polycaprolactone materials for tissue engineering

Accepted Manuscript Enzymatic degradation of graphene/polycaprolactone materials for tissue engineering Eoin Murray, Brianna C. Thompson, Sepidar Sayyar, Gordon G. Wallace PII:

S0141-3910(14)00389-9

DOI:

10.1016/j.polymdegradstab.2014.10.010

Reference:

PDST 7482

To appear in:

Polymer Degradation and Stability

Received Date: 29 August 2014 Accepted Date: 13 October 2014

Please cite this article as: Murray E, Thompson BC, Sayyar S, Wallace GG, Enzymatic degradation of graphene/polycaprolactone materials for tissue engineering, Polymer Degradation and Stability (2014), doi: 10.1016/j.polymdegradstab.2014.10.010. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Enzymatic Degradation of Graphene/Polycaprolactone materials for Tissue Engineering Eoin Murray1,3, Brianna C. Thompson2,3*, Sepidar Sayyar3, and Gordon G. Wallace3* 1

RI PT

Institute for Sports Research, Nanyang Technological University, Singapore School of Mechanical and Aerospace Engineering, Nanyang Technological University, Singapore 3 ARC Centre of Excellence for Electromaterials Science, Intelligent Polymer Research Institute, University of Wollongong, Australia * Corresponding author – [email protected], [email protected] 2

SC

Keywords: polycaprolactone, graphene, enzymatic degradation, lipase, composite, tissue engineering

M AN U

1. Abstract

2. Introduction

EP

TE D

Graphene/polycaprolactone composites have proven to be promising substrates for biodegradable tissue engineering scaffolds for electro-responsive tissue types. The degradation behaviour of these materials will be critical to any future application. To that end, the effect of chemically converted reduced graphene oxide (CCG) on the enzymatic degradation of graphene/polycaprolactone composites in phosphate buffered saline was examined. Two types of graphene/ polycaprolactone composites were tested; a simple blend and our previously developed covalently-linked composites. A number of graphene concentrations of each type were tested. Covalently linked graphene/polycaprolactone (cPClCCG) showed a consistent degradation profile maintaining the graphene:PCL ratio throughout the degradation process. However, the mixed blended sample (mixPCl-CCG) showed inconsistent graphene loss indicative of non-homogeneous dispersion throughout the polymer matrix. Increasing the graphene concentration up to 1 wt% did not change the rate of degradation but at higher concentrations degradation was slowed. The degradation products were also shown to be non-toxic to the proliferating cells.

AC C

Tissue engineering scaffolds should provide mechanical support and physicochemical cues for the growth of cells to replace tissue. [1] Materials for such scaffolds should possess appropriate mechanical properties, chemical and biological compatibility and ideally should degrade in an appropriate time frame. [1, 2] Due to its biocompatibility, processability, and consistent degradation profile, polycaprolactone (PCl) has frequently been used as a long-term degradable implantable scaffold material for tissue engineering. [3-6] The primary degradation pathway of pristine PCl is the hydrolysis of ester bonds leading to chain scission resulting in formation of shorter chains, oligomers and caproic acid. [7] This takes place in the amorphous bulk of the material and is dependent on its crystallinity and water transport into the material to enable hydrolysis. [4, 8-12] However, due to its semi-crystalline and hydrophobic nature, the rate of hydration and subsequent hydrolytic cleavage is low and hence the rate of degradation via this pathway is relatively slow (up to 4 years). [7, 10, 13] Enzymatic degradation, on

ACCEPTED MANUSCRIPT

the other hand, occurs at the surface of the polymer with lipases and esterases attaching to the polymer surface before hydrolyzing surface ester bonds in a much shorter time frame forming shorter chain polymers and oligomers. Degradation, while dependent on the enzyme used and the polymer composition and crystallinity, is generally complete after approximately 12 days and yields, among other fragments, ε-hydroxy caproic acid which, without clearance, can alter the pH of the host. [14-18]

