Development of Drug-Eluting Surgical Sutures for the Wound- Healing Process Using Melt Extrusion Technology

Xiaoxuan Deng

A thesis submitted for the degree of Master of Science at the University of Otago, Dunedin, New Zealand.

February 2020

ABSTRACT

Incorporation of an anti-inflammatory drug with polymer materials to produce suture products is important especially when the suture is used for internal organs and/or tissues. In this work, poly(ε-caprolactone) (PCL), polyethylene glycol 400 (PEG400), chitosan and keratin were used as biopolymer matrix to encapsulate 5, 15, 30 wt% diclofenac potassium (DP) to achieve controlled drug release.

At first, polymer alone was extruded by hot- melt extrusion technique under various temperatures and at different PCL/PEG400/chitosan-keratin ratios. The products were analysed in terms of FTIR characterization, thermal properties and mechanical properties. According to the characterization analysis of polymer sutures, the optimal polymer formulation (PCL: PEG400: chitosan/keratin= 80:19:1 wt%) (Group 1) was chosen to manufacture sutures with drug DP content at 5, 15, 30 wt%. In addition, PCL/PEG400/chitosan (Group 2) and PCL/PEG400/keratin (Group 3) were also fabricated with drug DP to create sutures, the polymer and drug weight ratios were 95:5, 85:15 and 70:30 wt%. Drug release features were investigated via the drug dissolution test. MTT assay, live/dead assay and scratch assay were also applied to study the biocompatibility of drug embedded sutures.

Results showed that sutures can be extruded with a smooth surface and uniform thickness under temperature of 63 ± 1 °C. It was observed from DSC and TGA studies that completely amorphous and miscible solid dispersions were prepared. FTIR analysis indicated that the presence of hydrogen bonds between the polymers improved material miscibility. Tensile properties of the sutures were clearly affected by PEG400, chitosan and keratin addition. The optimal formulation of tensile strength was obtained when PEG400: chitosan/keratin was 19:1 wt%. Rapid drug release rate was observed at different stages of group 1, 2, 3. For group 1, the release occurred rapidly occurred after 1 h 45 mins, whereas group 2 exhibited fast release between 1 h 45 mins and 3 h 45 mins. Group 3 displayed a sustained release where a fast release occurred from the onset to 3 h 45 mins. However, the final concentration of drug release in group 1 was lower than the other two groups, which can be predicted that a longer drug release time would be in an extended period compared to the other two groups. According to biocompatibility test, group 1 demonstrated very high cell viability at 72 h. Especially

sample F3, which contained 30 wt% drug, and the ratio of PCL/PEG/chitosan-keratin was 80/19/1 w/w. It achieved the highest wound healing rate in the scratch assay.

iii

ACKNOWLEDGEMENTS

I would first like to thank my supervisor Associate Prof. Azam Ali of the Centre for Bioengineering and Nanomedicine, Department of Food Science. He has been my supervisor since the first day I came to University of Otago to start my postgraduate study. His advice and expertise were priceless to me and he was willing to answer my questions and doubts at any of his free time. Owing to his help and suggestions, I could work on this interesting research topic, and will pursue knowledge on this path further. He encouraged me in my work and guided me on the right track when I was confused. We had fortnightly meeting on Thursday, which was really helpful.

I would also like to thank Dr. Maree Gould of the Centre for Bioengineering and Nanomedicine, Department of Food Science, who was never tired of explaining virous cell culture concepts and analysis technique. Helping me familiar with operating procedure for MTT assay, live/dead assay and scratch assay, and recommending me with Prism software which made the graphic building and statistical analysing much easier. I could not complete my data analysis in a relatively short time without her great help. On the other hand, she is also a friendly and funny person, who always treats people with kindness and lots of harmless jokes. I never lose smile on my face whenever talking to her. I will forever be grateful to Ian Stewart of the Department of Chemistry whose immense help and guidance allowed me to carry out TGA and DSC experiments. Many thanks also to fellow PhD students Tajul Islam for his patient explaining whenever I felt confused about experimental steps or concepts, and Stephen Giteru for his training for mechanical testing.

I would like to thank my friend Bhavini Patel, who invited me to go to gym together and shared life experience with me, I often felt inspired just by chatting with her. I must also express my infinite gratitude to Dino Milotic who sacrificed lots of time reading and editing my thesis, giving me detailed feedback. He also gave me suggestion and encouraged me when I needed it the most. Finally, I have to thank my parents who provided me this great opportunity to come to University of Otago studying Master of

Science. This accomplishment would not have been happened without their encouragement and financial support.

TABLE OF CONTENTS

Abstract ...... ii Acknowledgements ...... iv Table of Contents ...... vi List of Tables ...... ix List of Figures ...... x List of Abbreviations ...... xii 1 Introduction ...... 1 1.1 Antimicrobial Sutures ...... 1 1.2 Classification of Suture Material ...... 4 1.2.1 Absorbable Sutures ...... 6 1.2.2 Non-Absorbable Sutures ...... 9 1.3 Sutures with Active Pharmaceutical Ingredients (APIs) ...... 11 1.4 Suture Structure ...... 14 1.4.1 Monofilament and Multifilament (Braided) Sutures ...... 14 1.4.2 Barbed Sutures ...... 15 1.4.3 Smart Sutures ...... 16 1.5 Fabrication Process ...... 16 1.5.1 Electrospinning ...... 24 1.5.2 Melt Extrusion ...... 26 1.5.3 Coating ...... 28 1.6 Conclusion ...... 29 2 The Current Project ...... 31 2.1 Fabrication of Drug-Eluting Sutures with Hot- melt Extrusion Technique...... 31 2.2 Materials Selected for This Study ...... 32 2.3 Aims and Objectives...... 33 3 Materials and Methods ...... 35 3.1 Materials ...... 35 3.2 Equipment ...... 35 3.3 Suture Preparation for Control Group ...... 36 3.4 Fourier Transformation-Infrared Spectroscopy (FT-IR) ...... 39 3.5 Thermal Gravimetric Analysis (TGA) ...... 40 3.6 Differential Scanning Calorimetry Analysis (DSC) ...... 41

3.7 Mechanical Properties ...... 42 3.8 Suture Preparation for Formulation Group ...... 43 3.9 Drug Release Studies ...... 46 3.9.1 Standard Preparation ...... 46 3.9.2 In vitro Drug Release Study ...... 46 3.10 Biocompatibility Tests...... 48 3.10.1 Cell Lines and Maintenance...... 48 3.10.2 Cell Viability in Response to Suture Materials ...... 51 3.10.3 Fluorescence Microscopy ...... 54 3.10.4 Wound Healing in Response to Suture Materials ...... 56 4 Results ...... 58 4.1 Suture Preparation for the Control Groups ...... 58 4.1.1. Melt Extruder ...... 58 4.1.2 Melt Extrusion at 63±1 °C ...... 59 4.1.3 Melt Extrusion at Various Temperatures ...... 63 4.2 Summary of Biomaterial Production and Suture Generation ...... 64 4.3 Fourier-Transform Infrared Spectroscopy ...... 65 4.4 Thermal Testing ...... 69 4.4.1 TGA ...... 69 4.4.2 DSC ...... 73 4.5 Tensile Testing ...... 78 4.6 Summary of Suture Characteristics ...... 82 4.7 Suture Preparation for the Formulation Groups ...... 83 4.8 Drug Dissolution Testing ...... 86 4.8.1 Calibration Curve ...... 86 4.8.2 Drug Release Profiles ...... 88 4.9 Summary of Drug-Eluting Suture Fabrication and Drug Dissolution ...... 92 4.10 Biocompatibility Analysis ...... 93 4.10.1 MTT Assay ...... 93 4.10.2 Live/Dead Assay ...... 100 4.10.3 Scratch Assay ...... 110 4.11 Summary of Biocompatibility Tests ...... 113 5 Discussion ...... 114 5.1 Analysis of Suture Fabrication and Characteristics ...... 115 5.1.1 Hot- melt Extrusion Parameters ...... 115 5.1.2 Characterisation of Polymer Sutures...... 117 5.1.3 Polymer Composites as Drug Carriers ...... 120

vii

5.2 Analysis of Drug Dissolution with Products Produced by Solid Dispersion Technique ...... 121 5.2.1 Solid Dispersion ...... 121 5.2.2 Dissolution Behaviour ...... 122 5.3 Analysis of Biocompatibility of the Sutures...... 124 5.3.1 Effect of Drug-Eluting Sutures on Cell Proliferation ...... 125 5.3.2 Effect of Drug-Eluting Sutures on Cell Viability ...... 126 5.3.3 Effect of Drug-Eluting Sutures on Cell Migration ...... 127 5.4 Comparison of Suture Sample with Commercial Product ...... 128 5.5 Critical Analysis of Methodology and Limitations ...... 128 5.6 Future Directions ...... 130 5.7 Conclusion ...... 131 References ...... 133

viii

LIST OF TABLES

Table 1. General Comparison of Suture Materials and Properties ...... 5 Table 2. The Summary of Drug-Eluting Suture Fabrication Processes ...... 17 Table 3. Control Group (C) (weights/ 10 g in total) ...... 37 Table 4. Formulation Group (F) (weights/ 10 g in total) ...... 44 Table 5. Control Group ...... 52 Table 6. Experimental Group ...... 52 Table 7. The Physical Observations of Extruded Polymer Fibres ...... 60 Table 8 Control Group (contains PCL+PEG400+Chitosan/keratin) at Various Temperatures...... 63 Table 9. Functional Groups of PCL ...... 65 Table 10. Summary of Differential Scanning Calorimetry Analysis of Samples ...... 73 Table 11. Endothermic Transition for Each Thermal Property ...... 74 Table 12 Exothermic Transition for Each Thermal Property ...... 74 Table 13. Summary of Mechanical Testing Analysis of Suture Samples...... 79 Table 14. The Summary of the Suture Properties with Different Formulations ...... 82 Table 15. Summary of Suture Preparation for Group 1 ...... 84 Table 16. Summary of Suture Preparation for Group 2 ...... 84 Table 17. Summary of Suture Preparation for Group 3 ...... 84 Table 18. Absorbances of Known Concentrations of Drug in Phosphate Buffer at pH 7.4...... 86 Table 19. Concentrations of Dissolution Media at Different Time Intervals...... 89

LIST OF FIGURES

Figure 1. Manufacturing Procedure of Drug-Eluting Sutures ...... 4 Figure 2. Commercialised Suture Materials ...... 5 Figure 3. Barbed Suture ...... 16 Figure 4. (a) Uniaxial Electrospinning and (b) Coaxial Electrospinning ...... 25 Figure 5. Core-Sheath Structure ...... 25 Figure 6. Melt Extrusion Process ...... 26 Figure 7. Summary of The Study Design ...... 34 Figure 8. Suture Material Preparation without Drug ...... 37 Figure 9. Mixing Extruder ...... 38 Figure 10. Extruding Process ...... 38 Figure 11. ALPHA FTIR (Bruker) ...... 39 Figure 12. TGA Q50 ...... 40 Figure 13. DSC Q2000 ...... 41 Figure 14. Tensile Analyser ...... 42 Figure 15. Suture Material Preparation with Drug ...... 45 Figure 16. Groups 1, 2, 3 Mixtures ...... 45 Figure 17. Suture Immersion ...... 47 Figure 18. UV-vis Spectrophotometer ...... 47 Figure 19. Thermo Scientific Liquid Nitrogen Tank & Water Bath...... 49 Figure 20 Herasafe KS Class П Biosafety Cabinet ...... 49

Figure 21 CO2 Incubator & Flasks of Complete DMEM ...... 50 Figure 22. Handled Automated Cell Counter ...... 50 Figure 23. Plate Design for MTT Assay ...... 53 Figure 24. UV-Vis Plate Reader ...... 53 Figure 25. Plate Design for Live/Dead Assay ...... 55 Figure 26. Olympus IX71 Inverted Microscope ...... 55 Figure 27. Plate Design for Scratch Assay ...... 57 Figure 28. Ceti Magnum Trinocular Compound Microscope ...... 57 Figure 29. Extrudate of Sample C3 at 63±1 °C with Length of 38 cm...... 60

Figure 30. Extrudates of Sample C4 & C5 at 63±1 °C ...... 61 Figure 31. Extrudate of Sample C5 at 63±1 °C with Length of 186 cm...... 61 Figure 32. Extrudates of C5 at Variable Temperatures ...... 62 Figure 33. Temperature Controller of The Micro-Extruder...... 62 Figure 34. FTIR Spectra of Individual Samples and Overlay Graph ...... 68 Figure 35. TGA Curves of Individual Samples and Overlay Graph...... 72 Figure 36. DSC Curves of Individual Samples and Overlay Graph ...... 77 Figure 37. Tensile Properties of Suture Samples ...... 80 Figure 38. Ductile Failure of Suture Samples ...... 81 Figure 39. Young’s Modulus of Suture Samples ...... 81 Figure 40. Extrudates of Suture Samples Contained Drug ...... 85 Figure 41. Standard Curve of Drug in Phosphate Buffer at pH 7.4...... 87 The calibration curve was used for calculating the unknown concentrations with known absorbances in the drug dissolution experiment...... 87 Figure 42. Dissolution of Group 1 (F1, F2, F3) in Phosphate Buffer at pH 7.4 ...... 90 Figure 43. Dissolution of Group 2 (F4, F5, F6) in Phosphate Buffer at pH 7.4 ...... 90 Figure 44. Dissolution of Group 3 (F7, F8, F9) in Phosphate Buffer at pH 7.4 ...... 91 Figure 45. Comparison of F3, F6 and F9 in Phosphate Buffer at pH 7.4 ...... 91 Figure 46. HaCat Cells ...... 95 Figure 47. Purple Formazan Crystals of HaCat Cells ...... 95 Figure 48. Colour Changing of Sutures After Adding MTT Solution ...... 96 Figure 49. Representative standard curve of MTT assay ...... 96 Figure 50. Cell Number for Group 1 via MTT Assay ...... 97 Figure 51. Cell Number for Group 2 via MTT Assay ...... 98 Figure 52. Cell Number for Group 3 via MTT Assay ...... 99 Figure 53. Live/Dead Assay with Fluorescence Microscopy...... 107 Figure 54. Fluorescence of Live/Dead Cells on Sutures...... 107 Figure 55. Ratio of Live/Dead Cells in Response to Different Sutures at 24, 48 and 72h ...... 109 Figure 56. Images of Gap Closure in Wound Healing Assay in Response to Sutures ...... 111

LIST OF ABBREVIATIONS

°C Degree Celsius µl Microliter ANOVA Analysis of Variance API Active Pharmaceutical Ingredient CB Calcein Blue cDMEM Complete DMEM CHX Chlorhexidine DMEM Dulbecco Modified Eagle Medium Dic Diclofenac DP Diclofenac Potassium FCS Fetal Calf Serum FDA Food and Drug Administration FTIR Fourier-Transform Infrared Spectroscopy HA Hyaloplasm Acid HaCat Immortalized Keratinocytes h Hours HCL Hydrogen Chloride HME Hot- melt Extrusion LVFX Levofloxacin mg Milligram mL Millilitre mm Millimetre MTT 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide nm Nanometre PAH Poly(allylamine hydrochloride) PBS Phosphate Buffered Saline PCL Polycaprolactone PDS Polydioxanone PEG Polyethylene Glycol

xii

PGA Polyglycolic Acid PI Propidium Iodide PLLA Poly(lactic acid) RC Regenerated Collagen SSI Surgical Site Infection Tg Glass Transition Temperature Tm Melting Point Temperature UV Ultra-violet

xiii

1 INTRODUCTION

Surgical sutures are medical devices used for wound closure to hold body tissue together. Implanting sutures inside the human body may lead to an inflammatory response and/or postoperative surgical site infection (SSI), resulting in additional requirements such as pharmaceutical treatments. For example, oral administration of drugs often requires a high dosage due to low drug bioavailability and lack of specificity. Moreover, non-specific drug administration may have side-effects on tissues and organs that may or may not be the goal of the proper treatment. In this case, local drug delivery is essential (Champeau et al. 2017).

