balance of theweight loss and gain.The FTIR spectra at the characteristic region (1600–

1700 cm�1) from PLGA-doped PCL films in osteogenic

0)-doped polycaprolactone in cell culture medium (n ¼ 5): (a) at different

9.80E+01

9.90E+01

1.00E+02

1.01E+02

1.02E+02

1.60E+03 1.62E+03 1.64E+03 1.66E+03 1.68E+03 1.70E+03

Wavenumber (cm-1)

Ref

lect

ance

(%)

9.80E+01

9.90E+01

1.00E+02

1.01E+02

1.02E+02

1.60E+03 1.62E+03 1.64E+03 1.66E+03 1.68E+03 1.70E+03

Ref

lect

ance

(%)

Wavenumber (cm-1)

PCL/PLGA (70:30)

0D 3D 7D 14D 28D

PCL/PLGA (90:10)

0D 3D 7D 14D 28D

rom PLGA-doped PCL films in osteogenic medium for 28 days.

ARTICLE IN PRESSZ.G. Tang, J.A. Hunt / Biomaterials 27 (2006) 4409–44184412

medium for 28 days (Fig. 3), indicated that the controlfilms at 0 day in osteogenic medium did not demonstratedistinctive bands in this region. However, the sample filmsin osteogenic medium registered a series of distinctivebands: 1690, 1685, 1675, 1670, 1662, 1653, 1647, 1636, 1628and 1624 cm�1. The prominent bands included 1685, 1670,1653, 1647 and 1636 cm�1. These bands associated withPCL/PLGA (100/0) and PCL/PLGA (90/10) showedenhanced reflections with increasing ageing time up to 14days, which had lost their prominence by 28 days ofageing. Interestingly, variations occurred with PCL/PLGA(80/20) and PCL/PLGA (70/30); the bands from thesefilms showed enhanced reflections with increased ageingtime up to 28 days, except the samples at 14 days ofageing.

The FTIR spectra at the characteristic region(1500–1550 cm�1) from PLGA-doped PCL films in osteo-genic medium for 28 days (Fig. 4), demonstrated that thecontrol samples (0 day) did not have any distinctive bandsin this region. However, the samples in the osteogenicmedium exhibited a series of distinctive bands: 1541, 1534,1528, 1522, 1517 and 1508 cm�1. The bands associated withPCL/PLGA (100/0) and PCL/PLGA (90/10) demonstratedenhanced reflections with increasing ageing time up to 14days, the prominence of which was lost at 28 days ofageing. This was comparable to the bands associatedwith the FTIR spectra at the characteristic region

Fig. 4. FTIR spectra at the characteristic region (1500–1550cm-1) fr

(1600–1700 cm�1) (Fig. 3); the bands associated withPCL/PLGA (80/20) and PCL/PLGA (70/30) showedenhanced reflections with ageing up to 28 days, exceptfor samples at 14 days of ageing. A slight band shift at1508 cm�1 to a lower wave number was found in samplesaged to 28 days.From the FTIR spectra at the characteristic region

(1200–1360 cm�1) from PLGA-doped PCL films in osteo-genic medium for 28 days (Fig. 5) a distinctive new bandwas found at 1340 cm�1 associated with the ageing processand time.The FTIR spectra at the characteristic region

(1000–1200 cm�1) from PLGA-doped PCL films in osteo-genic medium for 28 days (Fig. 6) revealed characteristicbands at 1187, 1170, 1130, 1110, 1090, 1068 and 1047 cm�1.Interesting changes associated with PLGA doping andmedium aging were correlated to bands at 1187 and 1170,1130, and 1110 and 1090 cm�1.The changes with bands at 1187 and 1170 cm�1 were

related to material composition such as the percentage ofdoping. The PCL/PLGA (100/0) demonstrated that theband at 1187 cm�1 was stronger than that at 1170 cm�1.Ageing in osteogenic medium changed the relative reflec-tions of both bands. Reverse relative reflections between1187 and 1170 cm�1 were found when samples were aged to7 days. Equivalent relative reflections between 1187 and1170 cm�1 were found when samples were aged to 14 and

om PLGA-doped PCL films in osteogenic medium for 28 days.

ARTICLE IN PRESS

Fig. 5. FTIR spectra at the characteristic region (1200–1360cm�1) from PLGA-doped PCL films in osteogenic medium for 28 days.