M AN U

SC

RI PT

Tissue engineering of electro-reponsive cells has been shown to be improved by electrical stimulation. [19, 20] The introduction of conducting fillers to such a well-studied tissue engineering matrix as PCl has been shown to result in conducting biocompatible composites which can be used to proliferate cells under electrical stimulation. [21, 22] However, the filler used must be biocompatible and have an appropriately low percolation threshold for conductivity so as not to adversely affect the degradation profile of the polymer scaffold. Graphene has shown promise as a filler for this application, as recent reports suggest it can enhance cell proliferation [23] and can be cleared by renal excretion, phagocytosis and/or endocytosis. [24-27] In addition, the use of chemically converted reduced graphene oxide (CCG) has been reported to improve conductivities by orders of magnitude even at very low concentrations in a polymer matrix. [22] However, it is unknown what effects the addition of graphene would have on the degradation rates and products of a polymeric tissue engineering material.

TE D

Previous studies on the effect of nanofillers on the degradation of polycaprolactone (mostly nanoclays) and other polyesters have shown that addition of nanofillers can have a positive or negative effect on the degradation rate depending on the individual filler type and its effect in crystallisation and hydrophobicity. [28, 29] It has been shown that graphene increases the number of crystallization nucleation sites and thus changes the size and number of the spherulite crystalline regions in polycaprolactone. [22] Both degradation pathways can be affected by the crystallinity as degradation occurs in the amorphous regions first. In addition, reduced graphene oxide is hydrophobic in nature affecting the overall hydrophobicity of the composite. It is also unclear what effect graphene sheets would have on the toxicity of the degradation products.

AC C

EP

In this work we compare the enzymatic degradation of pristine polycaprolactone and graphene/polycaprolactone composites produced using two methods introduced in our previous papers [6, 21, 22]; a simple mixing method to produce a blended composite and a chemical route to produce graphene sheets covalently linked to the polymer matrix. The physical results of this degradation and the toxicity of the byproducts are examined. 3. Materials and Methods

3.1 Materials - N,N-dimethylformamide (DMF), methanol N,N'-dicyclohexylcarbodiimide (DCC), 4dimethylaminopyridine (DMAP), Dulbecco’s modified Eagle's medium (DMEM), methanol, dichloromethane, polycaprolactone (MW 80,000) and triethylamine were sourced from Sigma-Aldrich and used as received. Graphite powder was obtained from Bay Carbon. Milli-Q water (DI water) with a resistivity of 18.2 mΩ/cm was used in all preparations. 3.2 Synthesis – All composites were produced using polycaprolactone of average molecular weight of 80,000. Chemically converted reduced graphene oxide (CCG) was produced via the chemical oxidation of

ACCEPTED MANUSCRIPT

RI PT

graphite using a modified Hummer’s method and subsequent reduction and dispersion that resulted in a 0.5mg/ml graphene dispersion in DMF as previously described. [30] Blended graphene/polycaprolactone (mixPCl-CCG) composites were prepared by mixing the required amount of polycaprolactone in the CCG dispersion at 70 °C and precipitation into cold methanol. Covalently linked graphene/polycaprolactone composites (cPCl-CCG) were prepared by adding the required quantities of DCC and DMAP to a solution of polycaprolactone in a CCG dispersion at 70 °C and subsequent precipitation into cold methanol. Both synthetic procedures and the analysis of both systems are described in detail in our previous publication. [21, 22, 30]

M AN U

SC

3.3 Degradation – Lipase from Pseudomonas sp. (Type XIII) was purchased from Sigma Aldrich with a specific activity of 15 U/mg. The enzyme was used as received, with no further purification. The enzyme was used at 8 U/mL (0.53 mg/mL) in 0.1M phosphate buffered saline (PBS) for all experiments, based on the method presented by He et al. [31]. Samples were hot-pressed at 100 °C to yield an approximate thickness of 0.1 mm, and then discs of samples were prepared using a 6 mm diameter hole punch. Discs of the materials were weighed and placed into eppendorf tubes, and 1 mL of the 8 U/mL enzyme solution was placed into each tube. Samples were immediately transferred to a 37 °C waterbath and incubated for between 3 to 96 hours, with enzyme replacement performed every 24 hours. After the incubation time, each sample was removed, rinsed in water, dried overnight and then weighed. After weighing, the properties of polymer discs were determined post-degradation, and enzyme solution/breakdown product solutions were also retained for toxicity testing.