Conjugating polymers with bioactive agents can achieve the purpose of controlling release kinetics and improving the targeting efficacy. Controlled release systems help to ensure a high drug concentration in the wound area (Lee et al. 2017). The utility of biodegradable sutures to deliver drugs not only eliminates the need of implanting foreign material inside the wound sites, but also effectively reduces surgical infection (Joseph et al. 2017). Sutures can be incorporated with an active pharmaceutical ingredient (API) , which is the component of any drug that produces the intended effects, to achieve controlled and optimal drug release at the wound site (Champeau et al. 2017).

1.1 Antimicrobial Sutures

Staphylococcus aureus (S. aureus) and Escherichia coli (E. coli) are the most popular organisms used in the trialling of new suture applications (Henry-Stanley et al. 2010). S. aureus, a class of gram-positive bacteria, may colonize the surface of implants. E. coli is a gram-negative organism, and is capable of triggering wound infections (Augustine and Rajarathinam 2012). Both organisms may cause complications in terms of wound healing when the suture is implanted inside the human body (Viju and Thilagavathi 2013b, Dennis et al. 2016). The hydrophobic or hydrophilic property of sutures determines the bacterial attachment preference. Most pathogens tend to attach to hydrophobic surfaces and have more stable viability under less humid environments (Ercan et al. 2018). However, Dhom et al. (Dhom et al. 2017) showed that the slightly hydrophilic suture component, Vicryl®, had the

highest bacterial colonization, which was ascribed to its macrostructure. Thus, the macrostructure enabled the bacteria-containing liquid to be absorbed into the suture strand, which made cleaning difficult, leading to massive pathogenic growth (Dhom et al. 2017).

Any artificial suture surface is prone to the attachment of microbial cells. Under moist conditions, these cells may attach and proliferate very quickly leading to the formation of a biofilm. This biofilm enables microbial cells to survive in a harsh environment, subsequently reducing susceptibility to antibiotics. In this situation, the suture material and wound site are both contaminated because of the formation of the biofilm (Joseph et al. 2017). The prevention of biofilm formation requires a suture with a biphasic release profile. This suture has a fast-antimicrobial release phase in order to retard biofilm formation initially, followed by a slow, prolonged release for maintaining a sufficient antibiotic concentration (Alvarez- Paino, Munoz-Bonilla, and Fernandez-Garcia 2017).

There are two common methods to combine API with suture materials. The first one is adding API during the suture manufacturing stage; the second is post-addition of API into ready- made sutures (Champeau et al. 2017). The first method can be achieved by two fabrication processes: electrospinning and hot- melt extrusion. In these two processes, API is dispersed within the suture materials. The post-addition method can be achieved with process of dip- coating or soaking (Champeau et al. 2017). Among them, the dip-coating method is the most prevalent strategy which involves coating the suture in a prepared drug solution. The suture simply absorbs the drug onto the surface, which leads to low drug loading and drug release efficiency (Lee et al. 2017). However, coating APIs on the suture surfaces results in rapid drug release (Champeau et al. 2017). Thus, blending APIs with suture materials during fabrication, such as hot- melt extrusion method, is more desirable when sustained drug release is required.

The most important surgical suture characteristics for enhancing wound healing include: tensile strength, knot security, tissue reactivity and handling properties (Chu et al. 1996). The handling properties include pliability (i.e. structural flexibility) and friction force (i.e. the slipperiness of a suture during operation). Higher friction force may cause stronger tissue drag, and in extreme cases may lead to the difficulty of passing a suture through tissues, resulting in severe tissue injury (Debbabi and Abdessalem 2015).

Additional antimicrobial agents may be used for repression of surgical site infections (SSIs). There are two main antibacterial strategies for using sutures. The first strategy uses cationic biopolymers to passively coat the suture surfaces to prevent bacterial attachment (Serrano et al. 2015). The second strategy incorporates APIs, such as silver or antibiotics, alongside suture materials, which releases API from the surgical suture after implantation within the human body (Serrano et al. 2015). This method of drug release is also called drug-eluting and is used for removing bacteria suspended in/around tissues and organs. While the former strategy offers the benefit of biocompatibility, the latter strategy is more effective.

Drug-eluting sutures have been developed using a variety of fabrication methods, such as dip-coating, melt extrusion and electrospinning (Dennis et al. 2016). Figure 1 summarises the current manufacturing processes of drug-eluting sutures. Drug-eluting suture development has several problems, such as low tensile strength, maintenance of drug release rate, and commercial applications (Arora et al. 2019). Poly(lactic acid) (PLLA), Polyethylene glycol (PEG), and model drug levofloxacin (LVFX) have been applied to fabrication sutures with wet electrospinning technique. Formulations were increased with PEG content from 1% to 4%, which showed an initial burst release followed by a sustained release. However, the suture with the highest concentration of PEG had the largest bacteria-killing potential at 24 hours (Kashiwabuchi et al. 2017). Therefore, the design of polymer-drug formulation is very important in terms of bacterial exhibition. The application of PEG as suture material was also investigated its impact on suture tensile strength. However, the varying concentrations of PEG did not attribute any significant effect to LVFX and PLLA (Dhom et al. 2017). To be able to tune the mechanical properties of sutures by adjusting formulations, chitosan, keratin and cellulose, or a selected combination of either two of them can be considered as addition materials (Tran and Mututuvari 2015).

Figure 1. Manufacturing Procedure of Drug-Eluting Sutures 1.2 Classification of Suture Material

Suture materials that include absorbable and non-absorbable sutures may both be classified as synthetic or natural sutures (Figure 2). The summary of suture materials is shown in Table 1. Natural nonabsorbable materials such as cotton, silk and linen are less preferred since they are prone to SSIs. However, synthetic absorbable sutures have predictable degradation and relatively lower contamination potential (Lee et al. 2017). Thus, incorporating drugs with synthetic absorbable sutures becomes essential, especially in relation to internal organs or tissue application.

Table 1. General Comparison of Suture Materials and Properties

Suture Tensile Tissue Origin Absorbability Configuration material strength reactivity Collagen Natural Absorbable Twisted Low High Silk Natural Non- Braided Low High absorbable Line Natural Non- Twisted High High absorbable Cotton Natural Non- Twisted Low High absorbable Polydioxanone Synthetic Absorbable Monofilament High Low Polyglyconate Synthetic Absorbable Monofilament Moderate Low Poliglecaprone Synthetic Absorbable Monofilament Low Low Polyglycolic Synthetic Absorbable Monofilament/ Moderate Low acid Braided Polyglactin Synthetic Absorbable Braided Moderate Low 910 Polyester Synthetic Non- Braided High Low absorbable Polyamide Synthetic Non- Monofilament/ Moderate Low absorbable Braided Polypropylene Synthetic Non- Monofilament Moderate Low absorbable Polybutester Synthetic Non- Monofilament High Low absorbable Stainless steel Synthetic Non- Monofilament/ High Low absorbable Braided

Figure 2. Commercialised Suture Materials

1.2.1 Absorbable Sutures

Absorbable sutures are defined as those that lose 50% tensile strength within 60 days of their application (Dennis et al. 2016). The reduction of tensile strength is attributed to proteolytic or hydrolytic degradation, depending on the natural or synthetic absorbable sutures (Erçin and Karahan 2018).

Polymer sutures are effective when used in surgical applications if they meet the exacting needs of mechanical properties and the degradation rate. Claude et al. (2007) used rat models to study vascular anastomose using microsurgery. Even though there were no significant differences in macroscopic, histologic and functional assessments between absorbable and nonabsorbable sutures, the local tissue inflammatory reaction was more serious when nonabsorbable sutures were used in rat tissues (Claude et al. 2007).

1.2.1.1 Natural Absorbable Suture Material

Natural absorbable suture materials include catgut, which is derived from sheep or goat intestines and contains over 99% pure collagen. There are two catgut forms sold commercially; plain catgut and treated catgut tanned by chromium trioxide (called chromic catgut) (Pillai and Sharma 2010). Chromic catgut has stronger mechanical properties, a lower degradation rate than plain catgut, and may also reduce tissue reactions (Gabel et al. 2000). Catgut sutures are slowly degraded by proteolytic enzymes within the human body. The peak tensile strength of catgut sutures occurs <4 days post implantation, and completely loses tensile strength after 2 weeks (Pillai and Sharma 2010). Catgut suture are known to produce a strong inflammatory reaction the first three days after implantation (Bichon, Borloz, and Cassano-Zoppi 1984). Moreover, the absorption rate of catgut sutures is slower than the suture degradation rate (Chellamani, Veerasubramanian, and Balaji 2013), causing mechanical strength loss faster than wound healing, leading to complications (Tajirian and Goldberg 2010).

Regenerated collagen (RC) has been reported to have low immunologic activity. Specifically, RC sutures made from bovine long flexor tendons have been exclusively utilised in microsurgery applications. RC may be produced by two methods; enzymatic digestion of native collagen-rich tissues or extracting the tissues with salt solutions (Pillai and Sharma

2010). RC may be enhanced by making it spinnable, including crosslinking, grafting and blending (Krishnamoorthy et al. 2014, Hirano et al. 2000). Additionally, the mechanical and thermal properties of RC filaments may be improved by maturation during the incubation process (Pillai and Sharma 2010). For example, a study prepared RC from cattle skin through wet-spinning method, showing that the spun fibres had excellent biocompatibility with improved cell adhesion and proliferation (Wang et al. 2016).

1.2.1.2 Synthetic Absorbable Suture Material

Poly(p-dioxanone) (PDS) is a monofilament suture that has been recently developed due to its potential for excellent tensile strength and knot properties. It also demonstrates low tissue reactivity and enhanced hydrolysis. When PDS was used in x and y, the tissue reactivity was significantly lower than when x and y sutures were used. However, its melting point is low which may cause pyrolysis during spinning manufacturing process (Lee et al. 2014). Compared with poly(glycolic-co-lactic acid) (Vicryl®) and poly(glycolic acid) (Dexon®), PDS exhibits lower inflammatory reaction rates (Middleton and Tipton 1998). Commercially available PDS is sold in monofilament form, such as PDS®II and PDM ®. This suture may maintain > 50% of tensile strength after 4 weeks of implantation (Im et al. 2007). Additionally, PDS is also manufactured as a commercially available barbed suture product. For instance, Quill™ SRS is a bidirectional barded suture product used to close deep subcuticular layers and to close tissues in deeper layers, such as abdominal closures and paediatric cardiac procedures (Greenberg 2010).

Polyglyconate sutures are commercial monofilament products fibres without coating (Maxon®). This suture, consisting of glycolic acid and trimethylene carbonate (1,3-dioxan-2- one) building blocks, is sterilized by ethylene oxide (Chu 2013), which has been found to effectively reduces the body’s inflammatory response to suture application using this product. Moreover, Polyglyconate degrades via hydrolysis ~60 days after implantation inside the body. The degradation process continues for up to 180 days after implantation. This suture generates the best knot security compared to the other synthetic absorbable sutures and has been successfully used for soft tissues or ligation, including cardiovascular tissue and peripheral vascular surgery (Chellamani, Veerasubramanian, and Balaji 2013).

Poliglecaprone sutures, more commonly known as Monocryl®, are absorbable, monofilament sutures comprised of glycolide and epsilon caprolactone copolymers (PGCL) (Langley- Hobbs 2013). This type of suture typically loses 50% of its tensile strength within 7 days after implantation and 100% tensile strength loss within three weeks. Therefore, it is only used for wounds with fast healing (Chu 2013). However, it has good knot security and handling properties attributed to its reduced pliability and poor structural memory (Chellamani, Veerasubramanian, and Balaji 2013). Moreover, it has been reported that Poliglecaprone has significantly lower microbial adherence rate than nonabsorbable multifilament and monofilament sutures (Chu 2013). Previous studies found that using monofilament Poliglecaprone sutures for wound healing caused less hypertrophic scar formation compared to multifilament Vicryl RapideTM. Even though absorbable sutures aren’t normally used for superficial wound healing, Poliglecaprone has been recommended for healing injuries to the face, ear, abdomen etc, as it provides great healing properties (Trott 2012).

Poly(glycolic acid) (PGA), is the simplest biodegradable linear aliphatic polyester, and comes in monofilament, braided or coated forms. These sutures are widely used to treat contaminated wounds (Hochberg, Meyer, and Marion 2009). The only mode of degradation is hydrolysis, which comprises of two steps; i) firstly, ester bonds are hydrolysed by the water, allowing water to diffuses into the amorphous regions, ii) secondly, the entire polymer structure dissolves by the scission of the crystalline section (Pillai and Sharma 2010). After implantation in the human body, hydrolytic enzymes begin the degradation of PGA. Enzymes involved in the degradation of PGA include esterase, trypsin, and chymotrypsin (Makino, Arakawa, and Kondo 1985). The degradation products of PGA are non-toxic and are expelled by the body via urine (Pillai and Sharma 2010). Moreover, once implanted, PGA’s tensile strength is reduced by up to 80% in 14 days and alkaline conditions further decrease knot pulling strength (Chellamani, Veerasubramanian, and Balaji 2013). Further studies are warranted to explore the incorporation of monofilament PGA sutures with APIs to achieve a more controlled degradation and to reduce site infections near sutures (Lee et al. 2014).

Polyglactin 910, also known as Vicryl®, is a braided suture that consists of copolymers of glycolic acid and lactic acid in a 90/10 mol/mol multifilament form. This suture is coated with calcium stearate and can maintain its original tensile strength < 14 days post

implantation, followed by gradual degradation between 100 to 120 days. Many researchers have employed coating techniques to cover Polyglactin 910 with bioactive agents, resulting in less tissue drag and a lower rate of wound infection (Chellamani, Veerasubramanian, and Balaji 2013). Vicryl® rapid is a Polyglactin 910 variant with improved degradation rate compared to Vicryl®. Panacryl® is another variant formed by the copolymer of glycolic acid and lactic acid but with higher ratios of these two acids, which reduces the rate of suture degradation inside the human body (Pillai and Sharma 2010).

1.2.2 Non-Absorbable Sutures

Non-absorbable sutures consist of non-biodegradable biomaterials, such as silk, nylon, polyester and metal. They are more mechanically durable than absorbable sutures. The application of non-absorbable sutures is limited to skin, fascia or tendons where tissues require higher tensile strength (Lee et al. 2017). Despite the excellent mechanical properties of non-absorbable sutures provide to wound sites. The major disadvantage of using nonabsorbable sutures is that they are required a second procedure to remove the sutures after the wound heals (Joseph et al. 2017).

1.2.2.1 Natural Non-Absorbable Sutures

The braided form of silk is used as a natural nonabsorbable surgical suture material. It is dyed black for better visibility and is usually coated with oil, wax or silicone for improved toughness. Despite enhancements, silk sutures have the lowest tensile strength compared to all the currently available sutures (Hochberg, Meyer, and Marion 2009). Additionally, studies have shown that the braided sutures using silk material may introduce pathogens into the wound site, leading to infection (Wu et al. 2016). However, silk sutures show a high performance in serum and have good water-repellent properties. Silk sutures are mainly used to reduce inflammation once placed in the wound site and are not responsible for antibacterial action (Chellamani, Veerasubramanian, and Balaji 2013).

Linen sutures are most commonly produced in the twisted multifilament form flax (Fong et al. 2015), and they may also be coated with silicone and polyvinyl solutions (Chellamani, Veerasubramanian, and Balaji 2013). Linen sutures do not lose strength after implantation

and have the ability to gain 10-20 % tensile strength in an aqueous environment due to their non-absorbable property, suggesting that a wet linen suture is able to endure high stress while securing wound closure (Cherif et al. 2013).

Cotton, extracted from cotton plants, is coated with wax to be used for surgical suture applications. A cotton suture typically loses 50% tensile strength within half a year and can completely lose strength within two years (Ratner et al. 2004). Like linen sutures, cotton sutures may gain roughly 10% tensile strength when moist. Wound site infections and tissue reactions may occur frequently using cotton suture due to its high capillarity (Chellamani, Veerasubramanian, and Balaji 2013). In addition, cotton suture material can generate static electricity causing it to cling to surgical linen (Ratner et al. 2004). Therefore, the application of cotton sutures is limited.