Fig. 6. FTIR spectra at the characteristic region (1000–1200 cm�1) from PLGA-doped PCL films in osteogenic medium for 28 days.

Z.G. Tang, J.A. Hunt / Biomaterials 27 (2006) 4409–4418 4413

ARTICLE IN PRESSZ.G. Tang, J.A. Hunt / Biomaterials 27 (2006) 4409–44184414

28 days. The PCL/PLGA (90/10) exhibited equivalentrelative reflections between 1187 and 1170 cm�1 for all thetime periods, whereas a stronger band at 1187 cm�1 wasfound in all PCL/PLGA (80/20). When 30% PLGA wasdoped in PCL the ageing process changed the relativereflections of both bands at 1187 and 1170 cm�1 in such away that a stronger band at 1187 cm�1 was found in thecontrol samples and equivalent bands of both in samplesaged to 3 and 7 days, and a stronger band at 1170 cm�1 insamples aged to 14 and 28 days.

The changes with the bands at 1130 cm�1 were seen inPCL/PLGA (80/20) and PCL/PLGA (70/30). This bandfrom PCL/PLGA (80/20) almost immediately disappearedin the ageing process. It could be found in PCL/PLGA (70/30) aged in 3 days and had completely disappeared at 7days.

Finally, the changes with bands at 1110 and 1090 cm�1

were also material composition related. Ageing did notintroduce changes with these bands in PCL/PLGA (100/0)(a stronger 1110 cm�1). The PCL/PLGA (90/10) demon-strated a stronger 1090 cm�1 except for samples aged (astronger 1110 cm�1) to 7 days. The stronger 1090 cm�1 inPCL/PLGA (80/20) was reversed completely in the ageingprocess. For PCL/PLGA (70/30), the stronger 1090 cm�1

changed into equivalent bands when aged to 7 and 14 daysand recovered the stronger 1090 cm�1 band when aged upto 28 days.

Fig. 7. Scanning electron micro-graphs of PLGA-doped polycaprolactone

aged in cell culture medium for 28 days (scale bar ¼ 5mm).

The scanning electron micro-graphs of PLGA-dopedPCL soaked in cell culture medium at 28 days (Fig. 7)demonstrated cracking in all control samples. Nano-poreswere observed in samples of PCL/PLGA (70/30). Samplesin cell culture medium for 3 days demonstrated nano-porous structures in the region of PCL except for PCL/PLGA (70/30), which presented a smooth surface. Com-paratively, the PCL/PLGA (80/20) generated a moreporous surface feature. The smooth surface versus nano-porous surface progressed further and at 7 days, samples ofPCL/PLGA (100/0) and PCL/PLGA (90/10) demonstratednano-porous surface features, whereas samples of PCL/PLGA (80/20) and PCL/PLGA (70/30) exhibited smoothsurface features.By 14 days almost all samples had become smoother

with a reduced topography, except the PCL/PLGA (100/0)which still demonstrated a rough surface. Samples aged for28 days showed unique smooth surface features, the PLGAdots in samples of PCL/PLGA (70/30) demonstratedhighly porous structures.For the surface profiles of PLGA-doped PCL after

incubation in cell culture medium for 28 days (Fig. 8),pure PCL was not selected because there were no PLGAdots in the material for comparison. The samples whichcontained no PLGA spherical dots were considered ascontrol samples. Samples aged for 3 days demonstratedmorphological differences which were indicative of theeffects of degradation. Morphological differences wereobserved in samples of PCL/PLGA (70/30). A ring ofmorphological changes appeared at the junction of thePLGA dot with the PCL substrate. These were notobserved in samples of PCL/PLGA (90/10) and PCL/PLGA (80/20). Samples aged for 7 days demonstratedmorphological differences in all PLGA-doped PCL.Small dots in PCL/PLGA (90/10) were either washedaway or completely eroded. A ring of morphologicalchanges were observed in samples of PCL/PLGA (80/20)similar to that observed in samples of PCL/PLGA (70/30)after 3 days. A substantial area of erosion of PLGAwas observed in samples of PCL/PLGA (70/30). Samplesaged to 14 days exhibited strong evidence of the collapseof the PLGA dots in all the sample groups. In samplesof PCL/PLGA (90/10), all the PLGA dots were degraded,leaving pits. In samples of PCL/PLGA (80/20), the sur-face of the PLGA dot was deformed due to the internaldegradation of PLGA. In samples of PCL/PLGA (70/30),the PLGA dot was almost completely degraded. Samplesaged for 28 days indicated that the erosion of PLGAmight extend into the PCL bulk material. The residualacids might proceed down to the PCL substrate andproduce holes in samples of PCL/PLGA (90/10). Thisacid-induced degradation even destroyed the junctionof PCL aggregation in samples of PCL/PLGA (80/20).The effect of the acid induced degradation was confir-med in samples of PCL/PLGA (70/30). The acid-induced erosion also proceeded into PCL close to thePLGA dot.