AC C

EP

TE D

3.4 Analysis – All characterization was performed on the discs used for degradation after washing with DI water and air drying at room temperature unless otherwise stated. Raman spectra were recorded on a Jobin Yvon Horiba HR800 Raman microscope using a 632 nm laser line and a 300-line grating. Scanning electron microscopy (SEM) images were collected with a field-emission SEM instrument (JEOL JSM7500FA). Samples were sputter-coated (EDWARDS Auto 306) with a thin layer of platinum (≈15 nm thickness). Thermal gravimetric analysis (TGA) was performed in using TGA Q500, TA Instruments with a heating rate of 10 °C under a nitrogen atmosphere. Differential scanning calorimetric (DSC) analysis was performed on a DSC Q100, TA Instruments. 5 to 8 mg of the sample was pre-sealed into an aluminum pan and first heated to above the melting temperature (Tm) of the polymer (100 °C), then cooled to 0 °C at 10 °C/min, the temperature increased to above 100 °C at 10 °C/min. All sonication was done using a Branson Digital Sonicator (S450D, 500 W, 40 % amplitude). Contact angle measurements were by application of a 1 μl water droplet onto the disc surfaces using the sessile drop technique. Images of the droplets were captured DataPhysics OCA20 Goniometer and analysed using SCA21 software (DataPhysics). Three or more measurements were taken on each surface, and the mean and standard deviation were calculated. 3.5 Toxicity of degradation products – The degradation media from lipase-catalysed PCl or PCl-CCG degradation (consisting of soluble and insoluble degradation products and lipase in PBS) and PBS/lipase controls (containing no degradation products) were stored at -20 °C prior heat deactivation of the enzyme at 65 °C for 2 hours. After this, the solutions were filter-sterilised using a 0.2 µm filter (retaining only soluble degradation products), and then added to culture media at 10 %(v/v). Inhibition of growth of L-929 cells was used to determine if the breakdown products from PCl, cPCL-CCG or mixPCl-CCG were

ACCEPTED MANUSCRIPT

RI PT

toxic or caused changes to the fibroblast cell line metabolism affecting growth. L-929 mouse fibroblast cells (NCTC clone 929 [L cell, L-929, derivative of Strain L], ATCC® CCL-1™) were plated at 1 x 103 cells/mL in 0.1 mL of unmodified media (DMEM + 10 % FBS + 1 % penicillin/streptomycin) and cells were incubated for 24 hours at 37 °C in humidified 5 % CO2. After 24 hours, the media was completely removed and 0.1 mL media containing 10%(v/v) of the degradation products or PBS/enzyme control, and the cells were returned to the incubator. After 48 hours the cell number in each well was measured using a standard Pico Green assay (Life Technologies), and the cell numbers were compared to the PBS/lipase control. 4. Results and Discussion

4.1 Mass loss

TE D

M AN U

SC

Graphene nanosheets have been shown to function as multiple crystallization nucleation centres and change the underlying crystal structure when used as a filler in polymer matrices. For this reason the incorporation of graphene causes large increases in the mechanical strength and modulus of polycaprolactone at very low levels. We have recently shown that composites produced with dispersed chemically reduced graphene are also biocompatible and conduct electricity at very low percolation thresholds (< 0.1 wt% graphene) [22]. In addition, the covalent attachment of the graphene nanosheets to the polymer matrix results in a far better dispersed composite with much improved mechanical and electrical properties without compromising the biocompatibility or processability of the polymer. Thus, additive fabrication techniques were able to be used to form made to measure three-dimensional scaffolds ostensibly for tissue engineering using these materials [21]. However, the inherent change in the crystal structure on addition of graphene which is responsible for the improved mechanical properties may affect the degradation profile of these materials. Enzymatic degradation using lipases affords a practical, accelerated overview of the degradation process in these materials.