1.2.2.2 Synthetic Non-Absorbable Sutures

Polyester sutures are nonabsorbable multifilament braided structures comprised of polyethylene terephthalate. These sutures are commonly utilized in the coated form as it adds variation to their application. Various coatings, such as polybutylate, Teflon and silicone may reduce tissue drag (Chellamani, Veerasubramanian, and Balaji 2013). These types of sutures have extremely high tensile strength, ranking only lower than metal, and very little to no loss in tensile strength observed after implantation (Herrmann 1971). Furthermore, these sutures provide sustained support for slowly healing tissues and produce minimal tissue reactivity (Hochberg, Meyer, and Marion 2009). These are generally used for soft tissues such as cardiovascular and ophthalmic applications.

Polyamide was the first developed synthetic suture. It can be applied in both monofilament and multifilament forms, giving it flexibility in possible applications. The multifilament form of polyamide sutures is commonly used in dermatologic surgery. However, even though it demonstrates excellent handling properties during operation, the tissue reactions are frequently observed (Debbabi and Abdessalem 2015). When in monofilament form, the widely used polyamide suture, Nylon, may lose 30 % of tensile strength within two years after implementation, giving it excellent long-term degradation resistance. However, in multifilament form the entire tensile strength may degrade within six months (Chellamani, Veerasubramanian, and Balaji 2013).

Polypropylene is available in the monofilament form and it is formed by catalytic polymerisation of propylene. It is the best suture for the skin healing process (Chellamani, Veerasubramanian, and Balaji 2013). As it is easy to implant and remove from the skin due to its naturally smooth surface. it may maintain the same tensile strength after implantation. And it is also able to pass through skin tissue and induce minimal inflammatory responses. However, the smooth surface may reduce knot security (Hochberg, Meyer, and Marion 2009).

Polybutester is a monofilament suture that has thermoplastic properties and is composed of polybutylene, polyglycol and polytetramethylene terephthalates (Chellamani, Veerasubramanian, and Balaji 2013). Polybutester has sufficient tensile strength to support tissue healing processes (Hochberg, Meyer, and Marion 2009). It does not have significant suture memory and will not remain its shape in the package like nylon and polypropylene. Therefore, polybutester exhibits excellent handling properties and higher knot security. Moreover, polybutester sutures demonstrate lower chance of causing hypertrophic scarring than nylon sutures (Trott 2012).

Stainless steel, the metallic suture that can be applied in both monofilament and multifilament structure. It has the highest tensile strength and knot security compared to the other available suture materials. It also maintains its tensile strength throughout the entire healing process. Moreover, it is easily sterilised using the autoclaving process. But it tends to cut tissues in the wound site, and knot tying is difficult to perform with stainless steel sutures (Chellamani, Veerasubramanian, and Balaji 2013).

1.3 Sutures with Active Pharmaceutical Ingredients (APIs)

Many researchers have worked on sutures incorporating APIs. Amongst them, some existing studies have developed fatty acid or silver nanoparticle coated silk sutures to ensure a slow- release drug delivery system (Wu et al. 2016). Chen et al. (2015) coated the braided silk sutures by mixing levofloxacin hydrochloride and PCL and found excellent antimicrobial activity against both S. aureus and E. coli (Chen et al. 2015).

Chitosan is an ideal candidate to facilitate wound closure as it has excellent antimicrobial and scar preventive characteristics along with the low toxicity. The antimicrobial mechanism operates by binding positively charged amine groups with negatively charged microbial surfaces (Muzzarelli and Muzzarelli 2005), which affects cell permeability and leads to leaking of bacterial cell cytoplasm and apoptosis. Viju and Thilagavathi (Viju and Thilagavathi 2013b) studied the effect of varying concentrations of chitosan on the braided silk sutures, and they observed that the increase in chitosan concentrations resulted in knot strength improvement. Furthermore, agar diffusion experiments demonstrated that higher chitosan concentration inhibited both E. coli and S. aureus bacteria better than lower concentrations (Viju and Thilagavathi 2013b). Chitosan-collagen blends allow additional suture properties, such as allowing collagen to be spinnable. The hydrogen bond between chitosan and collagen modifies the triple helix structure of collagen and enable mixing at the molecular level (Chen et al. 2010). Optimal mechanical and biological properties may be achieved with chitosan-collagen blends of 4 to 1 collagen to chitosan (Wang et al. 2016).

Reinbold et al. (2017) developed a novel hydrophobic, bioactive agent of totarol to act as an antibacterial coating agent on surgical sutures in order to reduce SSIs (Reinbold et al. 2017). Even though the results showed that the totarol coating inhibited bacterial biofilm formation compared to uncoated sutures, the inhibition zones of the agar diffusion test were generally small, attributed to the poor diffusion of totarol within the sutures (Reinbold et al. 2017). To improve the efficacy of totarol diffusion, other suture materials such as PCL, PLA and PGA may be considered as alternatives (Liu et al. 2020).

Nitric oxide (NO) is a natural antibacterial agent found to facilitate wound healing. The gaseous delivery of NO to wound sites is not practical, as it is a diatomic free radical, making it reactive. A way to deliver NO in solid form would be ideal (Lowe et al. 2014). Solid delivery of NO to tissues is achieved by the incorporation of NO into a suture, which produces a thromboresistant antimicrobial material able to provide enhanced healing to both chronic and acute wounds (Lowe et al. 2014, Tummalapalli et al. 2016). Lowe et al. (Lowe et al. 2014) utilized the melt spinning technique to control the release of NO. The suture was coated with PCL. The result showed that melt-spun sutures had high tensile strength, and PCL coated sutures significantly decreased NO release compared with the uncoated suture, which makes it easier to achieve sustained NO release (Lowe et al. 2014).

Triclosan is an antiseptic agent that has antibacterial and antifungal properties. It has been fabricated with several synthetic materials such as polydioxanone and polyglactin to produce absorbable sutures (Leaper et al. 2011, Leaper et al. 2017). Triclosan coated sutures offer high antimicrobial drug concentrations to SSIs, especially preventing the formation of biofilms. Microbial activity is retarded in the presence of triclosan due to its ability to inhibit bacterial fatty acid synthesis and reproduction (Leaper et al. 2017). Justinger et al. coated triclosan to Vicryl Plus® sutures and polydioxanone sutures respectively. The results showed that triclosan coated Vicryl Plus® sutures reduced wound infection significantly (Justinger et al. 2009). However, a study of 26 patients undergoing breast reduction surgery found that nine out of sixteen patients had dehiscence in triclosan-coated suture group (Deliaert et al. 2009), which questions the effectiveness of the triclosan-coated materials in wound healing. However, since triclosan has been widely used in healthcare and cosmetic products, a high percentage of triclosan has been detected in the urine of US citizens (Han, Lim, and Hong 2016), which may lead to gradual microbial resistance inside the human body (Carey et al. 2016).

Octenidine has been studied as an alternative API to triclosan. For example, octenidine coated PGA sutures, with fatty acid carriers, were tested for their antimicrobial efficacy on S. aureus. When compared with the triclosan control group (Vicryl Plus®), the inhibition zones observed were smaller on octenidine coated sutures than on the control group for up to nine days, which can be attributed to the lower solubility of octenidine, leading to slower drug release. An in vitro drug release study showed that octenidine coated sutures have a sustained release profile similar to triclosan coated sutures (Obermeier et al. 2018). It was also shown that octenidine coated sutures had no considerable effect on human metabolic activities and that they meet the ISO standard of biocompatibility through the cytotoxicity tests. Overall, the antimicrobial efficacy of octenidine coated sutures was only slightly lower than the commercial triclosan coated sutures, and the low toxicity of octenidine coated sutures makes them an ideal candidate for clinical applications (Obermeier et al. 2015).

Obermeier et al. (2015) found that Chlorhexidine (CHX) and octenidine, in conjunction with only fatty acid carriers (palmitic and lauric acid), may be viable alternatives for triclosan (Obermeier et al. 2015). They have a broad antimicrobial spectrum and biocompatibility, which effectively prevents the growth of most gram-negative and gram-positive bacteria. The

antimicrobial mechanisms of these two active agents are similar, as both use positively charged API molecules that bind with negatively charged bacterial cells, resulting in microbe membrane leakages (Obermeier et al. 2015). Similarly, other research showed that the antimicrobial efficacy lasts for 9 days for octenidine coated sutures, and up to 5 days for CHX coated sutures (Obermeier et al. 2018). Researchers found that octenidine and chlorhexidine coatings had higher antimicrobial activity than commercial Vicryl Plus® across varying solubilities in aqueous solutions (Obermeier et al. 2018). In ascending order of solubility, triclosan had the lowest solubility, followed by octenidine and CHX. Additionally, CHX- laurate sutures had very few viable adhered bacteria, even though incubation condition had higher bacterial concentration, which suggests excellent antimicrobial activity for CHX- laurate (Obermeier et al. 2018). Authors recommended this type of suture as a supplement to Vicryl Plus® for future clinical applications.

1.4 Suture Structure

1.4.1 Monofilament and Multifilament (Braided) Sutures

Monofilament sutures are single strands of fibres, whereas multifilament sutures consist of several strands braided together. Generally, multifilament sutures have good handling properties and create better knots than monofilament fibres, as monofilament sutures provide higher frictional force and have greater stiffness. Monofilament sutures have weaker knot strength and break more easily (Lee et al. 2014).

Additionally, braided sutures have much higher capillarity than non-braided sutures, allowing fluid to pass through the suture thread (Joseph et al. 2017), The capillarity of braided sutures makes them more prone to tissue inflammation/ immune reaction than monofilament sutures, as fibre capillarity has been associated with a more suitable habitat for bacteria (Busse 2016, Joseph et al. 2017). Nevertheless, larger inhibition zones were observed when a multifilament suture was coated with APIs compared to the monofilament sutures (Reinbold et al. 2017), suggesting that capillarity plays a role in drug delivery/dosage. The twisted structure of braided sutures offers a larger surface area for the coating, and allows the APIs to be absorbed more abundantly than in monofilament sutures. Therefore, multifilament sutures have greater

structural flexibility and provide greater surface area for coating than monofilament sutures, which makes multifilament sutures better for anti-microbial action.

1.4.2 Barbed Sutures

Barbed sutures (Figure 3), are bidirectional or one-directional/unidirectional surgical sutures without a knot. Unidirectional barbed sutures are made with monofilament fibres. The barbs on the suture body are unidirectional and the needles are swaged onto one end (Figure 3a). The opposite end is secured by an anchor or knot to prevent movement to the opposite direction where barbs are facing (Greenberg 2010). Similar to unidirectional sutures, bidirectional barbed sutures are also made from monofilament fibres. Barbs are formed by cutting the suture around the perimeter in a helical pattern. The two group of barbs face each other on two sides of the suture according to the midpoint (Figure 3b). The needles are swaged on two ends of the length for suture anchoring (Greenberg 2010).

Barbed sutures have several advantages over traditional monofilament and multifilament sutures. For example, barbed structures allow for secure closure of multiple skin layers without knot tying, which enhances the tensile strength of the suture (Tummalapalli et al. 2016). Moreover, barbed sutures retard bacterial attachment and growth, which reduces tissue inflammation and wound infection (Joseph et al. 2017). In a study, three types of the sutures (monofilament, multifilament and barbed sutures) were coated with an antibiotic substance and placed in a contaminated wound site. it was reported that the barbed suture had the highest bacterial inhibition (Dhom et al. 2017). Even though barbed sutures have advantages in clinical applications, the design of the barb front can accidentally puncture the surgical glove and result in the transference of infections to the surgeons and cause further wound site infections to the patients (Dennis et al. 2016). Furthermore, the cutting-type barbs may potentially weaken the suture core, thinning the suture diameter and reducing the tensile strength (Umranikar et al. 2017, Greenberg and Goldman 2013). However, with careful handling, barbed sutures may exert excellent antimicrobial activity in securing wound healing of skin tissues.

(a) One directional barbed suture

(b) Bidirectional barbed suture

Figure 3. Barbed Suture 1.4.3 Smart Sutures

Suture ‘smartness’ is a term used to describe the ability of the suture to return to the original shape once its shape has been altered. These sutures can be stretched at below critical temperatures before implanting. Firstly, the suture is applied loosely to the wound site, then, once the temperature rises either under body conditions or with the aid of external sources, the suture returns to its original shape. Smart sutures show good pliability and mechanical property for achieving self-tightening (Dennis et al. 2016). In terms of knot making, a tight knot may cause the damage of healthy cells, leading to skin necrosis. On the other hand, a loosely sealed wound line would introduce foreign substances to infiltrate wound site (Duarah et al. 2018). Therefore, a smart suture that can self-tighten under body temperature is desired.

1.5 Fabrication Process

There are three commonly used procedures in manufacturing surgical sutures; electrospinning, hot- melt extrusion and coating (Figure 1). Electrospinning and hot- melt extrusion allow APIs to disperse throughout the suture body in order to achieve the controlled release of drugs. The coating includes dip coating and layer-by-layer coating. Dip coating is the immersion of the surgical suture into the API solution. The layer-by-layer coating is a layer by layer deposition of APIs. In both cases, the drug is coated on the surface of the suture to achieve fast release. The summary of these three fabrication processes is listed in Table 2.

Table 2. The Summary of Drug-Eluting Suture Fabrication Processes

Manufacturing Research topic Suture Bioactive The polymer Suture Solubility Tensile API release Reference Process material agent and/or drug structure/ strength profile amount Suture diameter The comparison of Poly (L- Tetracycl PLLA/ TCH Structure: Absorbable Low Blend: initial (He, Huang, blend and coaxial lactic ine (wt%): Monofilam burst release and Han 2009) electrospinning acid) hydrochl 10/3 (Blend) ent & core- followed by a techniques by (PLLA) oride shell very slow incorporating model (TCH) PLLA /TCH release. drug into polymer (wt%): Diameter: fibres 10/10 Blend: 692 Core-shell: 8/10 nm initial burst 5/10 (Core- Core-shell: release followed Electrospinning shell) 261~1907 by a sustained (uniaxial and nm release. coaxial) The comparison of Poly (L- Cefotaxi PLLA/ Structure: Absorbable Moderate Blend: initial (Hu, Huang, blend and coaxial lactic me CFX-Na Braided burst release and Liu 2010) electrospinning acid) sodium (wt%): techniques by (PLLA) (CFX- 8:3 (Blend) Diameter: Core-shell: braiding Na) Blend: sustained manufactured fibres PLLA/ 0.34 mm release and followed by CFX-Na Core-shell: coating with chitosan (wt%): 0.35 mm solution 8:30 (Core- shell)

Manufacturing Research topic Suture Bioactive The polymer Suture Solubility Tensile API release Reference Process material agent and/or drug structure/ strength profile amount Suture diameter The development of poly(lacti bupivaca PLGA/ Structure: Absorbable Low An accelerated (Weldon et al. a local anesthetic- c-co- ine bupivacaine Monofilam release profile 2012) eluting suture system glycolic hydrochl HCL (wt%): ent with PLGA sutures acid) oride 10/5 (PLGA) 10/10 Diameter: 10/15 70~300 10/22 μm The development of poly(lacti ibuprofe Ibuprofen/P Structure: Absorbable High Sustained drug (Lee et al. controlled pain relief c-co- n LGA (wt%): Core- release profile 2013) drug sutures by glycolic 0/100 sheath physically acid) 10/90 assembling sheet (PLGA) 15/85 Diameter: Electrospinning (contains PLGA and 380 ~ 420 (uniaxial and drug) with μm coaxial) commercially available suture (VICRYL® W9114,) The development of ⅰ. poly-L- Aceclofe Aceclofenac/ Structure: Absorbable Low Aceclofenac (Padmakumar core-sheath yarns, lactic nac and PLGA Core- loaded: burst et al. 2016) with PLLA as core acid insulin (wt%): sheath release and combination (PLLA) 50/50 Insulin loaded: (50:50) of PLGA ⅱ. poly Diameter: an initial burst and model drug as lactic-co- Insulin/PLG 200 ~ release and sheath glycolic A (wt%): 300 μm followed by a acid 3/97 sustained (PLGA) release