ARTICLE IN PRESS

Fig. 8. Surface profiles of PLGA-doped polycaprolactone aged in cell culture medium for 28 days (scale bar ¼ 10mm).

Z.G. Tang, J.A. Hunt / Biomaterials 27 (2006) 4409–4418 4415

4. Discussion

Ageing degradable polymer films in phosphate-bufferedsaline (PBS) is a conventional method used primarilybecause of its simplicity and therefore robust versatility.The weight loss of polymers increases with ageing, howeverthis might not be true when samples are placed in a morecomplex and complete media that contains proteins. In thisstudy, the weight loss of degradable films in cell culturemedium demonstrated more dynamic changes; indicatingweight loss at the beginning, followed by weight gain. Theweight loss of the samples did not follow the theory derivedfrom the degradation of polymers in PBS[24]. Increasing thePLGA doping of PCL films did not achieve more weightloss as the theory predicted because the accumulation ofacidic products could accelerate the degradation. Weightgain might be a unique phenomenon when complete cell

culture media is used. Cell culture media’s in which cells areactually cultured are a mixture of proteins, amino acids,and other nutrient molecules in addition to numerousessential salts. Weight gain might be correlated with thedeposition of substances from the cell culture media.Deposition of substances in osteogenic medium was

observed just by comparison of the weight loss and weightgain. From Fig. 2, all PLGA-doped PCL films did notcontinuously increase their weight loss during ageing up to14 days. Equilibrium favoured molecule deposition insteadof degradation and molecule release. Fingerprints of theproteins from FTIR spectra analysis proved the existenceof proteins on the surfaces of all the films.FTIR spectra were used to determine the secondary

structures of proteins. The commonly used band fordetermination of protein secondary structure is the amideI region (1620–1690 cm�1) [25]. The amide I region covers

ARTICLE IN PRESSZ.G. Tang, J.A. Hunt / Biomaterials 27 (2006) 4409–44184416

different C ¼ O stretching frequencies arising from differ-ent secondary structures (a-helix, b-sheet, turn andunordered structures) and has no major interfering bandsfrom other structures in proteins. Arakawa et al. [26]described the evidences of lysozyme in terms of FTIR. Theamide II (1534 cm�1) was assigned to the dried state ofproteins and that at 1542 cm�1 was assigned to thehydrated form of proteins. The double bond stretches inaromatic side chains of the proteins were demonstrated inthe characteristic bands in the region of 1500–1600 cm�1.In this region, if any characteristic bands appeared, itwould be an indication of molecule deposition. Accordingto Wang [25], the amide III band (1200–1330 cm�1) hasalso been used in determining the secondary structures ofproteins. It has better resolution for secondary structuresand no water interference (about 1640 cm�1).

The FTIR analysis confirmed the deposition of proteins(Figs. 3–5). No doping (PCL/PLGA (100/0)) or smalleramounts of doping such as PCL/PLGA (90/10) demon-strated the greatest deposition by 14 days of ageing asindicated by weight loss, weight gain and FTIR spectraanalysis. The higher doped films such as 20% and 30%PLGA also achieved the highest deposition by 14 days ofageing as indicated by weight loss and weight gain,however it was interesting that these more highly dopedfilms did not have the same potential to attract proteins at14 days based on the evidence from the FTIR spectraanalysis. This may be related to the degradation of PLGAmore than PCL. The deposition reduced and eventuallydisappeared with the degradation and dissolution ofPLGA. PCL/PLGA (100/0) and PCL/PLGA (90/10) weresimilar in terms of FTIR. The different sizes of PLGA dotsmight be crucial, however, this is an interesting discussionat this stage that requires further research to investigatethis explanation.