AC C

EP

The effect on graphene on the previously reported enzyme-catalysed degradation of PCl [31] was examined over 96 hours for PCl and composite materials (Figure 1). The absolute mass loss was found to be a more reliable value for comparisons between discs, as small variations in the thickness affected the weight of the disc without greatly changing the surface area. As the enzyme-mediated degradation was a surface phenomenon, the mass loss was limited more by the surface than the mass, and so the % mass loss values varied widely from sample to sample, while the absolute mass loss showed was much more consistent. In order to rule out enzyme-limitation of degradation, the concentration of the enzyme was doubled and the rate of degradation checked at 24 hours. As shown in Figure S1, increasing the concentration of lipase did not increase degradation at 24 hours. For PCl alone, and cPCl-CCG0.1%, cPCl-CCG1% and mixPCl-CCG5% composites, most mass was lost over the first 24 hours, with absolute losses of 1.2-1.5 g per disc, or 50-60% of the starting weights (Figure 1a). cPCl-CCG5% showed a much slower and more consistent rate of degradation over the 96 hours, demonstrating a different degradation profile to both composites with lower graphene loadings (PCl alone, and cPCl-CCG0.1%, cPCl-CCG1%) and composite with a comparable graphene loading made by simple mixing and not covalent modification (mixPCl-CCG5%). Observation of the discs showed that they

ACCEPTED MANUSCRIPT

b)

TE D

M AN U

SC

a)

RI PT

degraded from the outer edges in, with the thickness also reducing over time (Figure 1b). At later stages of degradation, the degradation media was observed to have cloudy white to grey precipitate (assumed to be small polymer/graphene particulate matter eroding from the surfaces), except for the mixPClCCG5%, which was observed to contain dark black material (see Figure S2). The precipitates, as well as images of the discs in Figure 1b, suggests that the mixPCl-CCG5% had shed more graphene, or larger particles of composite had been removed from the disc, compared with the covalently assembled composites.

AC C

EP

Figure 1 – Lipase-catalysed degradation of PCl-CCG composites. a) Absolute mass loss of the polymer composite discs over 96 hours for each sample, with each point representing the average of duplicate measurements, and the error bars showing the standard deviation. b) Representative images of 6 mm diameter discs of composite materials before degradation, and after various degradation times.

4.2 Morphology, wettability and resistivity (surface properties) Electron microscopy shows the surface morphology of the pristine, undegraded materials (Figure 2). The addition of graphene increases the apparent crystallinity with graphene nanosheet edges visible underneath a polymer layer. On degradation the amorphous regions of the polymer are degraded first as they offer easier access for hydrolysis and the surfaces of all materials show deep pitting and porosity. In the composites, graphene nanosheets appear to be exposed following the removal of the covering amorphous polymer.

RI PT

ACCEPTED MANUSCRIPT

SC

Figure 2 – Scanning Electron Micrographs of pristine and 24 hour degraded composite samples. The scale bar represents 3 µm.

M AN U

Contact angle increases for all materials as degradation progresses, indicating an increase in hydrophobicity, despite the increase in surface roughness demonstrated in Figure 2. The pristine polymer increases from 90 ° to over 130 ° over 96 hours of degradation due to the removal of the more hydrophilic amorphous regions and shorter chains (Figure 3a). Over the same time frame the contact angle of the graphene composites increase to over 140 °, as hydrophobic graphene sheets are exposed following the removal of the covering amorphous polymer.

b)

AC C

EP

a)

TE D

In addition, these exposed and semi-exposed graphene sheets result in a decrease in the surface resistivity of the composite materials to a value closer to their bulk conductivity (Figure 3b), with all materials exhibiting an initial decrease in resistivity of over two orders of magnitude by 6 hours, followed by a plateau on further sample degradation. The mixPCl-CCG5% showed higher resistivity than cPCl-CCG5%, and the resistivity of the mixed composite increased slightly with degradation time after 6 hours.