Manufacturing Research topic Suture Bioactive The polymer Suture Solubility Tensile API release Reference Process material agent and/or drug structure/ strength profile amount Suture diameter The development of Poly(ε- Nanohyb PCL/HT- Structure: Absorbable Low PCL/HT-Dic: (Catanzano et a controlled release caprolact rid of Dic/HT/Dic Monofilam controlled al. 2014) suture system by one) hydrotalc (wt%): ent release. The incorporating drug (PCL) ite and ⅰ. 75/0/8/17 complete release molecules in Diclofen ⅱ. 82/18/0/0 Diameter: happened after lamellar ac (HT- ⅲ. 75/17/0/8 300 μm 55 days hydrotalcite-like Dic) ⅳ. 82/0/0/18 compounds The development of poly(capr chlorhexi CHX/PCL Structure: Absorbable Low An initial rapid (Scaffaro et al. antimicrobial sutures olactone) dine (wt%): Monofilam release followed 2013) by melt spinning (PCL) diacetate 1/99 ent by a slow- different (CHX) 2/98 release Melt extrusion concentration of 4/96 Diameter: chlorhexidine 300±10 diacetate with PCL μm The development of polylacti Cytosine CpG ODN/ Structure: Absorbable Not Sustained (Intra et al. sutures with anti- c acid- – PLGA Monofilam investiga release over 35 2011) tumor properties by co- phosphor (wt%): ent ted days melt extruding glycolic othioate– 0.1/99.9 PLGA with acid guanine Diameter: CpG ODN (PLGA) oligonucl 0.6 mm eotides (CpG ODN)

Manufacturing Research topic Suture Bioactive Additional The Suture Solubility Tensile API release Reference Process material agent material polymer structure/ strength profile and/or Suture drug diameter amount Melt extrusion The development ⅰ. Nitric ⅰ. AN/VIM Structure: Absorbable High PCL coated (Lowe et al. of delayed- Acryloni oxide copolymer Monofilam sutures 2014) release suture trile-1- (NO) (wt%): ent exhibited a system by melt co- 82/18 significantly spinning nitric vinylimi Diameter: delayed release oxide (NO) dazole ⅱ. PCL/ 366 nm of NO compared incorporated (AN/VI chloroform with uncoated copolymer and M) (wt%): sutures followed by copolym 3/97 coating with PCL er ⅱ. PCL Coating The development Silk, Ibuprofe Poly(allyla - Structure: Non- Moderate An initial rapid (Wang, Chen, of anti- n mine Monofilam absorbable release in first and Sun 2009) inflammatory hydrochlor ent 4h, followed by suture system via ide) (PAH) a sustained layer-by-layer & Diameter: release for 10 depositing PAH Hyaluronic 391 ± 30 days and HA on silk acid nm surgical sutures, sodium then coated with (HA) ibuprofen

Manufacturing Research topic Suture Bioactiv Additiona The polymer Suture Solubility Tensile API release Reference Process material e agent l material and/or drug structure/ strength profile amount Suture diameter The ⅰ. silk, Chlorhe ⅰ. CHX Monofilamen Absorbable/ - - (Harnet et al. development of ⅱ. xidine Homopol concentration t/ non- 2009) antimicrobial polyeste (CHX) ypeptides : 20% (w/v) multifilament absorbable suture via layer r [poly-L- by layer ⅲ. glutamic depositing copolym acid homopolypeptid er of (PGA) es on surgical glycolid ⅱ. Poly-L- sutures, then e and L- lysine coated with lactide (PLL)] chlorhexidine The comparison PGA chlorhex - Chlorhexidin Monofilamen Absorbable - a (Matl et al. Coating of Vicryl Plus® idine e/ Octenidine t continuous 2009) and PGA and concentration release surgical sutures octenidi s (w/v): lasted for that were coated ne ⅰ. 2%, 5%, 96h with 15% chlorhexidine ⅱ. 10%, 21% and octenidine The development Silk Levoflox - The Structure: Non- High continuous (Chen et al. 2015) of antibacterial acin concentration Braided absorbable drug release surgical silk hydrochl of LVFX-HCl within 5−6 suture by coating oride in the coating Diameter: days silk strands with (LVFX- solution was 0.3~0.4 mm PCL and LVFX- HCl) 3000 μg/mL HCL

Manufacturing Research topic Suture Bioactive Additional The Suture Solubility Tensile API release Reference Process Material agent material polymer structure/ strength profile and/or Suture drug diameter amount The study of the Silk Tetracycli TCH Multifilame Non- Moderate A low initial rapid (Viju and characterisation of ne concentratio nt (braided) absorbable release followed Thilagavathi tetracycline hydrochlo ns (w/ by a sustained and 2013a) hydrochloride drug ride v): 0.1, 0.5, slow release over incorporated silk (TCH) and 1% a prolonged sutures period of time The study of the Silk Chitosan Chitosan Multifilam Non- High - (Viju and effect of different concentrati ent absorbable Thilagavathi concentrations of ons (w/v): (braided) 2013b) chitosan on silk 1%, 2% Coating surgical sutures and 3% The development ⅰ. Totarol PLGA - Monofilam Non- - Controlled (Reinbold et of antibacterial Resonlon ent/multifil absorbable release profile al. 2017) sutures by ®, 75 cm ament coating the USP 3/0 combination of ⅱ. totarol and Ethibond PLGA Excel, 75 cm, USP 3-0

Manufacturing Research topic Suture Bioactive Additional The Suture Solubility Tensile API release Reference Process Material agent material polymer structure/ strength profile and/or Suture drug diameter amount The development PGA octenidin Fatty acid: Drug / Monofilam Absorbable - Delayed drug (Obermeier et of antimicrobial e palmitic or fatty acid ent release and al. 2015) surgical sutures dihydroc lauric acid (wt%): lasted for 9 days by using hloride 20/80 components of 40/60 fatty acid and 60/40 octenidine as coatings The development PGA chlorhexi Fatty acid: Drug / Multifilam Absorbable - Octenidine (Obermeier et of antimicrobial dine palmitic or fatty acid ent coated lasted for al. 2018) sutures by diacetate lauric acid (wt%): (braided) 9 days. Coating coating PGA and 20/80 Chlorhexidine sutures with octenidin 40/60 coated lasted for chlorhexidine e 60/40 5 days and octenidine in dihydroc fatty acid carriers hloride The study of the Silk Silver - Multifilam Non- - - (Baygar et al. silk sutures that are nanopartic AgNO3 absorbable Solution: 50 ent 2019) coated by silver les (braided) nanoparticles, in mL (1 mM) order to avoid the microbial colonization on the sutures

1.5.1 Electrospinning

In the production of sutures, electrospinning is used to achieve porous nanofiber sutures. There are two types of electrospinning; uniaxial and coaxial electrospinning, which will be described below. An electrospinning device consists of a syringe, a high voltage emitter and a collector. The drug/polymer mixture is injected into the syringe and electrospun into nanofibers. This technique can fabricate suture materials into porous nanofibers with diameters from 5 to 500 nm (Champeau et al. 2017). The utilisation of a high electric field enables the formation of large surface areas (Champeau et al. 2017, Joseph et al. 2017). Even though the various advantages of electrospinning, electrospun nanofibers generally have poor mechanical performance, as they are in nonwoven form but generally, the thinner the fibre diameter, the higher the mechanical strength (Hu, Huang, and Liu 2010).

In uniaxial electrospinning, the drug migrates to the surface of the fibre and forms high concentrations, as shown in Figure 4 (a). This is commonly seen when small hydrophilic molecules are applied to the matrix. Consequently, an ideal controlled drug release cannot be achieved, as a burst release appears in this circumstance. In coaxial electrospinning, the drug is encapsulated inside the core of the fibre and surrounded by a polymer-formed shell to generate a core-shell structure. To achieve this, the drug and the polymer are loaded in separate syringes and the drug solution is injected into the centre of the polymer loaded capillary, as shown in Figure 4(b). This approach is particularly useful for immiscible drugs, organic solvent-sensitive or electric discharge-sensitive drugs. However, once the drug starts eluting, the tensile strength of the fibre tends to decrease, which is a drawback for wounds that require long-term support (Champeau et al. 2017).

Figure 4. (a) Uniaxial Electrospinning and (b) Coaxial Electrospinning

He et al. (2009) applied both blend and coaxial electrospinning processes to study antimicrobial drug-eluting sutures with tetracycline hydrochloride (TCH) and PLLA. In the blend electrospinning process, drug and polymer were dissolved in hexafluoroisopropanol (HFIP). In the coaxial electrospinning process, TCH and PLLA were the core and the shell respectively, which formed the core-shell structure (Figure 5). While the release profile of the blend electrospun sutures showed an initial burst followed by a sustained release, the core-shell sutures operated a controlled release. Additionally, the presence of the drug in both blend and coaxial electrospinning had a negligible effect on the mechanical properties of sutures, suggesting that the drug formed molecular interactions with polymers (He, Huang, and Han 2009).

Figure 5. Core-Sheath Structure

Most of the heat-sensitive compounds cannot be utilised with the melt spinning technique because the procedure uses high temperatures. Padmakumar et al. (2016) thus used the electrospinning technique to manufacture drug-loaded sutures. This approach can enable stable drug loading and a controlled drug release rate. However, electrospun fibres often demonstrate low mechanical properties. The fibre strength may improve with post-processing at high-temperatures, but the integrity of heat-sensitive drugs may be altered under these harsh conditions (Padmakumar et al. 2016). To avoid this problem, Padmakumar et al. (2016) utilised a PLLA core, and the drug-loaded PLLA sheath was electrospun onto the core. With this method, they were able to reduce the core diameter, maintain the drug function, and enhance the mechanical strength of the fibre (Padmakumar et al. 2016). The same research group has developed an electrospinning technique to produce a slow, controlled drug release system. They showed that nanofibers would form in the yarn, which caused the drug molecules to interact with the large surface area of the inter-fibres tightly, therefore, achieving a slowly controlled drug release system (Padmakumar et al. 2019).

Lee et al. (2013) demonstrated a novel drug-eluting suture system consisting of a drug- delivery sheet and a surgical suture. The sheet contained PLGA and was loaded with the drug Ibuprofen. The sheet was manufactured via electrospinning at a voltage of 15 kV, and was then physically braided and cured on the suture at 47°C. This system controls not only the drug delivery, but also maintains the mechanical strength. Moreover, tailoring the nanofibrous sheet is necessary for avoiding the impact on the mechanical strength of the suture and adjusting the drug release profile (Lee et al. 2013).

1.5.2 Melt Extrusion

Melt extrusion (Figure 6) is a widely applied technique that can manufacture monofilament sutures. This method has advantages compared to conventional preparation processes. For instance, the drug administration time can be reduced which then lowers the amount of drug usage for manufacturing (Li et al. 2013). Melt extrusion technique can disperse different API in a matrix at a molecular level, and can increase the solubility of poor water-soluble or hydrophobic compounds (Catanzano et al. 2014).

Figure 6. Melt Extrusion Process

Melt extrusion has wet and melt spinning. These techniques are simple to implement, and the APIs can be homogeneously distributed into the suture cross-section, achieving prolonged drug release (Tummalapalli et al. 2016). Moreover, where other techniques require solvents, this technique is solvent-free, which is more environmentally friendly and cheaper to produce. Besides, this one-step process may be scaled-up, making it easy to move from laboratory to commercial production. However, the drawback of this method is that often the increase in the drug content leads to the decrease in the mechanical properties of the suture. In comparison to wet spinning, melt spinning provides several advantages during suture

production. For instance, the filaments manufactured via melt-spun are usually without voids (Tummalapalli et al. 2016) However, it is generally accepted that high melting point polymers cannot be fabricated with thermolabile drugs (Champeau et al. 2017). Nevertheless, a previous study proposed that incorporating hydrotalcite-like compounds with APIs that have poor thermal stability can effectively maintain their physiochemical integrity (Catanzano et al. 2014). With this technique, melt extruding thermolabile drugs with polymers would become more common.

Studies looking into drug release found that the intercalation of API molecules in lamellar hydrotalcite-like compounds may obtain sustained drug release profiles (Catanzano et al. 2014, Perioli, Ambrogi, et al. 2011, Perioli, Posati, et al. 2011). For example, melt-spun PCL fibres which incorporated with diclofenac (Dic) were studied (Catanzano et al. 2014). In the Dic nanohybrid, the API was intercalated with a synthetic hydrotalcite to form hydrotalcite- like compounds (HT-Dic). The result demonstrated that the mechanical property and the drug release profile of the sutures largely depend on the Dic-loaded form. Furthermore, the combination of both Dic and HT-Dic even allowed further adjustment of the drug release rate, a release profile beyond 70 days was achieved (Catanzano et al. 2014). This study suggests a possible production of a more advanced suture that can achieve a sustained release of anti-inflammatory drugs by using hydrotalcite-like compounds.

The melt spinning technique was also used to fabricate PCL monofilaments incorporated with chlorhexidine diacetate (CHX) (topical anti-infective agent) (Scaffaro et al. 2013). Compared to PCL/HT-Dic loaded sutures, when the concentration of PCL/CHX was 4%, the sustained release of PCL/CHX sutures occurred for 6 days compared with the 70 days that PCL/HT- Dic achieved. An interesting observation was that even though CHX in pure form showed toxicity to human fibroblast cells, the PCL/CHX compound exhibited excellent cell viability and low toxicity, suggesting that the interaction between the polymer and CHX inhibits the potential toxicity that CHX may have in live tissues (Scaffaro et al. 2013).

The immunostimulant, Cytosine–phosphonothioate–guanine oligonucleotides (CpG ODN), was mixed and melt-extruded with polylactic acid-co-glycolic acid (PLGA). The collected sutures showed an intact structure in scanning electron microscopy images and reached diameters ranging from 50 to 300 μm. CpG ODN had a sustained release of up to 35 days. In a mice model, the mortality was significantly higher when CpG ODN-loaded sutures were

utilized for wound closure compared to the application of polyglycolic Acid (Dexon®) sutures, polyglactin 910 (Vicryl®) sutures or a combination of both (Intra et al. 2011).

The melt spinning method was employed to fabricate an acrylonitrile-co-1-vinylimidazole (AN/VIM) copolymer that incorporated nitric oxide (NO). In order to control the release rate of NO, the melt-spun monofilaments were dip-coated with poly(ɛ-caprolactone) (PCL) solution. The formed porous structure enables prolonged NO release from AN/VIM fibres up to 3 days, compared with 3 hours release of uncoated fibres (Lowe et al. 2014). As mentioned before, NO is effective as an anti-inflammatory coating. Thus, this method is able to apply NO to suture fabrication despite its limitations mentioned earlier.

1.5.3 Coating

The simplest approach to add API to a manufactured suture is coating suture surface with the desired drug (Champeau et al. 2017). It is relatively cheap and tensile strength of the suture is maintained. The drug release profile can be adjusted by controlling the property of the bioactive agent, such as its porosity, thickness and degradation rate. Coatings generally adhere to sutures through absorption, ionic interaction, and van der Waals forces. Those commercially available antimicrobial sutures such as Vicryl Plus®, PDS Plus® and Monocryl Plus® are all coated with triclosan. The property of triclosan produced sutures with smooth surfaces, which reduces tissue drag and immune reaction (Tummalapalli et al. 2016).

The dip-coating method is the most prevalent strategy which involves coating the suture in a prepared drug solution. Another coating method has also received attention known as layer- by-layer deposition. This coating method is similar as dip-coating, the difference is that the layer-by-layer method requires the suture to be dip-coated into the single API solution or multiple different API solutions for several times to achieve excellent antimicrobial effects (Dubas, Wacharanad, and Potiyaraj 2011). The suture simply absorbs the drug onto the surface, leading to sufficient drug loading and drug release efficiency (Lee et al. 2017). However, coating APIs on the suture surfaces may result in rapid drug release (Champeau et al. 2017).

Matl et al. (2009) used the dip-coating method to coat Pollyglactin-910 sutures with a combination of CHX and octenidine with palmitic and lauric acid acting as a carrier. The result was compared with commercial Vicryl Plus® and showed that even though the lipid-

based drug delivery system improved the surface smoothness and reduced the tissue drag during sewing, the experimental group demonstrated excellent antimicrobial activity, and the sustained drug release lasted long enough to achieve anti-infection. However, in vitro studies using L929 mouse fibroblasts showed that the cytocompatibility of modified sutures was inferior compared to Vicryl Plus® (Matl et al. 2009).