Direct information about the degradation of PCL wasalso obtained through ATR-FTIR analysis. According tothe work of Elzein et al. [27], the band at 1190 cm�1

referred to n(C–O): this vibrator is almost in the chain axisdirection (approximately 301 tilt) which provided a goodsignal in the thin film spectrum. This signal was noted inthe work of Tang et al. [22] as a strong indication ofpreferential aggregation of PCL molecules. The band at1180 cm�1 is close to the band at 1170 cm�1, which wasassigned as n’s(C–O–C): it indicates a preferential orienta-tion of this vibrator in the surface plane. In PLGA-dopedPCL, the band at 1170 cm�1 is more associated with PCLthan the band at 1190 cm�1.

Comparison of this pair of bands at 1187 cm�1 and1170 cm�1 was made by denoting the stronger andequivalent bands to indicate the changes of the conforma-tions of PCL molecules. The conformational changes ofPCL molecules in PCL/PLGA (100/0) demonstrated thatchanges in 0% PLGA-doped films occurred with ageingand the greatest weight loss was achieved at 7 days. Thedoping of 10% PLGA into PCL controlled the conforma-tional changes of PCL molecules, so weight loss might not

be related to the changes of PCL molecules. As was thecase for 10% PLGA doping, the 20% PLGA doping didnot affect the conformations of the PCL molecules.Comparing the conformational changes of PCL moleculesin the 30% PLGA doping with the 0% PLGA doping, thepattern of conformational changes of PCL molecules inPCL/PLGA (70/30) films was quite similar to that in PCL/PLGA (100/0) films. The significant differences were thecomplete reversal of the stronger band at 1187 and1170 cm�1. The PCL/PLGA (70/30) aged at 14 and 28days had a stronger band at 1170 cm�1, indicating the lossof PLGA.The use of FTIR spectra in the analysis of the

degradation of PLGA can be found in recent publications.The characteristic bands at 1090, 1131 and 1185 cm�1 forthe stretching vibration of the C–O–C were assigned toPLLA homopolymer [28], with supporting evidence pro-vided in the publication of C- atiker et al. [29]. Thecharacteristic peaks at 1267, 1185, 1129, 1092 and1044 cm�1 were correlated with the original polymer andcopolymers of PLLA and PLGA. It is clear that thepresence of the band at 1100 cm�1 correlated with PLGA[30] and the presence of the band at 1087 cm�1 with PLLA[31]. The direct evidence about the degradation of PLGAinvolves the analysis of the bands at 1130, 1100 and1090 cm�1. In PLGA-doped PCL, the band at 1090 cm�1 ismore associated with PLGA than the band at 1110 cm�1.A comparison of the pair of bands at 1110 and

1090 cm�1 was made by denoting the stronger andequivalent bands to indicate the changes of the conforma-tions of PLGA molecules. Theoretically there are noconformational changes of PLGA molecules in PCL/PLGA (100/0) films and the results in Fig. 6 confirmedthe theoretical deduction. The conformational changes ofPLGA molecules in PCL/PLGA (90/10) films weredetected only at day 7. The conformational changes ofPLGA were not a significant phenomenon in PCL/PLGA(90/10).The characteristic band at 1090 cm�1 was enhanced with

the increase of doping of PLGA. The switchover of1090–1110 cm�1 as a strong band was an indication ofthe loss of PLGA without disrupting the PCL. Thishappened in PCL/PLGA (80/20). The differences betweenPCL/PLGA (80/20) and PCL/PLGA (70/30) were the lossof PLGA in PCL/PLGA (70/30), that disrupted PCL by7 days, which was supported by the comparison of thecharacteristic bands at 1170 and 1187 cm�1.Overall the FTIR spectral analysis demonstrated that the

loss of PLGA in PLGA-doped PCL was prominent whenthe percentage of doping was more than 10%. The loss ofPLGA in PLGA-doped PCL exhibited a decrease of thebands at both 1190 and 1090 cm�1.It was surprising that the micro-structures on the

surfaces are not a key parameter of most degradationstudies. A visual examination of samples degraded for 5weeks in saline was carried by Li et al. [24]. Light micro-graphs of the top surface of the polymer blends have been