Figure 3 – a) Contact angle of water with PCl-CCG composites before degradation, and after 6 hours and 96 hours of enzymatic degradation. b) shows the change in resistivity of mixPCl-CCG5%, cPCl-CCG1%, and cPCl-CCG5% composites with degradation time. 4.3 Crystallinity and thermal properties (bulk properties)

ACCEPTED MANUSCRIPT

TE D

M AN U

SC

RI PT

Differential scanning calorimetry has shown that the addition of graphene does not affect the melt temperature but results in very large increases in the crystallization temperature from 19 °C in pristine PCl to 34.5 °C in the cPCl-CCG5% composite (Figure 4a and Table 1). This is due to graphene nanosheets functioning as multiple crystallisation nucleation centres changing the spherulite size and dispersion throughout the material. Hydrolytic degradation of PCl is dependent on the crystallinity with degradation occurring preferentially in the amorphous regions. Degradation over 48 hours results in no change in the melt temperature for PCl alone or composites, which remains between 55 and 60 °C for the pristine polymer and the graphene/polymer composites (Table 1). However, after 16 hours of degradation the crystallization temperature of polycaprolactone increases from 18.6 °C pre-degradation to 30.7 °C (Figure 4a) and does not increase further up to 48 hours. Over the same time frame the crystallization temperature for graphene composites change only 4-6 °C (Figure 4b). This is indicative of the difference in the relative crystallinity and the change in crystallinity following the preferential degradation of the amorphous part of the polymer and the composites.

EP

Figure 4 - Differential scanning calorimetry of a) PCl and b) cPCl-CCG5% before (solid) and after 16 (--) and 48 (··) hours degradation.

AC C

Table 1 - Average melt and crystallization temperatures of PCl and PCl-CCG composites before and after 48 hours of degradation

PCl cPCl-CCG 0.1% cPCl-CCG 1% cPCl-CCG 5% mixPCl-CCG

Melt Temperature (°C) Pre-degradation Post-Degradation 57.2 55.5 55.6 55.9 57.1 56.4 59.6 56.4 57.1 55.1

Crystallisation Temperature (°C) Pre-degradation Post-Degradation 18.6 30.7 31.9 35.7 33.8 39.5 34.5 39.4 34.8 39.8

ACCEPTED MANUSCRIPT

TE D

M AN U

SC

RI PT

Thermal gravimetric analysis (TGA) was performed on the remaining discs of pure PCl and the PCl-CCG composites at different time points along the degradation process. All samples show a sharp monotonic weight loss between 340 and 420 °C which can be assigned to the decomposition of the original polymer chain length (Figure 5), indicating that any shorter chains are removed from the degraded material. In addition, as CCG weight losses are minimal in this temperature range, the residual weight after full decomposition of the polymer can be assigned to the graphene content. The graphene percentage calculated from TGA analysis of the composites prepared by covalent attachment method (cPCl-CCG5%) are very consistent around 5 % across the degradation time indicating a consistent rate of graphene removal. However, the graphene percentage in the blended mixtures (mixPCl-CCG5%) increases substantially from 5 % to almost 19 % indicative of a non-consistent rate of graphene removal and hence a poorly dispersed material. This is supported by the observation of larger, darker particulate material in the degradation media (Figure S2), and slight increases in resistivity over degradation time (Figure 3b) for mixPCl-CCG5% relative to cPCl-CCG5%.

EP

Figure 5 - Thermal gravimetric analysis of a) mixPCl-CCG5% and b) cPCl-CCG5% after 0, 6, 24 and 72 hours of enzymatic degradation. The residual weight after full decomposition of the polymer can be attributed to graphene content.