Wang et al. (2009) applied the layer-by-layer deposition method. One example is that a silk fibre was coated with poly(allylamine hydrochloride) (PAH) and dextran, along with hyaloplasm acid (HA), layer by layer to develop an anti-inflammatory suture system. The mixture was then soaked into an Ibuprofen solution. It was found that the sustained drug release lasted for 10 days in 0.9% normal saline (Wang, Chen, and Sun 2009). Another example is that braided natural and synthetic sutures, such as silk, polyester and PLLA sutures, were coated with the incorporation of polyelectrolyte and CHX multilayer films. It was observed that sutures with CHX-functionalized films exhibited a significant impact on preventing bacterial colonisation. Interestingly, there was no bacterial attachment on sutures for up to 7 days (Harnet et al. 2009).

There was a novel formulation proposed by Chen et al. (2015). They coated sutures with levofloxacin hydrochloride (LVFX-HCL) and poly (ɛ-caprolactone) (PCL) (Chen et al. 2015). The coating process was composed of two-dipping and two-rolling steps, which controlled the thickness of the sutures. The experiment involved four different groups of sutures, such as coated shell and strands, coated braided yarns, uncoated braided silk sutures, and PCL-coated sutures without LVFX-HCL. The result demonstrated that all study groups had the antimicrobial properties, and the group of LVFX-HCL and PCL coated shell and strands showed higher potential for clinical applications (Chen et al. 2015). In the following study, the same research group investigated the effect of the number of coating repetitions on suture characteristics. They found that the increase of the number of repetitions led to the decrease of knot-pull tensile strength and the bending stiffness, which can be attributed to the thicker diameter (Chen et al. 2015).

1.6 Conclusion

Polymers play a vital role in medical applications. Their high flexibility gives surgical sutures excellent physical and mechanical properties. Recently, biodegradable polymers have

received much attention, since they can be degraded inside the human body through proteolysis degradation or hydrolytic degradation without post-surgery extraction, which significantly reduces the chance of surgical infection and prolonged patient suffering.

It is widely accepted that within the drug-eluting suture process, that adding API during manufacturing is the most ideal, because it is a one-step process and does not require much equipment, which can reduce cost and allow mass production.

Melt extrusion processes should be employed whenever conditions are applicable. As it does not require a solvent removal step, and it meets environmentally friendly requirements. However, in some cases, a thermolabile API is required for the drug-eluting suture applications. The high temperature in the melt extrusion process would destroy the structure of the thermolabile drug. Thus, electrospinning is a better choice.

Even though both melt extrusion and electrospinning can manufacture drug-eluting sutures with a drug content up to 20%, the higher drug amount leads to a lower tensile strength of the sutures. However, in some suture applications, high drug content is required. In order to maintain the tensile strength and meet the high drug amount requirement, coating can be applied.

During the drug-eluting suture engineering, the tensile strength of the sutures has been widely studied. However, detailed mechanical properties during surgery such as knot-pulling strength are rarely studied by researchers, which requires more attention in future studies.

2 THE CURRENT PROJECT

2.1 Fabrication of Drug-Eluting Sutures with Hot- melt Extrusion Technique

At present, there are no studies that have investigated the biopolymer blend of PCL/PEG400/chitosan/keratin for surgical suture applications, in particular, using diclofenac potassium, as the target drug. Brianezi et al. used PCL/PEG/chitosan to control the release rate of gentamicin for skin wound healing (Brianezi et al. 2018). This medical device demonstrated the ability of inhibiting the growth of E. coli and S. aureus. Another study extruded PLA/PCL blend with Ethyl Ester L-Lysine Triisocyanate. The results showed that the higher drug loads resulted in decreased suture ductility (Visco et al. 2019). Usually for absorbable synthetic suture studies, researchers focus more on PLA-based sutures (He et al. 2014, Tyler et al. 2016, da Silva et al. 2018, Liu et al. 2019). Furthermore, the recruitment of the hot- melt extrusion technique for surgical suture manufacturing has not obtained much attention (Catanzano et al. 2014). Most researchers focus on suture coating to achieve outer layer drug release (Baygar et al. 2019, Chen et al. 2015, Obermeier et al. 2015, Reinbold et al. 2017). However, using this coating technique, it is difficult to control the drug release rate, due to the simple attachment of the drug onto suture surfaces (Lee et al. 2017).

This study was designed to explore the novel biomaterials of PCL, PEG400, chitosan and keratin mixed in various ratios with the anti-inflammatory drug diclofenac potassium. These were extruded using the hot- melt extrusion method. The hot- melt extruder was used for dissolving the biopolymers and the drug, creating an amorphous solid dispersion state, which ensured that the biopolymers and drug were mixed at a molecular level to achieve homogeneity. The high temperatures during the hot- melt extrusion may be seen as a limitation, as most drugs are not stable under high temperatures. Therefore, a low extruding temperature is essential. The most suitable biopolymer was PCL, due to its thermoplastic property with a low melting point and glass transition temperature (Dash and Konkimalla 2012a). This study explored the potential of PCL, PEG400, chitosan and keratin to serve as

a more complex and robust absorbable surgical suture, which can also reduce the surgical site infection and inflammation.

2.2 Materials Selected for This Study

Poly(ɛ-caprolactone) (PCL) has been widely utilized in biomedical applications such as controlled release drug delivery systems and absorbable surgical sutures (Malikmammadov et al. 2018). It is a synthetic biodegradable polyester which with a molecular weight ranging from 3,000 to 80,000g/mol. This material has a low melting point of 59-64°C, and the glass transition temperature is 60°C (Dash and Konkimalla 2012a). PCL is an ideal candidate for melt processing, due to its biocompatibility and biodegradability (Malikmammadov et al. 2018). It is also one of the biopolymers that has been approved to be used in the biomedical application by Food and Drug Administration (FDA) (Azimi et al. 2014).

According to previous studies, PCL is compatible with a wide range of pharmaceutical ingredients, which allows drug distribution uniformly throughout the polymer matrix. Moreover, its long-term degradation property enables the drug release to last for several months (Dash and Konkimalla 2012a). Diclofenac potassium is a nonsteroidal anti- inflammatory drug, and has analgesic, antipyretic and anti-inflammatory activity, is poorly water-soluble, and served as the model drug. The drug content was 5, 15 and 30 wt% in the polymer matrix. Additionally, the versatility of PCL allows it to be blended with many other polymers, such as PEG (Grossen et al. 2017), to tune the physical, chemical and mechanical properties of devices.

Polyethylene glycol (PEG), has a high solubility in water. Due to its high hydrophilicity, it has been used to improve the hydrophilicity of block copolymers, and enhance water uptake and porosity, as well as to modify the degradation properties of copolymers (Stanković et al. 2013, Stanković et al. 2014). During the melt extrusion process, PEG has been successfully utilized as a plasticizer to improve the solubility of a drug in oral administration (Stanković, Frijlink, and Hinrichs 2015). Therefore, the combination of PCL with natural and/or synthetic polymers may yield different functionalities.

PCL is also compatible with natural polymers like chitosan and keratin (Tan et al. 2015). Chitosan is a polysaccharide found in crab shells and which can be obtained from chitin through the process of deacetylation (Dash and Konkimalla 2012a). It has been applied to

many biomedical purposes due to the relatively cheap price, antimicrobial effects, biodegradability and non-toxicity (Tanase and Spiridon 2014). Longer polymers of chitosan (with high molecular weight) may be employed as matrix tablet retardants, while shorter polymers of chitosan (with low molecular weight) have the ability to improve the water solubility of active ingredients in drug release systems (Wolf 2010).

PCL has also been used to improve the mechanical properties of keratin, and keratin by itself is easily degraded and unable to retain structural integrity in aqueous environments alone (Edwards et al. 2015). Keratin is the main component of feathers. It contains high quantities of cystine and hydroxyl amino acids such as serine (Rouse and Van Dyke 2010). As such, keratin offers a wide range of noncovalent and covalent interactions, which makes this natural polymer difficult to damage when paired with a polymer (Tanase and Spiridon 2014).

In light of the literature findings, a ratio of 50:50 by weight of chitosan/keratin (Tran and Mututuvari 2015, Vasconcelos and Cavaco-Paulo 2013, Tran, Duri, and Harkins 2013, Sando et al. 2010) was selected for this study, as this ratio was seen to be optimal in enhancing the properties of the desired suture (Tran and Mututuvari 2015). Furthermore, PCL and PEG were added to the blend as well so that the suture has strong basic properties, such as increasing solubility of poor water-soluble drug from adding PEG400 (D’souza and Shegokar 2016), and improving ductility of the sutures due to the addition of PCL (Dash and Konkimalla 2012b). Thus, a suture was designed that may have a low enough melting point to be produced by melt extrusion method, have low manufacture cost, and have additional properties of having a great drug carrying capacity.

Our hypothesis was that the suture made from PCL/PEG400/chitosan/keratin/ drug not only has excellent physicochemical, mechanical and biological properties, but also can exhibit controlled drug release.

2.3 Aims and Objectives

The overall aim of this study was to develop an absorbable drug-eluting surgical suture using a blend of PCL/PEG400/chitosan/keratin and diclofenac potassium (DP) drug. The aim was to melt extrude biopolymers and drug DP with the potential for applications in reducing surgical site infection and inflammation.

The study design includes: 1) fabricating biopolymer sutures via the hot- melt extrusion method; 2) characterising the sutures’ physical microarchitecture, chemical makeup, thermal and mechanical properties; 3) manufacturing of biopolymer sutures which incorporate the drug through the hot- melt extrusion technique; 4) determining drug release profile; and 5) investigating the cytotoxicity of the drug-eluting sutures (Figure 7).

To achieve this aim, the primary objectives were as follows:

1) Optimize the parameters of hot- melt extrusion techniques for obtaining strong and uniform sutures with/without the drug.

2) Characterise the physical, chemical and mechanical properties and thermal properties of the biopolymer sutures that may provide excellent tensile strength for surgical applications.

3) Determine the drug dissolution profiles with drug embedded sutures.

4) Asses the in vitro biological activity of the drug-eluting sutures.

Figure 7. Summary of The Study Design

The objectives are from (1) to (5). (1) Fabrication of sutures that contain polymers only. (2) Characterisation of melt-extruded sutures with FTIR, TGA. DSC and tensile testing methods. (3) Fabrication of sutures that contain both polymers and model drug. (4) Drug dissolution of sutures is conducted in PBS solution at pH 7.4. (5) Biocompatibility of sutures contain model drug is investigated through MTT, Live/Dead and cell migration assays.

3 MATERIALS AND METHODS

3.1 Materials

The main material of the sutures was the medical-grade polycaprolactone (CAPPA® 6506; CAS no: 24980-41-4, Perstorp Holding AB, Sweden). This material is a thermoplastic homopolymer with an average molecular weight of 50kDa with a melting point of 58-60° and a melt flow index of 11.3-5.2. The second material used was colourless liquid poly(ethylene glycol) (Kollisolv® PEG E 400; no.: 25322-68-3, Sigma-Aldrich, USA) with a molecular weight of 400 g/mol (PEG400), Tg= -20 °C, Tm= 4 °C. Keratin powder (hydrolysed, code: 69430-36-0) was supplied by Qingdao Aurora Chemical Co., Ltd., China (Mainland). Chitosan (degree of deacetylation, 0.91; Mw=8.9×104 Da) was purchased from Weseta international, Shanghai, China. Diclofenac potassium (Voltaren®) was purchased from Countdown pharmacy, New Zealand. 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Invitrogen, New Zealand) was used to investigate cytotoxicity of suture materials and diclofenac potassium drug. Dulbecco Modified Eagle Medium phenol red media (Gibco™, Thermo Fisher Scientific, New Zealand), Dullbecco’s phosphate buffered saline (DPBS, Gibco™, New Zealand) and Trypsin-EDTA (Gibco™, Life Technologies Corporation, New Zealand) were used for cell media, and cell detachment, respectively, during cell culture work. For live/ dead assay, calcein blue AM (CAS no.: 54375-47-2, Sigma-Aldrich, USA) and propidium iodide (Cat no.: 421301, BioLegend, USA) were used for fluorescence observation.

3.2 Equipment

Laboratory designed mini melt-extruder (Billaa-Tex®, TX, USA) was used to prepare sutures that were with and without the drug. Differential scanning calorimeter (DSC) Q2000 (TA Instruments, MA, USA) and thermogravimetric (TGA) analyser Q50 (TA Instruments, MA, USA) were employed to estimate the thermal properties of suture products. Water bath (PolyScience, Illinois, USA) was applied to keep the temperature at 37.4 ± 2 °C for the drug dissolution test. UV-vis spectrophotometer (Ultrospec 3300 pro, Amersham Bioscience, TX, USA) was required for drug concentration analysis. Bio-Rad Benchmark Plus plate reader (,

USA) was used to measure the absorbance of cells in 96-well plates. Fourier Transform Infrared Spectrometry (FTIR) analysis was used to determine the molecular bonds in the selected polymers and carried out using an Alpha II Fourier Transform Infrared Spectrometer (Bruker Optik, Germany). The FTIR absorbance spectra were obtained using OPUS software (version 5.0, Bruker Optik, Germany). HaCat cells growing in the presence of melt-extruded sutures were imaged using an optical Olympus IX71 Inverted Microscope (Olympus Corporation, Japan), and the images were analysed using the ImageJ v1.52 software package (National Institute of Health, USA). GraphPad Prism 8.3.1 software (USA) was used for graphical and statistical analysis of cytotoxicity assays.

3.3 Suture Preparation for Control Group

Detailed fabrication procedure is shown as the following steps:

i. Moisture absorption: before starting the fabrication process, PCL, keratin and chitosan were all placed in a desiccator for absorbing moisture of the polymers.

ii. Five control groups were prepared by pre-weighing 10 g of the mixture of PCL, PEG, keratin and chitosan at different ratios (Table 3). Firstly, keratin and chitosan were mixed at (50:50) and extracted a certain amount that was required by each formulation. Secondly, PCL was added into the keratin-chitosan blend, followed by dripping the required amount of PEG into the mixture. Finally, the polymer compound was graded thoroughly (Figure 8).

iii. The five control groups were then prepared by melt extruding in a mixing extruder (Figure 9). During the feeding process, the pressure was increased by pushing the plug, allowing the material to pass through the feeder and enable the screw that was inside the body of the extruder to contact the material fully (Figure 10). Melt blending was performed at 60±1, 63±1, 66±1, 69±1, 72±1, 75±1 °C for 5-20 min while the roller speed remained unchanged. The extrudates from the die were physically drawn into fine fibres. All samples after preparation were placed at room temperature (25 °C).

Table 3. Control Group (C) (weights/ 10 g in total)

Control groups, including PCL only; PCL: PEG400 at a ratio of 80:20 wt% PCL: PEG400 and chitosan-keratin blend mixed together at ratios of 80:19:1, 80:18:2, 80:16:4 wt%.

C PCL (g) PEG400 (g) Chitosan-Keratin (g) (50:50) C1 8 - - C2 8 2 - C3 8 1.9 0.1 C4 8 1.8 0.2 C5 8 1.6 0.4

Figure 8. Suture Material Preparation without Drug

Keratin and chitosan were mixed together with a grinder at a weight ratio of 50:50, followed by blending with PCL and PEG before feeding into the extruder.

Figure 9. Mixing Extruder

A mixing extruder is composed of a feeder, a rotator, a plug and a temperature controller. Motor button only can be turned on once the device has reached the operating temperature.

Figure 10. Extruding Process

During suture fabrication, the plug was placed into the feeder. One hand had to press down the handle to increase the surface contact between materials and the screw inside the rotator. The other hand was used to slowly pull the extrudate, making it into a thin fibre shape.

3.4 Fourier Transformation-Infrared Spectroscopy (FT-IR)

Figure 11. ALPHA FTIR (Bruker)

In order to confirm the existence of interactions between the API and the carriers, each extrudate was pressed into a thin film and calibrated to background spectra. The FTIR spectra was obtained by measuring a wavelength ranges from 650 to 4000 cm-1 with a resolution of 4 cm-1. The final spectrum was the mean of 20 scans. The ATR spectra were converted to absorbance spectra using OPUS software (version 5.0, Bruker Optik, Ettlingen, Germany).