ARTICLE IN PRESSZ.G. Tang, J.A. Hunt / Biomaterials 27 (2006) 4409–4418 4417

published [32] but the images were not as clear as thedescription. In this study, the morphology changes wereobtained using scanning electron microscopy and aninterferometric microscopy at every time point. The resultsindicated that the micro-structures changed with ageingtime. Ageing introduced new nano-porous structures andthe nano-porous structures became increasingly less porousand eventually smoothed out. This might contribute to theweight gain and substrate deposition. At day 3, PCL/PLGA (70/30) films achieved the most weight loss due tothe nano-porous structures of the PLGA dots and theacidic acceleration of degradation. The number of osteo-blast cells on this sample was not due to the nano-porousstructures but the deposition of proteins as indicated inFig. 7.

The smoothening effect from protein deposition wasreported in recent investigations. Denis et al. [33] studiedthe protein adsorption on model surfaces with controllednano-topography and chemistry and found that collagenwas deposited onto hydrophilic surfaces, forming a stiffer,thinner and smoother adlayer without supramolecularassemblages. Brevig et al. [34] studied the effects ofalbumin-coated surfaces on leukocyte adhesion andactivation. Deposition introduced a much smootheralbumin layer on the relatively hydrophilic TCPS modelsurface compared to a coarse layer with albumin aggre-gates on the hydrophobic PS model surface. Carbone et al.[35] studied the response of cancer cells on gelatine-coatednano-structured TiO2 films. The gelatine coating wascharacterised by a close, uniform packaging of proteins,resulting in a rather smooth surface irrespective of thepreparation conditions.

Hydrophilic surfaces attracted protein deposition andformed a smoother adlayer of native proteins. Nano-structured surfaces enhanced the deposition of protein.Deposition of biological substances from the culturemedia in this study produced the smooth surface first inthe most hydrophilic PCL/PLGA (70/30) and then with theless hydrophilic PCL/PLGA (80/20), PCL/PLGA (90/10)and PCL/PLGA (100/0), respectively. The smooth surfaceof the PCL/PLGA (70/30) films was direct evidencetogether with surface amide I, II and III for proteinsfrom the FTIR spectral analysis. At day 7, PCL/PLGA(90/10) films achieved the greatest weight loss, whichwas also due to the formation of nano-porous structureson the surface at day 3 as indicated in Fig. 7. Similar tothe PCL/PLGA (70/30) films, PCL/PLGA (90/10) filmsattracted a substantial number of osteoblast cells atday 7 due to the deposition of proteins. The formationof the osteogenic surface in PCL/PLGA (80/20) couldbe explained in the same way as it was for PCL/PLGA(90/10).

Hydrophobic surfaces attracted protein deposition andformed a relatively coarse adlayer of supramolecularassemblages of proteins. This might be the reason whyPCL/PLGA (100/0) films attracted less osteoblast cells.The progressive loss of PLGA in PCL/PLGA (70/30) might

weaken the induction of PCL/PLGA (70/30) to the growthof osteoblast cells.The interferometric micro-graphs showed evidence of the

loss of PLGA micro-dots. Degradation at the boundary ofPLGA micro-dots was found in PCL/PLGA (70/30) filmsafter 3 days incubation. These phenomena occurred insamples of PCL/PLGA (90/10) and PCL/PLGA (80/20)films after 7 days incubation, when progressive loss ofPLGA was observed in PCL/PLGA (70/30) films. Theprogressive loss of PLGA might counter the growth ofosteoblast cells [23]. In samples aged for 14 days, a massiveloss of PLGA was found in PCL/PLGA (70/30) films. Theloss of PLGA in PCL/PLGA (90/10) and PCL/PLGA(80/20) films was not so dramatic at that time. Loss ofPLGA micro-dots appeared to be complete by 28 daysincubation. Acidic catalyst degradation was observed in thedomains of PCL in close proximity to PLGA micro-dots.Changing the amount of doping of PLGA on the surface

of PCL films controlled the degradation of PLGA andthe release of acidic products. Nano-porous structures wereintroduced in PCL domains, which then attracted thedeposition of biomolecules and proteins, resulting inthe adhesion and growth of osteoblast cells. The osteoblastcells favoured a smooth surface coated with substancesfrom the media that enhanced the attachment and growthof the cells.