AC C

4.4 Toxicity of degradation products

The toxicity of the soluble products in the degradation media were tested after heat-inactivation of the lipase protein. Adherent L-929 fibroblast cells were exposed to media with addition of 10 % (v/v) degradation products for 48 hours (with a control exposed to 10% (v/v) PBS/lipase inactivated in the same manner) before the cell number was measured by Pico Green Assay and the number of cells growing well exposed to degradation products compared to the number of cells in the control wells. For PCl alone, the 6, 24 and 72 hour degradation media decreased cell number relative to the PBS-treated control, with longer degradation time leading to a greater effect on the cells (down to 65 % of the cell density for the 96 hour degradation media). This effect may be related to the decrease of pH associated with formation of acidic PCl degradation products, however pH of the media was measured and minimally adjusted to pH 7.4 with 0.1 M sodium hydroxide before addition to cell culture.

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Samples containing graphene showed less toxicity than the PCl alone, with covalently formed composites showing a slight trend for more graphene leading to higher final cell number, and a nonlinear relationship between degradation time and decrease in cell number. All covalently-modified composite degradation products showed an increase in cell number of the 24 hour degradation products, with up to a 60% increase over the control cells. An increase in cell number is likely due to increased nutrients or growth factors in the media to increase metabolism and growth rate of the cells, indicating that breakdown products may be acting as a nutrient source for the fibroblasts. The mechanism by which graphene may influence degradation product formation or media conditions to achieve this is unknown, however graphene oxide has been shown to enhance cell growth in some instances [32-34], where the graphene isn’t in an aggregated state. This may also explain the lack of increased cell density for the mixPCl-CCG5% sample.

Figure 6 – Toxicity of degradation products at various degradation times measured by comparing fibroblast cell number (determined by Pico Green Assay) after 48 hours exposure to 10 % (v/v) degradation media and comparison to a PBS/lipase control. Each point represents the mean of 4 replicates ± S.E. of the mean.

4. Conclusions Controllable degradation to non-toxic products is a key property of polymer composites for tissue engineering. To this end, accelerated degradation profiles of polycaprolactone and PCl/graphene

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

composites, produced either by mixing of components, or covalent attachment of the graphene to the polymer, were obtained using lipase enzymes and a number of relevant physical parameters assessed throughout the degradation process. The rate of degradation of covalently-linked composites was not significantly affected with graphene loadings below 5 wt%, and no changes in the degradation rate were observed for higher loadings in the mixed sample. Analysis of the surface properties of the composites showed that while surface roughness increased after enzymatic degradation, the hydrophobicity and conductivity both increased. The melt temperature of the composites was unaffected by incorporation of graphene or degradation, however the crystallisation temperature increased with both graphene inclusion (mixed or covalently attached) and degradation, suggesting that both treatments changed the crystallinity of the materials. Overall, physical characterisation suggested that the covalent route to composite formation led to a more homogenous dispersion of graphene throughout the material (in agreement with our earlier work [22]), which led to more consistent degradation of the material – the graphene content stayed at 5% for the covalent material, while it increased from 5% up to 19% for the mixed material demonstrating inconsistent graphene removal. The degradation products of the composite materials were found to exhibit less inhibition to cell metabolism and proliferation than the degradation products of PCl alone, with higher graphene loadings actually increasing cell number over 48 hours compared to controls. The controllable non-toxic degradation in conjunction with the notable physical and electronic properties confirm that covalently-linked polycaprolactone graphene composites are ideal materials for the development of scaffolds for tissue engineering of electro-responsive cells.

TE D

5. Acknowledgements

AC C

6. References

EP

The authors acknowledge use of facilities within the UOW Electron Microscopy Centre. The authors gratefully acknowledge the support of the Australian Research Council through the Superscience Fellowships and provision of funding through the Australian Laureate Fellowship. We also gratefully acknowledge the support of the ARC Centre of Excellence for Electromaterials Science (ACES) for provision of equipment used throughout this work.