3.5 Thermal Gravimetric Analysis (TGA)

Figure 12. TGA Q50

Thermogravimetric analyser (TGA Q50, TA instruments, MA, USA) was utilized to study the chemical stability of the carriers and API at elevated temperatures. Samples were heated from 20 °C (room temperature) to 800 °C at 10 °C/min and held at an isotherm for 3 min (Mohamed et al. 2008). The chambers were flushed with nitrogen at a flow rate of 30 ml/min during the testing.

3.6 Differential Scanning Calorimetry Analysis (DSC)

Figure 13. DSC Q2000

DSC can be utilized to study drug-polymer interactions and microparticles as it provides information such as melting point, enthalpy of crystallization and glass transition temperature (Chen et al. 2016).

Thermal properties of the extrudates were characterized by using Differential Scanning Calorimetry (DSC Q2000, TA instruments, MA, USA). The instrument was calibrated with high purity aluminium standards. Samples (5-10 mg) were sealed in aluminium pans with a pierced lid. A constant nitrogen flow of 30 ml/min was maintained to provide a constant thermal blanket within the DSC cell, thus eliminating thermal gradients and ensuring the validity of the applied calibration standard from sample to sample (Palazi et al. 2018, Li, Guo, Fan, et al. 2013). Samples were equilibrated for 1 min at 0 °C after which the temperature has increased to 200 °C at 10 °C/min, isotherm was maintained for 1 min, and the sample was cooled to -70 °C at the same rate, isotherm was maintained for 1 min. The sample then was reheated from -70 °C to 0 °C at 10 °C/min (Mohamed et al. 2008).

3.7 Mechanical Properties

Figure 14. Tensile Analyser

Before the testing, the diameter of each suture sample was measured three times using a digital calliper and a mean diameter was calculated and entered in the Exponent Connect Software v7.0 (Stable Micro Systems, UK) for data analysis. The tensile properties of straight sutures were investigated with a 50 kg load cell, and at a strain rate of 100 mm/min, using a gauge length of 15 mm, at 23 °C. The grips were set to a distance such that the suture material was taut, with minimal force (<0.1 N) applied before testing began. For each sample, at least five specimens are analysed. If the fibre broke in the clamp, the test was repeated. If the fibre slipped out of the clamp during testing, that sample was not included in the calculations of mean tensile failure load. If the rupture of sutures happened close to the centre of the active length, which indicates that the two points that were griped in the clamp had no effect on the results. The average results are reported with their standard deviation (Scaffaro et al. 2013, Marturello et al. 2014, Naleway et al. 2015, Hossain et al. 2014, Catanzano et al. 2014).

3.8 Suture Preparation for Formulation Group

Suture preparation for formulation group was conducted using the following steps:

i. Moisture absorption: Before starting the fabrication process, PCL, keratin and chitosan were all placed in a desiccator to absorb the moisture of polymers.

ii. Nine formulation groups were prepared by pre-weighing 10 g of the mixture of PCL, PEG, keratin and chitosan in different ratios shown in Table 4 (Figure 15): Firstly, 3 groups that contained PCL/PEG400/chitosan-keratin, PCL/PEG400/chitosan, and PCL/PEG400/keratin were prepared (Figure 16). For group 1, keratin and chitosan (50:50) were blended together by a certain amount that was required for each formulation. PCL (as white powder) was added into the keratin-chitosan blend, followed by dripping a certain amount of liquid polymer PEG into the mixture and then grinding the polymer compound thoroughly. The preparation of group 2 and group 3 followed similar instructions. Secondly, the drug tablets were manually ground into fine powder in a grinder. To achieve 5, 15 and 30 wt% of drug content, the ratio of polymer blend and drug were set as 95:5, 85:15, and 70:30 wt%. Briefly, a certain amount of polymer compound (1.9 g, 1.7 g or 1.4 g) was extracted, mixing with the required amount of drug powder (0.1 g, 0.3 g or 0.6 g), resulted in a total amount of 2 g (Table 4). Finally, the mixture was grounded thoroughly and fed into the extruder.

iii. The nine formulation groups were then prepared by melt mixing in a mixing extruder (Figure 9). During the feeding process, the material mixture was pushed by the plug to increase pressure, which allowed materials to pass through the feeder and enabled the screw to fully contact the materials. Melt blending was performed at 63±1°C for 5-20 min with no change in roller speed. The extrudates from the die were physically drawn into fine fibres. The extruding order was group 2, F4, F5, F6, group 1, F1, F2, F3, group 3, F7, F8, F9. The reason that extruding groups 1, 2, 3 were placed between different polymer groups was to eliminate the possibility of introducing unwanted material into the desired polymer group. All samples after preparation were placed at room temperature (25 °C).

Table 4. Formulation Group (F) (weights/ 10 g in total)

According to the characterisation study of the control group sutures, PCL: PEG400: chitosan/keratin= 8:1.9:0.1 (Group 1) was confirmed to be the ideal weight ratio group. PCL/PEG/chitosan (Group 2) and PCL/PEG/keratin (Group 3) were designed to compare with Group 1.

F PCL (g) PEG400 Chitosan- Chitosan Keratin (g) Diclofenac (g) keratin (g) (g) (g) (50:50) F1 8 1.9 0.1 - - 0.1 F2 8 1.9 0.1 - - 0.3 F3 8 1.9 0.1 - - 0.6 F4 8 1.9 - 0.1 - 0.1 F5 8 1.9 - 0.1 - 0.3 F6 8 1.9 - 0.1 - 0.6 F7 8 1.9 - - 0.1 0.1 F8 8 1.9 - - 0.1 0.3 F9 8 1.9 - - 0.1 0.6

Figure 15. Suture Material Preparation with Drug

Keratin and chitosan were mixed together at a weight ratio of 50:50, followed by blending with PCL and PEG. Finally, certain amount of drug was added into mixture and blended thoroughly.

Figure 16. Groups 1, 2, 3 Mixtures

The mixtures were in white powder forms. All the mixtures tended to form lumps, due to the presence of liquid polymer PEG400.

3.9 Drug Release Studies

3.9.1 Standard Preparation

To begin the experiment, phosphate-buffered saline (PBS) solution was prepared by pouring 1 packet of dry powder of phosphate-buffered saline into 1000 ml of distilled water. After stirring with a magnetic stirrer, the PBS was completely dissolved in the water with pH of 7.4 ± 0.2. For calibration purposes, 100 mg drug was dissolved in 100 ml PBS solution with pH 7.4 and this served as the stock solution. From this stock solution, 5 ml was pipetted into a 10 ml volumetric flask and diluted up to 10 ml with PBS. Followed by taking 5 ml solution out of the 10 ml volumetric flask and diluted to another 10 ml volumetric flask. The aforementioned process was repeated for 3 more times to obtain a sequential dilution with concentrations of 0.0312, 0.0625, 0.125, 0.25, 0.5 mg/ml respectively. The dilutions were analysed for absorbance at 276 nm using the UV-Visible spectrometer. The curve generated by concentrations and wavelengths exhibited behaviour over the full range of concentrations.

3.9.2 In vitro Drug Release Study

For the drug release study, nine different types of sutures were immersed in nine 50 ml falcon tubes which contained 30 mL 0.01 M of PBS buffer. Each tube contained sutures of 2 g based on Table 4. The reference tube contained sutures extruded from all the polymers (i.e. PCL, PEG400, keratin and chitosan). Tubes were immersed in a water bath (Daihan Scientific WCB-11) (Figure 17) for various periods of time under gentle stirring. The temperature of the solution was maintained at 37.4 ± 2 °C at all times.

The released media (5 mL) was extracted and replaced with the same volume of fresh PBS at time intervals of every 0.25, 0.5, 1, 2, 3 hr (i.e. 15mins, 45mins, 1h 45mins, 3h 45mins, 6h 45mins). The samples were analysed via measuring the absorbance at 276 nm wavelength with a UV-vis spectrophotometer (Ultrospec 3300 pro, Amersham Bioscience) (Figure 18) and then compared to the predetermined calibration curve to obtain concentrations. The reference cuvette contained PBS solution and the extracted media from the reference tube.

Figure 17. Suture Immersion

Nine 50 mL centrifuge tubes were placed in a water bath. Each tube contained 2 g of sutures

Figure 18. UV-vis Spectrophotometer

3.10 Biocompatibility Tests

3.10.1 Cell Lines and Maintenance

Human keratinocytes cell line (HaCat ATCC® CRL-2404) are the one of the most abundant cells in adult human skin, widely used in scientific research. HaCat was used as a model cell line to evaluate the cytotoxicity of the drug-embedded polymer sutures. Frozen stocks of HaCats were extracted from liquid nitrogen tank and then quickly warmed up in a water bath under temperature of 37°C (Figure 19). Cell proliferation procedure was conducted in a Herasafe 2030i biological safety cabinet (Thermo Fisher Scientific, USA) (Figure 20), which was disinfected with UV light and 70% ethanol.

Complete DMEM (cDMEM) was prepared with DMEM red media (pH:7.4) enriched with 10% foetal calf serum (FCS, Gibco™, Thermo Fisher Scientific, Waltham, MA, USA) and 1% penicillin and streptomycin (PenStrep, Sigma-Aldrich). Thawed HaCat cell line was then 2 transferred to a 75 cm flask containing cDMEM. The flask was kept inside a CO2 incubator ® ® (Heraeus Heracell ) (Figure 21), set at 37 °C with a 5% CO2 (95% air) atmosphere for a week, the media was changed once every three days. Once cells were passaged at 80% confluency, the media was removed and cells were washed with Dullbecco’s phosphate buffered saline (DPBS, Gibco™), and adding 3 mL Trypsin-EDTA (Gibco™, Life Technologies Corporation) to detach cells. Please note that the Trypsin-EDTA requires to be warmed in a water bath (Figure 19) at 37°C before using to prevent shocking cells. The flask was then incubated for 6 mins to allow for the trypsinization of the cells. Followed by inactivating trypsin with 6mL cDMEM. The solution was then removed to 15 mL tubes and the cell number was counted using a handheld automated cell counter (Scepter™) (Figure 22). Solution (1 mL) from the flask was re-subcultured in each new flask containing 25mL cDMEM. According to the cell number requirement for MTT, live/dead, scratch assays, the rest of cells were seeded into suture-placed plates for experiments as required.

Figure 19. Thermo Scientific Liquid Nitrogen Tank & Water Bath

Frozen stocks of HaCat cells were stored inside the nitrogen tank. To defrost the cell line, tubes needed to be immersed in the water bath at 37 °C.

Figure 20 Herasafe KS Class П Biosafety Cabinet

Complete DMEM solution preparation was conducted inside the biosafety cabinet with a sterile filter.

Figure 21 CO2 Incubator & Flasks of Complete DMEM

After completing cDMEM preparation, all the flasks were stored inside the CO2 Incubator.

Figure 22. Handled Automated Cell Counter

To ensure the accuracy of the results, handled automated cell counter was used to count the number of cells.

3.10.2 Cell Viability in Response to Suture Materials

Estimation of suture cytotoxicity was performed by direct contact of HaCat with polymer- drug suture materials. To prepare the samples, the fabricated polymer suture and polymer- drug suture were cut into 1 cm length (Obermeier et al. 2015, Umair et al. 2015). Washed with distilled water and sprayed with 70% ethanol. All these samples were exposed to ultraviolet light for 1 h before the addition of the cells to minimize contamination (Adomaviciute et al. 2018).

Cell viability of HaCat in response to sutures was determined using the MTT (3-(4,5- dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) tetrazolium reduction assay. This colorimetric assay determines the cell number by measuring the amount of purple formazan crystals formed when mitochondrial reductases (present in viable cells) reduce the yellow dye, MTT. This study was conducted in a 96-well plate. For formulation group and control group, 5 x 103 cells/ml were seeded into each well. Standard curve was ranged from 2.5 to 80 x 103 cells/ml cells (Figure 23). All the cells were left in the plates for 24 hours, attaching to the bottom of the each well. Cells were then treated for 24, 48 and 72 hours in the presence of sutures that are shown in Table 5 and Table 6. Each cultivation was performed in triplicate. Stock MTT was prepared at 5 mg/ml in PBS, 600 µL MTT stock was then extracted and diluted with 3 mL cDMEM colour free to achieve a final concentration of 0.4 mg/mL. After this, media was removed and 300 µL MTT solution was added into each well. Followed by incubating plates for 4 hours. After that, the MTT solution was extracted and the formazan crystals were left in plates. To dissolve the crystals, 100 μL of 10% SDS/1M HCL (100:1) were added to each well. After that, the plates were incubated at 37 °C overnight in the CO2 incubator. Absorbance results were obtained at a wavelength of 620 nm with Gen5 software (BioTek Instruments), using a UV-Vis plate reader (BioTek®, Synergy™ 2) (Figure 24)

Table 5. Control Group

To investigate the effect of polymers on the cell proliferation, pure PCL, PCL/PEG and PCL/PEG/chitosan/keratin extrudates were designed as control group.

C PCL (g) PEG400 Chitosan- Chitosan Keratin (g) Diclofenac (g) keratin (g) (g) (g) (50:50) C1 8 - - - - - C2 8 1.9 - - - - C3 8 1.9 0.1 - - -

Table 6. Experimental Group

Group 1, 2, 3 with various drug content were designed as experimental group.

Figure 23. Plate Design for MTT Assay

Experimental group, control group, and standard curves were assembled in two 96-well plates. Adding suture only group was to confirm whether the MTT solution would make colour change of sutures.

Figure 24. UV-Vis Plate Reader

3.10.3 Fluorescence Microscopy

Biocompatibility of the sutures was analysed in triplicate using the live/dead assay. Fluorescence staining was applied to observe the cells that were under direct contact with melt-extruded sutures. In order to investigate the number of dead cells and live cells at time intervals of 24, 48 and 72 h. Coverslips were immersed in 70% ethanol before placing into 24-well plates. Sutures were cut into 1 cm length for each well. Each well contained 500 µL DMEM media and 5 x 103 cells/ml cells (Figure 25). HaCat cells were treated for 24, 48, 72h through direct contact tests. After the treatment, a solution containing 4.5 µL calcein-AM (acetomethoxy derivate of calcein) and 18 µL ethidium homodimer-1 in 3 mL PBS was made. For some samples, 30 µL of the solution was pipetted onto a glass slide and covered with a coverslip, the side that contained cells faced the slide. For the rest of samples, the 30µL of the solution was pipetted onto each coverslip in the 24-well plates. Afterwards, images were taken using an IX71 inverted fluorescent microscope (Olympus) using a 10x objective (Figure 26). The analysis was done by using ImageJ software. The number of live or dead cells were quantified according to the following Eq. (1):

Percentage (%) = Live cells / (Live cells + Dead cells) x 100%

Figure 25. Plate Design for Live/Dead Assay

Formulation group, control group and cell only were seeded in 24-well plates.

Figure 26. Olympus IX71 Inverted Microscope

3.10.4 Wound Healing in Response to Suture Materials

The scratch assay (injury to the cell monolayer) was performed in triplicate, to quantify the spreading and migration capabilities of cells exposed to the samples. HaCat cells were seeded 3 in a 48-well plate at 40 x 10 cells/ml per well (Figure 27) and incubated in 5% CO2 at 37 °C for 24 h. After the cells grew to over 90% confluence, a straight scratch was created with a sterile 100 µl pipette tip. Afterwards, group 1 sutures were cut into 1cm length and placed into wells. Only group 1 with the formulation of PCL/PEG400/chitosan/keratin was chosen to be the formulation group due to its better performance in MTT and live/dead assays. Control group contained 3 suture groups: PCL only, PCL/PEG400 and PCL/PEG400/chitosan/keratin. Cells cultured with DMEM alone served as a control. Images of the closure scratch were captured at 0, 6 and 24 h using a Ceti Magnum Trinocular compound microscope (Figure 28). Before capturing the images, a dot was drawn on the bottom of each well to make the scratch area easy to find. The residual gap between the migrating cells was measured and expressed as percentage migration rate using the following Eq. (2):

Migration rate (%) = (Scratch width day 0 – Scratch width day 1) / Scratch width day 0 x 100%

Figure 27. Plate Design for Scratch Assay

Experimental group and control group are assembled in a 48-well plate.