5. Conclusions

The degradation of PLGA produced nano-porousstructures in the remaining PCL domains of PLGA-dopedPCL that favoured the deposition of proteins andbiomolecules. This resulted in a modified surface thatincreased the attachment and growth of osteoblast cells.The doping of PLGA in a controllable way onto PCLenhanced the attachment and growth of osteoblast cells upto a limit. It was possible to have too much PLGA doping,as was demonstrated with the 30% doping, which did notachieve the same results as the 10% and 20% dopings ofPLGA on PCL.

Acknowledgements

The authors gratefully acknowledge the support of thejoint UK Research Councils’ Interdisciplinary ResearchCollaboration in Tissue Engineering (BBSRC, EPSRC andMRC). The authors would also like to thank Ms. SandraFawcett for her assistance in scanning electron microscopyand Dr. Walter Perrie for his assistance in interferometricmicroscope.

References

[1] Engelberg I, Kohn J. Physico-mechanical properties of degradable

polymers used in medical applications: a comparative study.

Biomaterials 1991;12:292–304.

ARTICLE IN PRESSZ.G. Tang, J.A. Hunt / Biomaterials 27 (2006) 4409–44184418

[2] Middleton JC, Tipton AJ. Synthetic biodegradablepolymers as

orthopedic devices. Biomaterials 2000;21:2335–46.

[3] Ueda H, Tabata Y. Polyhydroxyalkanoante derivatives in current

clinical applications and trials. Adv Drug Deliv Rev 2003;55:501–18.

[4] Freyman TM, Yannas IV, Gibson LJ. Cellular materials as porous

scaffolds for tissue engineering. Progr Mater Sci 2001;46:273–82.

[5] Liao CJ, Chen CF, Chen JH, Chiang SF, Lin YJ, Chang KY.

Fabrication of porous biodegradable polymer scaffolds using a

solvent merging/particulate leaching method. J Biomed Mater Res

2002;59:676–81.

[6] Oh SH, Kang SG, Kim ES, Cho SH, Lee JH. Fabrication and

characterization of hydrophilic poly(lactic-co-glycolic acid)/poly(vinyl

alcohol) blend cell scaffolds by melt-molding particulate-leaching

method. Biomaterials 2003;24:4011–21.

[7] Peter SJ, Miller MJ, Yasko AW, Yaszemski MJ, Mikos AG. Polymer

concepts in tissue engineering. J Biomed Mater Res 1998;43:422–7.

[8] Lai HL, Abu’Khalil A, Craig DQM. The preparation and

characterisation of drug-loaded alginate and chitosan sponges. Int

J Pharm 2003;251:175–81.

[9] Mao JS, Zhao LG, Yin YJ, Yao KD. Structure and properties of

bilayer chitosan-gelatin scaffolds. Biomaterials 2003;24:1067–74.

[10] Frenot A, Chronakis IS. Polymer nanofibers assembled by electro-

spinning. Curr Opin Colloid Interface Sci 2003;8:64–75.

[11] Matthews JA, Wnek GE, Simpson DG, Bowling GL. Electrospinning

of collagen nanofibers. Biomacromolecules 2002;3:232–8.

[12] Luu YK, Kim K, Hsiao BS, Chu B, Hadjiargyrou M. Development

of nanostructured DNA delivery scaffold via electrospinning of

PLGA and PLA-PEG block copolymers. J Controlled Rel

2003;89:341–53.

[13] Xu CY, Inai R, Kotaki M, Ramakrishna S. Aligned biodegradable

nanofibrous structure: a potential scaffold for blood vessel engineer-

ing. Biomaterials 2004;25:877–86.

[14] Yoshimoto H, Shin YM, Terai H, Vacanti JP. A biodegradable

nanofiber scaffold by electrospinning and its potential for bone tissue

engineering. Biomaterials 2003;24:2077–82.

[15] Ang TH, Sultana FSA, Hutmacher DW, Wong YS, Fuh JYH,

Mo XM, et al. Fabrication of 3D chitosan-hydroxyapatite scaffolds

using a robotic dispensing system. Mater Sci Eng C—Biomimetic

Supramol Systems 2002;20:35–42.

[16] Hutmacher DW. Scaffold design and fabrication technologies

for engineering tissues—state of the art and future perspectives.

J Biomater Sci—Polym Ed 2001;12:107–24.