[1] Chen GP, Ushida T, Tateishi T. Scaffold design for tissue engineering. Macromolecular Bioscience. 2002;2:67-77. [2] Sabir MI, Xu X, Li L. A review on biodegradable polymeric materials for bone tissue engineering applications. Journal of Materials Science. 2009;44:5713-24. [3] Hutmacher DW. Scaffolds in tissue engineering bone and cartilage. Biomaterials. 2000;21:2529-43. [4] Lam CXF, Hutmacher DW, Schantz J-T, Woodruff MA, Teoh SH. Evaluation of polycaprolactone scaffold degradation for 6 months in vitro and in vivo. Journal of Biomedical Materials Research Part A. 2009;90A:906-19. [5] Lam CXF, Savalani MM, Teoh S-H, Hutmacher DW. Dynamics of in vitro polymer degradation of polycaprolactone-based scaffolds: accelerated versus simulated physiological conditions. Biomedical Materials. 2008;3.

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

[6] Cornock R, Beirne S, Wallace GG, Ieee. Development of a Coaxial Melt Extrusion Printing Process for Specialised Composite Bioscaffold Fabrication. 2013 Ieee/Asme International Conference on Advanced Intelligent Mechatronics2013. p. 973-8. [7] Bosworth LA, Downes S. Physicochemical characterisation of degrading polycaprolactone scaffolds. Polymer Degradation and Stability. 2010;95:2269-76. [8] Chen DR, Bei JZ, Wang SG. Study on biodegradation behavior of polycaprolactone microparticles. Acta Polymerica Sinica. 1999:620-2. [9] Chen DR, Bei JZ, Wang SG. Polycaprolactone microparticles and their biodegradation. Polymer Degradation and Stability. 2000;67:455-9. [10] Iwamoto A, Tokiwa Y. EFFECTS OF THE BLEND RATIO AND MELT VISCOSITY ON BIODEGRADABILITY OF PLASTICS CONTAINING AN ALIPHATIC POLYESTER. Kobunshi Ronbunshu. 1993;50:789-91. [11] Lam CXF, Teoh SH, Hutmacher DW. Comparison of the degradation of polycaprolactone and polycaprolactone-(beta-tricalcium phosphate) scaffolds in alkaline medium. Polymer International. 2007;56:718-28. [12] Tokiwa Y, Calabia BP. Biodegradability and biodegradation of polyesters. Journal of Polymers and the Environment. 2007;15:259-67. [13] Ju D, Han L, Li F, Chen S, Dong L. Crystallization, Mechanical Properties, and Enzymatic Degradation of Biodegradable Poly(epsilon-caprolactone) Composites with Poly(lactic acid) Fibers. Polymer Composites. 2013;34:1745-52. [14] Cook WJ, Cameron JA, Bell JP, Huang SJ. SCANNING ELECTRON-MICROSCOPIC VISUALIZATION OF BIODEGRADATION OF POLYCAPROLACTONES BY FUNGI. Journal of Polymer Science Part C-Polymer Letters. 1981;19:159-65. [15] Hou Y, Chen J, Sun P, Gan Z, Zhang G. In situ investigations on enzymatic degradation of poly(epsilon-caprolactone). Polymer. 2007;48:6348-53. [16] Iwamoto A, Tokiwa Y. ENZYMATIC DEGRADATION OF PLASTICS CONTAINING POLYCARPROLACTONE. Polymer Degradation and Stability. 1994;45:205-13. [17] Li F, Yu D, Lin X, Liu D, Xia H, Chen S. Biodegradation of poly(epsilon-caprolactone) (PCL) by a new Penicillium oxalicum strain DSYD05-1. World Journal of Microbiology & Biotechnology. 2012;28:2929-35. [18] Tokiwa Y, Jarerat A. Microbial degradation of aliphatic polyesters. Macromolecular Symposia. 2003;201:283-9. [19] Wallace GG, Higgins MJ, Moulton SE, Wang C. Nanobionics: The impact of nanotechnology on implantable medical bionic devices. Nanoscale. 2012;4:4327-47. [20] Wallace GG, Moulton SE, Clark GM. Appled physics: Electrode-cellular interface. Science. 2009;324:185-6. [21] Sayyar S, Cornock R, Murray E, Beirne S, Officer DL, Wallace GG. Extrusion printed graphene/ polycaprolactone/ composites for tissue engineering. Materials Science Forum, 2014. p. 496-502. [22] Sayyar S, Murray E, Thompson BC, Gambhir S, Officer DL, Wallace GG. Covalently linked biocompatible graphene/polycaprolactone composites for tissue engineering. Carbon. 2013;52:296-304. [23] Park SY, Park J, Sim SH, Sung MG, Kim KS, Hong BH, et al. Enhanced Differentiation of Human Neural Stem Cells into Neurons on Graphene. Advanced Materials. 2011;23:H263-+. [24] Zhang H, Peng C, Yang JZ, Lv M, Liu R, He DN, et al. Uniform Ultrasmall Graphene Oxide Nanosheets with Low Cytotoxicity and High Cellular Uptake. Acs Applied Materials & Interfaces. 2013;5:1761-7. [25] Yue H, Wei W, Yue ZG, Wang B, Luo NN, Gao YJ, et al. The role of the lateral dimension of graphene oxide in the regulation of cellular responses. Biomaterials. 2012;33:4013-21. [26] Mu QX, Su GM, Li LW, Gilbertson BO, Yu LH, Zhang Q, et al. Size-Dependent Cell Uptake of ProteinCoated Graphene Oxide Nanosheets. Acs Applied Materials & Interfaces. 2012;4:2259-66. [27] Yang K, Wan JM, Zhang SA, Zhang YJ, Lee ST, Liu ZA. In Vivo Pharmacokinetics, Long-Term Biodistribution, and Toxicology of PEGylated Graphene in Mice. Acs Nano. 2011;5:516-22.