Figure 28. Ceti Magnum Trinocular Compound Microscope

4 RESULTS

Before mixing the drug with the biopolymer materials to study the drug release profile, it was important to estimate the biopolymer sutures at the molecular level, and to investigate their thermal properties and mechanical properties. The overall aims were, 1) To optimise polymer formulations with hot- melt extrusion technique to provide a robust polymer matrix in which to embed drug particles. 2) To achieve solid dispersion of the polymer-carriers and drug with the hot- melt extrusion technique. 3) Optimise drug/polymer formulations using drug dissolution and biocompatibility assays. For these reasons, a wide range of polymer and drug ratios were investigated. The specific objectives were:

1. Polymer suture fabrication.

2. To investigate the characterisation of the biopolymer materials, using FTIR, DSC, TGA and tensile analyser.

3. To optimise the drug/polymer suture fabrication processes.

4. To investigate the drug release profile using a drug dissolution test.

5. To investigate the drug/polymer in vitro cytotoxicity, by using the MTT, live/dead and scratch assays.

4.1 Suture Preparation for the Control Groups

4.1.1. Melt Extruder

It is important to note that the only method to clean the melt extruder is to extrude the acquired material until the machine only contains the material that is wanted. For the machine cleaning purpose, PCL was used to purge the machine. At first, since the remaining material inside the device was unknown, the purging temperature was set higher than the melting point of PCL (80 °C), but nothing came out of the nozzle. The temperature was then gradually increased to 96 °C, 117°C and 150 °C. In the temperature increasing process, blackish material came out from the two holes that were on the connective cylinder of screws. At 150 °C, the

extrudate was entirely dark in colour. Once the machine was cool, the material would fix the screw and the machine could not rotate. Therefore, it required frequent cleaning during the purging process.

4.1.2 Melt Extrusion at 63±1 °C

To remove the residual material from previous the usage, the extruder was purged with pure PCL for 10 minutes, and the extrudates began to appear transparent and colourless. When the extruder contained only the required polymer, PCL, the temperature was adjusted the to reach 60±1°C followed by gradually increasing temperature to 63±1 °C, 66±1 °C, 69±1 °C, 72±1 °C, 75±1 °C. When the temperature reached 63±1 °C, the polymer powder was melted at the expected rate, which allowed the collection of fibres with an even surface. However, when materials were heated at 75±1°C, they became too liquid to collect into the fibre form.

The components of each control group are shown in Table 7. Extrudates were all obtained as fishing line-like fibres. The extrudates of pure PCL were soft, clear and transparent when coming out of the nozzle, turning to white solid fibres after being exposed to air for half a minute. The extrudates of C3 (0.6 mm), C4 (0.6 mm) and C5 (0.9 mm) were thicker than Control groups C1 (0.5 mm) and C2 (0.4 mm). The three groups (C3, C4, C5) were also not homogeneous, and had lumps on the fibre surfaces which were hard to notice without touching. It was difficult to obtain long fibres, the filament of C3 would break when the length reached about 38 cm (Figure 29). Both C4 and C5 extruded relatively long fibres (Figure 30), especially C5, which contained higher amounts of chitosan/keratin, extruded with fibre length of 186 cm (Figure 31).

Table 7. The Physical Observations of Extruded Polymer Fibres

Control group was melt-extruded at the temperature of 63±1 °C.

Keratin- PCL PEG400 Thickness C Chitosan(g) Physical properties (g) (g) (mm) (50:50) Homogeneous with high ductility; C1 8 - - 0.5 white C2 8 2 - 0.4 Homogenous; grey C3 8 1.9 0.1 0.6 Inhomogeneous; grey C4 8 1.8 0.2 0.6 Inhomogeneous; grey More stable extruding rate than C3 C5 8 1.6 0.4 0.9 or C4; inhomogeneous; white

Figure 29. Extrudate of Sample C3 at 63±1 °C with Length of 38 cm.

Figure 30. Extrudates of Sample C4 & C5 at 63±1 °C

Extrudate C5 demonstrated a more brittle physical property than extrudate C4, attributed to the higher content of chitosan/keratin.

Figure 31. Extrudate of Sample C5 at 63±1 °C with Length of 186 cm.

Figure 32. Extrudates of C5 at Variable Temperatures

When T= 63±1 °C, the extrudates were fine and extruded with stable extruding rate. When T= 66±1 °C, the extruding rate of extrudates became unstable. When T= 69±1, 72±1, 75±1 °C, extrudates were impossible to be collected in a fine filament form.

Figure 33. Temperature Controller of The Micro-Extruder

4.1.3 Melt Extrusion at Various Temperatures

Even a slight temperature change may largely affect the extrusion rate and extrudate properties. This was demonstrated by the control groups that contain PCL, PEG400 and chitosan/keratin. The resulting change in properties are shown in Table 8. When the temperature was set at 60±1 °C, the powder blend could not melt inside the extruder and no extrudate came out of the nozzle. Figure 32 shows extrudates of control group 5 at different temperatures. When the temperature was set at 63±1 °C, it was able to obtain fine fibres with a stable extruding rate. Raising the temperature to 66±1 °C, the extrudates were still relatively fine but under an unstable extruding rate, this caused difficulty in fibre collection. With the increase in temperature, extrudates were gradually liquefying and air pocket formation was observed followed by hearing a “bubble breaking sound”, which resulted in the impossibility of collecting products in filament form (temperature at 69±1 ,72±1 and 75±1 °C). Please note that the temperature fluctuation of 1 degree Celsius reflects the temperature controller readings, which fluctuated ±1 degree. The only way to adjust the temperature was to slightly turn the indicator and carefully observe the temperature change (Figure 33).

Table 8 Control Group (contains PCL+PEG400+Chitosan/keratin) at Various Temperatures

Control group C3, C4 and C5 demonstrated similar physical properties at temperatures of 60±1 °C, 63±1 °C, 66±1 °C, 69±1 °C, 72±1 °C, 75±1 °C.

Temperature °C Physical properties 60±1 Powder could not melt 63±1 Fine and long fibres 66±1 Relatively fine fibres but not under stable extruding rates 69±1 Liquid extrudates 72±1 Very liquid extrudates 75±1 Very liquid extrudates

4.2 Summary of Biomaterial Production and Suture Generation

The optimization results showed that:

⚫ It was important to purge melt extruder with a temperate higher than the melting point of PCL. The certain purge temperature depended on the previous materials used in the machine.

⚫ 63±1 °C is the optimum heating temperature for biomaterials to be melted effectively for extrusion.

⚫ Sutures that contain chitosan/keratin tend to be slightly thicker than sutures made with PCL only or PCL/PEG blend.

⚫ Sutures that contain higher amounts of PEG400 exhibit a greyish colour.

4.3 Fourier-Transform Infrared Spectroscopy

FTIR spectra of different material sutures are shown in Figure 34. PCL (C1) showed peaks around 2943 cm-1, 2866 cm-1, 1721 cm-1, 1293 cm-1, 1238 cm-1, 1163 cm-1, and 1106 cm-1 (Table 9). There was a significant difference in the spectra of PCL and the PCL/PEG blend (C2) demonstrated by a sharp peak at 3441 cm-1, indicating the presence of hydroxyl groups (OH). These spectral characteristics did not change with the addition of chitosan/keratin blend (C3, C4, C5).

Table 9. Functional Groups of PCL

Position (cm-1) Vibrator Abbreviation

2943 –C–H asymmetric stretching νas(CH2)

2866 –C–H symmetric stretching νs(CH2) 1721 –C=O stretching ν(C=O)

1293 C–O and C–C stretching in the crystalline phase νcr

1238 Asymmetric COC stretching νas(COC)

1163 C–O and C–C stretching in the amorphous phase νam 1106 C–O stretching ν(C–O)

Formulation FTIR Spectra

C–O/C–C –CH C–O a –CH –C=O COC C–O/C–C

C–O/C–C –OH b –CH –CH C–O –C=O COC C–O/C–C

Formulation FTIR Spectra

–OH C–O/C–C –CH c C–O –CH C–O/C–C –C=O COC

C–O/C–C –OH –CH C–O –CH –C=O d COC C–O/C–C

Formulation FTIR Spectra

–OH –CH C–O/C–C –CH e –C=O C–O COC C–O/C–C

Figure 34. FTIR Spectra of Individual Samples and Overlay Graph

Individual Spectra of pure PCL, PCL/PEG and individual spectra from different formulation of PCL/PEG/Chitosan/Keratin are shown, the spectra of the overlay of all the ingredients and samples are shown. a) PCL, b) PCL+PEG400, c) PCL+PEG400+chitosan-keratin=80:19:1 w/w, d) PCL+PEG400+chitosan-keratin=80:18:2 w/w, e) PCL+PEG400+chitosan-keratin=80:16:4 w/w, f) spectra of all the ingredients and samples. Data were acquired and graphed using OPUS software and shown in transmittance. Individual spectral images were used for peak analysis. Arrows indicate the peaks of interest.

4.4 Thermal Testing

4.4.1 TGA

TGA graphs of all the extruded samples are shown in Figure 35. Sutures were heated from 20 °C (room temperature) to 800 °C at 10°C/min and held at an isotherm for 3 min (Mohamed et al. 2008). All the samples showed a simple one-step decomposition profile with a single transition temperature, as demonstrated by TGA. The sample degradation profile which contains PCL: PEG: chitosan/keratin=80:16:4 (w/w) (C5) showed a higher thermal stability than the rest of samples, it started degrading at 377 °C and completely decomposed (0% w/w) at 778 °C, whereas pure PCL (C1) started degrading at 374 °C and degraded completely at 559 °C. With PCL: PEG4: chitosan/keratin= 8:18:2 (w/w) (C4), the sample was not fully degraded until after 800 °C. PCL and PEG blend (C2) showed the lowest thermal stability, it started to degrade at 325 °C, when the temperature reached 340 °C, half the sample degraded and fully decomposed at 477 °C. Overall, C3 exhibited a relatively slow and steady decomposition rate in comparison to the rest of the samples.

Formulation TGA Graphs

Figure 35. TGA Curves of Individual Samples and Overlay Graph a) PCL, b) PCL+PEG400, c) PCL+PEG400+chitosan-keratin=80:19:1 w/w, d) PCL+PEG400+chitosan-keratin=80:18:2 w/w, e) PCL+PEG400+chitosan-keratin=80:16:4 w/w, f) overlay graph of all the samples. Data were acquired and graphed using the TA Instruments Software Universal Analysis (TA Instruments, MA, USA). Individual graphs of pure PCL, PCL/PEG and PCL/PEG/chitosan/keratin are shown. The graph of the comparison of all the samples is shown.

4.4.2 DSC

The DSC profiles of the pure PCL and PCL-based composites are shown in Figure 36. All the samples were reheated to eliminate thermal history. Each sample included a glass transition (Tg), endothermic during heating (melting), and exothermic during cooling

(crystallization). The corresponding heat capacity (ΔCp) and Tg temperatures obtained from the DSC data of the pure PCL and its composites are shown in Table 10, 11, 12. Pure PCL

(C1) showed the highest Tg temperature at -68.91 °C, whereas the polymer composite of PCL: PEG: chitosan/ keratin=80:18:2 (w/w) (C4) demonstrated the highest heat capacity of 3.48 J/(g°C). When the ratio of PEG: chitosan/ keratin was 19:1 (w/w) (C3), crystallization point was the lowest compared to the other two ratios, which indicated that C3 has lower crystallinity.

Table 10. Summary of Differential Scanning Calorimetry Analysis of Samples

The glass transition temperatures and heat capacities of sample C1, C2, C3, C4 and C5 are shown below:

Thermal property Sample Temperature (°C) Heat capacity (J/(g°C) C1 -68.91 0.77 C2 -69.01 2.32 Glass transition C3 -69.31 3.14 C4 -69.41 3.48 C5 -69.20 3.40

Table 11. Endothermic Transition for Each Thermal Property

Thermal property Sample Temperature (°C) Melting onset C1 50.53 C2 51.31 C3 50.71 C4 50.22 C5 50.08 Melting peak C1 56.68 C2 55.96 C3 55.31 C4 54.57 C5 54.61

Table 12 Exothermic Transition for Each Thermal Property

Thermal property Sample Temperature 1 (°C) Temperature 2 (°C) Cooling onset C1 31.64 - C2 32.08 -30.79 C3 31.75 -32.38 C4 32.35 -32.52 C5 32.17 -31.6 Cooling peak C1 28.04 - C2 30.04 -34 C3 29.15 -36.08 C4 30.79 -35.39 C5 30.50 -34.55

Formulation DSC Graphs

Figure 36. DSC Curves of Individual Samples and Overlay Graph a) PCL, b) PCL+PEG400, c) PCL+PEG400+chitosan-keratin=80:19:1 w/w, d) PCL+PEG400+chitosan-keratin=80:18:2 w/w, e) PCL+PEG400+chitosan-keratin=80:16:4 w/w, f) overlay graph of all the samples. Data were acquired and graphed using the TA Instruments Software Universal Analysis (TA Instruments, MA, USA). Individual graphs of pure PCL, PCL/PEG and PCL/PEG/chitosan/keratin are shown. The graph of the comparison of all the samples are shown.

4.5 Tensile Testing

Tensile testing results were obtained by using the tensile analyser (Figure 14). The results of tensile testing were demonstrated with all major data sets (failure load, elongation at break, tensile strength, and Young’s modulus). Mean and standard deviation values of diameter (mm), failure load (N), elongation at break (%), tensile strength (MPa) and young’s modulus (MPa) are presented for all samples in Table 13.

Curves for all samples comparing stress to strain are shown in Figure 37 (a) and the values of tensile strength are shown in Figure 37 (b). The failure load and elongation of all samples are shown in Figure 38 (a, b). Young’s modulus results are displayed in Figure 39.

C5 has the highest failure load, which would make a good choice for high tension areas, such as scalp wound closures. C1 allowed for the most elongation before breaking, corresponding to its high ductility. Increased elongation would be advantageous in situations where a great deal of oedema is expected postoperatively. The values of the tensile strength of samples can determine the highest strength material independent of diameter. The stress-strain curves and tensile strength as shown in Figure 37 (a, b) exhibited that C1 and C3 have the highest tensile strength, which means C1, C3 have stronger mechanical properties than the rest of the samples. The higher Young’s modulus value of C3 demonstrated the higher stiffness of the sample (Figure 39).

Overall, C3 is a relatively strong and stiff material with proper elongation, which requires only a slightly higher strength to break it. Therefore, it is an ideal candidate for surgical suture material.

Table 13. Summary of Mechanical Testing Analysis of Suture Samples

Sample results of diameter, failure load, elongation at break, tensile strength and young’s modulus are shown with mean and standard deviation values.

Sample Diameter Failure Elongation Tensile Young’s (mm) Load (N) at break (%) strength (MPa) modulus (MPa) 0.547 4.8 123.4 20.7 (±0.907) 3.0 (±0.123) C1 (±0.091) (±0.318)

0.537 3.4 109.5 14.4 (±1.074) 2.7 (±0.110) C2 (±0.051) (±0.404) 0.618 5.5 66.9 20.2 (±0.244) 3.1 (±0.195) C3 (±0.040) (±0.141) 0.675 5.6 20.3 13.9 (±0.378) 2.4 (±0.110) C4 (±0.030) (±0.342) 0.701 6.5 19.9 16.4 (±0.612) 2.9 (±0.151) C5 (±0.036) (±0.228)

Figure 37. Tensile Properties of Suture Samples

Samples of pure PCL, PCL/PEG, and different formulations of PCL/PEG/chitosan/keratin were measured (a) stress versus strain before deformation; (b) the values of tensile strength. Each bar was presented with mean value and standard deviation.

Figure 38. Ductile Failure of Suture Samples

Samples of pure PCL, PCL/PEG, and different formulations of PCL/PEG/chitosan/keratin were measured for (a) failure load;(b) elongation at break. Each bar was presented with mean value and standard deviation.