[17] Leong KF, Cheah CM, Chua CK. Solid freeform fabrication of

three-dimensional scaffolds for engineering replacement tissues and

organs. Biomaterials 2003;24:2363–78.

[18] Vozzi G, Flaim C, Ahluwalia A, Bhatia S. Fabrication of PLGA

scaffolds using soft lithography and microsyringe deposition.

Biomaterials 2003;24:2533–40.

[19] Yan Y, Xiong Z, Hu Y, Wang S, Zhang R, Zhang C. Layered

manufacturing of tissue engineering scaffolds via multi-nozzle

deposition. Mater Lett 2003;57:2623–8.

[20] Tang ZG. Invited Lecture on Nanoscience and Tissue Engineering, in

iNano School, Autumn School 2005, Fuglsøcentret in Denmark,

from 7 to 11 October 2005.

[21] Castner DG, Ratner BD. Biomedical surface science: foundations to

frontiers. Surf Sci 2002;500:28–60.

[22] Tang ZG, Black RA, Rice JM, Hunt JA, Rhodes NP, Williams DF.

Surface properties of solvent cast poly(e-caprolactone) films for tissue

engineering consideration. Biomaterials 2004;25:4741–8.

[23] Tang ZG, Callaghan JT, Hunt JA. The physical properties and

response of osteoblasts to solution cast films of PLGA doped

polycaprolactone. Biomaterials 2005;26:6618–24.

[24] Li SM, Garreau H, Vert M. Structure-property relationships in the

case of the degradation of massive aliphatic poly (a-hydroxy acids) in

aqueous media, Part 1: Poly(DL-lactic acid). J Mater Sci: Mater Med

1990;1:123–30.

[25] Wang W. Instability, stabilization, and formulation of liquid protein

pharmaceuticals. Int J Pharm 1999;185:129–88.

[26] Arakawa T, Prestrelski SJ, Kenney WC, Carpenter JF. Factrors

affecting short-term and long-term stabilities of proteins. Adv Drug

Deliv Rev 2001;46:307–26.

[27] Elzein T, Nasser-Eddine M, Delaite C, Bistatc S, Dumas P. FTIR

study of polycaprolactone chain organization at interfaces. J Colloid

Interface Sci 2004;273:381–7.

[28] Teng C, Yang K, Ji P, Yu M. Synthesis and characterization of

poly(L-lactic acid)–poly(e-caprolactone) multiblock copolymers by

melt polycondensation. J Polym Sci Part A: Polym Chem

2004;42:5045–53.

[29] C- atiker E, Gumus-derelioglu M, Guner A. Degradation of PLA,

PLGA homo- and copolymers in the presence of serum albumin: a

spectroscopic investigation. Polym Int 2000;49:728–34.

[30] Oliveira JM, Miyazaki T, Lopes MA, Ohtsuki C, Santos JD.

Bonelikes/PLGA hybrid materials for bone regeneration: prepara-

tion route and physicochemical characterisation. J Mater Sci: Mater

Med 2005;16:253–9.

[31] Deng C, Rong G, Tian H, Tang Z, Chen X, Jing X. Synthesis

and characterization of poly(ethylene glycol)-b-poly(L-lactide)-

b-poly(L-glutamic acid) triblock copolymer. Polymer 2005;46:

653–9.

[32] von Recum HA, Cleek RL, Eskin SG, Mikos AG. Degradation of

polydispersed poly(L-lactic acid) to modulate lactic acid release.

Biomaterials 1995;16:441–7.

[33] Denis FA, Hanarp P, Sutherland DS, Gold J, Mustin C, Rouxhet

PG, et al. Protein adsorption on model surfaces with controlled

nanotopography and chemistry. Langmuir 2002;18(3):819–28.

[34] Brevig T, Holst B, Ademovic Z, Rozlosnik N, Røhrmann JH, Larsen

NB, et al. The recognition of adsorbed and denatured proteins of

different topographies by b2 integrins and effects on leukocyte

adhesion and activation. Biomaterials 2005;26(16):3039–53.

[35] Carbone R, Marangi I, Zanardi A, Giorgetti L, Chierici E, Berlanda

G, et al. Biocompatibility of cluster-assembled nanostructured TiO2

with primary and cancer cells. Biomaterials 2006;27(17):3221–9.