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

[28] Bikiaris DN. Nanocomposites of aliphatic polyesters: An overview of the effect of different nanofillers on enzymatic hydrolysis and biodegradation of polyesters. Polymer Degradation and Stability. 2013;98:1908-28. [29] Fukushima K, Abbate C, Tabuani D, Gennari M, Rizzarelli P, Camino G. Biodegradation trend of poly(epsilon-caprolactone) and nanocomposites. Materials Science & Engineering C-Materials for Biological Applications. 2010;30:566-74. [30] Gambhir S, Murray E, Sayyar S, Wallace GG, Officer DL. Anhydrous organic dispersions of highly reduced chemically converted graphene. Carbon. 2014. [31] He F, Li SM, Vert M, Zhuo RX. Enzyme-catalyzed polymerization and degradation of copolymers prepared from epsilon-caprolactone and poly(ethylene glycol). Polymer. 2003;44:5145-51. [32] Ku SH, Park CB. Myoblast differentiation on graphene oxide. Biomaterials. 2013;34:2017-23. [33] Liu Y, Huang J, Li H. Synthesis of hydroxyapatite-reduced graphite oxide nanocomposites for biomedical applications: oriented nucleation and epitaxial growth of hydroxyapatite. Journal of Materials Chemistry B. 2013;1:1826-34. [34] Kandiah K, Muthusamy P, Mohan S, Venkatachalam R. TiO2–graphene nanocomposites for enhanced osteocalcin induction. Materials Science and Engineering: C. 2014;38:252-62.

7. Supplementary Materials 1.6

1.2

TE D

1 0.8 0.6 0.4 0.2 0

PCL 80K Mix 1%

AC C

PCL

EP

Absolute weight loss (g)

1.4

PCL 80K DCC 0.1%

PCL 80K DCC 1%

8 U / mL 16 U / mL

PCL 80K DCC 5%

Figure S1 – Comparison of mass loss from composites after 24 hours incubation with different concentrations of lipase enzyme. No significant decrease in mass loss at 24 hours indicated that degradation of composites was not limited by use of 8U/mL Pseudomonas lipase.

ACCEPTED MANUSCRIPT

mixPCl-CCG5%

cPCl-CCG0.1%

cPCl-CCG1%

cPCl-CCG5%

RI PT

PCl alone

AC C

EP

TE D

M AN U

SC

Figure S2 – Degradation media after 96 hours degradation showing white precipitate of PCl degradation, grey precipitates for covalently modified cPCl-CCG composites and grey-black precipitate for mixPClCCG5%.