Figure 39. Young’s Modulus of Suture Samples

Samples of pure PCL, PCL/PEG, and different formulations of PCL/PEG/chitosan/keratin were measured for their stiffness. Each bar is presented with mean value and standard deviation.

4.6 Summary of Suture Characteristics

Table 14. The Summary of the Suture Properties with Different Formulations

Sampl Materials (w/w) Diamete Melting Glass Decompositio Tensile e r (mm) temperatur transitio n temperature strengt e (°C) n (°C) (°C) h (MPa) C1 PCL 0.5 56.68 -68.91 374.62 20.7 C2 PCL/PEG (80/20) 0.4 55.96 -69.01 325.64 14.4 C3 PCL/PEG/chitosa 0.6 55.31 -69.31 347.82 20.2 n-keratin (80/19/1) C4 PCL/PEG/chitosa 0.6 54.57 -69.41 369.10 13.9 n-keratin (80/18/2) C5 PCL/PEG/chitosa 0.9 54.61 -69.20 377.69 16.4 n-keratin (80/16/4) The comparison of characterisation of different suture formulations are listed in table 14. The studies concluded:

⚫ FTIR analysis showed that, except for the presence of hydroxyl groups (OH) in the spectrum of C2 (PCL/PEG400 blend) compared to C1 (PCL only), the addition of chitosan/keratin did not generate new molecular bonds (C3, C4, C5).

⚫ Sutures that contained PCL/PEG400/chitosan/keratin exhibited a relatively slow and steady decomposition rate in comparison to the PCL alone and PCL/PEG400 blend sutures.

⚫ All the samples presented a single Tg, which indicates the miscibility or interaction of molecules. And the addition of the PEG400/chitosan/keratin in the PCL reduced the melting point and glass transition temperatures.

⚫ Amongst all five samples, C3 (PCL/PEG400/chitosan/keratin blend) is the most ideal candidate for surgical suture application, due to its relatively strong and stiff material properties with proper elongation, which requires a slightly higher force to break.

4.7 Suture Preparation for the Formulation Groups

Polymers were split into three groups: group 1 contained PCL, PEG400, chitosan/keratin (Table 15); group 2 consisted of PCL, PEG400, chitosan (Table 16); group 3 was composed of PCL, PEG400, keratin (Table 17). Three polymer groups were extruded with drug DP respectively. The drug percentage in the compound was 5, 15, 30 wt%.

F1, F4 and F7 had the lowest drug content, the extruding process was consistent, thus, a uniform thickness was easy to achieve. Additionally, the extrudates were relatively white with a smooth surface. F2, F5 and F8 were extruded with a higher drug percentage. Products maintained uniform thickness but the colour turned yellowish. F3, F6 and F9 consisted of the highest drug content of 30%. Due to the drug additives, those with a lower melting temperature such as glucose and starch, and those with a higher melting temperature such as diclofenac potassium, extrudates were greyish with many visible white lumps on the surface, which caused the brittleness of the fibres (Figure 40). Interestingly, under the same drug load, fibres contained PCL/PEG400/chitosan were slightly thinner (0.8 mm) than fibres that contained PCL/PEG400/keratin (0.9 mm) and PCL/PEG400/chitosan/keratin (0.9 mm).

Table 15. Summary of Suture Preparation for Group 1

Group 1 contained PCL, PEG400, chitosan/keratin and drug, melt-extruded at 63±1 °C

F PCL (g) PEG400 (g) Keratin- Diclofenac Physical Chitosan(g) Potassium (g) properties (50:50) (Powder) F1 8 1.9 0.1 0.1 White, smooth F2 8 1.9 0.1 0.3 Yellow, smooth F3 8 1.9 0.1 0.6 Grey, rough

Table 16. Summary of Suture Preparation for Group 2

Group 2 contained PCL, PEG400, chitosan and drug, melt-extruded at 63±1 °C

F PCL (g) PEG400 (g) Chitosan Diclofenac Physical Potassium (g) properties (Powder) F4 8 1.9 0.1 0.1 White, smooth F5 8 1.9 0.1 0.3 Grey, smooth F6 8 1.9 0.1 0.6 Grey, rough

Table 17. Summary of Suture Preparation for Group 3

Group 3 contained PCL, PEG400, keratin and drug, melt-extruded at 63±1 °C

F PCL (g) PEG400 (g) Keratin Diclofenac Physical Potassium (g) properties (Powder) F7 8 1.9 0.1 0.1 White, smooth F8 8 1.9 0.1 0.3 Dark grey, smooth F9 8 1.9 0.1 0.6 Dark grey, smooth

Figure 40. Extrudates of Suture Samples Contained Drug

Group 1 (F1, F2, F3) contained PCL, PEG, chitosan, keratin and drug; group 2 (F4, F5, F6) contained PCL, PEG, chitosan and drug; group 3 (F7, F8, F9) contained PCL, PEG, keratin and drug, were melt-extruded at 63±1 °C. F1, F4, F7 contained 5% of drug; F2, F5, F8 contained 15% of drug; F3, F6, F9 contained 30% of drug.

4.8 Drug Dissolution Testing

4.8.1 Calibration Curve

A calibration curve was generated by measuring absorbances at 276 nm, based on known drug concentrations of 0.0312, 0.0625, 0.125, 0.25, 0.5 mg/mL (Table 18). The profile is shown at Figure 41.

Table 18. Absorbances of Known Concentrations of Drug in Phosphate Buffer at pH 7.4

Five concentrations of drug were measured at absorbance of 276 nm using the UV-Visible spectrometer. It can be seen that absorbance is increasing with higher concentration of drug.

Serial number Concentration (mg/mL) Absorbance at 276 nm 1 0.0312 0.013 2 0.0625 0.025 3 0.125 0.053 4 0.25 0.098 5 0.5 0.167

Calibration Curve

0.2 0.18 0.167 0.16 0.14 y = 0.3277x + 0.0077 0.12 0.098 R² = 0.9909 0.1 0.08 0.053

0.06 Absorbance276 nm 0.04 0.025 0.013 0.02 0 0 0.1 0.2 0.3 0.4 0.5 0.6 Concentration [mg/ml]

Figure 41. Standard Curve of Drug in Phosphate Buffer at pH 7.4.

The calibration curve was used for calculating the unknown concentrations with known absorbances in the drug dissolution experiment.

4.8.2 Drug Release Profiles

Nine formulations were divided into three groups which contained different polymer blends, as shown in table 19. Group 1= PCL/PEG400/chitosan/keratin/drug; Group 2= PCL/PEG400/chitosan/drug; Group 3= PCL/PEG400/keratin/drug. Each group comprised an embedded drug amount of 5, 15 and 30%. All sutures were placed into tubes that contained phosphate buffer at pH 7.4 and immersed into a water bath.

F1, F2 and F3 are composed of PCL, PEG, chitosan/keratin blend and drug. The drug content of F1, F2 and F3 was 5, 15, 30% respectively. Before 1.75 h, the drug dissolved slowly from the polymer matrices of F1, F2 and F3. After the point of 1.75 h, the drug release rate slightly increased until 6.75 h. Three formulations demonstrated similar drug release patterns, which indicated that the drug content had little impact on the polymer matrices (Figure 42).

F4, F5 and F6 consist of PCL, PEG, chitosan and drug. The drug content of F4, F5 and F6 was 5, 15, 30% respectively. F5 and F6 demonstrated similar drug release patterns, drug dissolution rate slightly improved between a time interval of 1.75 and 3.75h. On the other hand, F4 exhibited a sustained release at all times. Between 1.75 h and 3.75h, the drug concentration increased by 70% in F4 and F6. After 3.75h, the release rate slowed down, but was still quicker than the initial release rate. Surprisingly, F5 had a lower concentration than F4 the entire experimental time. However, at 3.75 h, the concentration gap between F4 and F5 narrowed dramatically, and so it can be predicted that a cross point will be seen after several hours (Figure 43), due to the relatively fast degrading material PEG400, since its degradation could last up to 3 days (Browning et al. 2014).

F7, F8 and F9 are composed of PCL, PEG, keratin and drug. The drug content of F7, F8 and F9 was 5, 15, 30% respectively. This group contained polymer blend of PCL, PEG400 and keratin. All the formulations exhibited a rapid drug release profile for the first 1.75 h, and the plateau state achieved at 3.75 h. These polymer groups generated a rapid drug release profile (Figure 44).

F3, F6 and F9 contained the highest amount of drug diclofenac (0.6 g) in their groups. Comparing these three formulations, F9 demonstrated the fastest drug release profile. The polymer matrices that contained PCL, PEG400, chitosan (F6) and PCL, PEG400, chitosan- keratin blend (F3) showed a sustained drug release profile. Moreover, the F3 exhibited more

stable controlled release rate than F6, which is an ideal candidate for controlling diclofenac release inside the human body (Figure 45).

Table 19. Concentrations of Dissolution Media at Different Time Intervals.

Group 1 (F1, F2, F3) consisted of PCL, PEG, chitosan, keratin and drug; group 2 (F4, F5, F6) contained PCL, PEG, chitosan and drug; group 3 (F7, F8, F9) was composed of PCL, PEG, keratin and drug. F1, F4, F7 contained 5% drug; F2, F5, F8 contained 15% drug; F3, F6, F9 contained 30% drug. Dissolution media were extracted at time intervals of 15 minutes, 30 minutes, 1 hours, 2 hours and 3 hours.

Time (h) 0.25 0.75 1.75 3.75 6.75

Concentration (mg/mL)

F1(mg/ml) 0.013 0.034 0.072 0.242 0.398 F2 0.187 0.218 0.327 0.559 0.739 F3 0.379 0.456 0.565 0.840 1.109 F4 0.294 0.395 0.492 0.709 0.874 F5 0.111 0.175 0.236 0.575 0.767 F6 0.327 0.477 0.565 0.977 1.200 F7 0.089 0.199 0.318 0.520 0.535 F8 0.099 0.242 0.428 0.620 0.688 F9 0.242 0.547 1.005 1.167 1.191

Figure 42. Dissolution of Group 1 (F1, F2, F3) in Phosphate Buffer at pH 7.4

Group 1 (F1, F2, F3) contained PCL, PEG, chitosan, keratin and drug. The drug content of F1, F2 and F3 were 5%, 15% and 30%, respectively.

Figure 43. Dissolution of Group 2 (F4, F5, F6) in Phosphate Buffer at pH 7.4

Group 2 (F4, F5, F6) contained PCL, PEG, chitosan and drug. The drug content of F4, F5 and F6 were 5%, 15% and 30%, respectively.

Figure 44. Dissolution of Group 3 (F7, F8, F9) in Phosphate Buffer at pH 7.4

Group 3 (F7, F8, F9) contained PCL, PEG, keratin and drug. The drug content of F7, F8 and F9 were 5%, 15% and 30%, respectively.

Figure 45. Comparison of F3, F6 and F9 in Phosphate Buffer at pH 7.4

Group 4 (F3, F6, F9) contained 30% of drug. The formulations of F3, F6 and F9 were PCL/PEG/chitosan/keratin /drug, PCL/PEG/chitosan/drug, and PCL/PEG/keratin/drug, respectively.

4.9 Summary of Drug-Eluting Suture Fabrication and Drug Dissolution

The suture generation and drug dissolution profiles indicated that:

⚫ Sutures that contained lower drug amounts demonstrated relatively white fibres with a smooth surface. Whereas, sutures with higher drug amount showed greyish colour with visible white lumps on the surfaces. Moreover, extrudates contained PCL/PEG400/chitosan were slightly less thick (0.8 mm) than fibres that contained PCL/PEG400/keratin (0.9 mm) and PCL/PEG400/chitosan/keratin (0.9 mm)

⚫ Group 3 (PCL/PEG/keratin/drug) demonstrated a rapid release profile. Whereas, Group 1 (PCL/PEG/chitosan/keratin/drug) and Group 2 (PCL/PEG/chitosan/drug) generated a controlled drug release profile. The release rate of Group 1 was even 23% slower than Group 2 between 1.75 and 3.75 h.

4.10 Biocompatibility Analysis

4.10.1 MTT Assay

In vitro biocompatibility/cytotoxicity of drug-treated sutures was determined using HaCat cells (Figure 46). According to the direct method, cells were initially seeded at a density of 5 x 103 cells/mL per well. Formulation groups and control groups were incubated for 24, 48 and 72 h in the presence of drug-treated or untreated sutures. Control cells were treated with media only. Purple formazan crystals were observed on HaCat cells after adding MTT (Figure 47). During the experiment, it was interesting to observe that sutures with drug turned purple after contact with MTT solution. Whereas, sutures without drug stayed white (Figure 48). For that reason, suture alone without cells were also tested to ensure that the suture itself did not contribute to the MTT results. Standard curves at 24, 48 and 72 h were generated by measuring absorbance at cell numbers of 5 x 103, 10 x 103, 20 x 103, 40 x 103, 80 x 103 cells/mL (Figure 49).

Group 1 (F1, F2, F3) were drug-treated polymer (PCL/PEG/chitosan/keratin) sutures, embedded with 5, 15, 30 wt% diclofenac potassium, respectively. Group 2 (F4, F5, F6) were drug-treated polymer (PCL/PEG/chitosan) sutures, embedded with 5, 15, 30 wt% diclofenac potassium, respectively. Group 3 (F7, F8, F9) were drug treated polymer (PCL/PEG/keratin) sutures, embedded 5, 15, 30 wt% diclofenac potassium, respectively. Control (C1, C2, C3) were polymer sutures containing pure PCL, PCL/PEG400, PCL/PEG400/chitosan/keratin, respectively. Cell is HaCat cells only. F3 (Group 1), F5 (Group 2) and F7 (Group 3) exhibited a significant decrease on cell proliferation at 24 h (p = 0.0002, p < 0.0001, p = 0.0141, respectively). Group 1 had a significant increase on cell proliferation at 48 and 72 h (p = 0.0001, 0.0002 respectively). Group 2 demonstrated a significant effect on the growth of cell number at 48 and 72 h (p = 0.0002, 0.0021 respectively). There was also a significant increase of group 3 on the viable cells at 48 and 72 h (p = 0.0039, 0.0002, respectively).

There was a significant increase (p < 0.05) in cell proliferation associated with drug-treated sutures at 48 and 72 h, compared to control samples and cell-only samples (Figure 50, 51, 52). At 24 h, it is interesting to note that F2, F4 and F6 showed much higher viability of cells compared to the control and cell-only samples. However, at 48h, the number of cells in these three formulations dropped under 20 x 103 cells (Figure 50, 51). All three formulation groups

(F1, F2, F3, F4, F5, F6, F7, F8, F9), exhibited similar cell viability with polymer only sutures (C1, C2, C3) at 48h (Figure 50, 51, 52). Nevertheless, at 72 h, only group 1 and group 3 with formulations F1, F2, F9 showed significantly increased cell viability compared to the control group (C1, C2, C3) and cell only (Figure 50, 52). Additionally, F1 achieved the highest cell number of 40,000 at 72 h (Figure 50).

Cell attachment in F1, F2 and F9 was significantly increased with the drug-treated sutures compared to the untreated sutures at 72 h. Drug content of these three formulations were 5, 15 and 30 wt%, respectively. It indicated that Group 3 polymer matrix was able to carry an increased drug amount. Whereas, Group 1 polymer matrix performs better cell-substrate compatibility with a lower amount of drug. Therefore, in order to further investigate the cell viability of group 1 and group 3, these two groups were selected for the live/dead assay.

Figure 46. HaCat Cells

Spindle shaped adherent cells that form a cobble stone pattern when confluent. Image taken with x microscope, untreated cells.

Figure 47. Purple Formazan Crystals of HaCat Cells

Crystals formed after HatCat cells were stained with MTT

(A) (B)

Figure 48. Colour Changing of Sutures After Adding MTT Solution

(A) Sutures contained polymer only; (B) Sutures contained polymer blend and drug.

Figure 49. Representative standard curve of MTT assay

Data were collected at 24, 48, 72 h and analysed via Prism software (R2=0.9890, 0.9535, 0.9865 respectively).

Figure 50. Cell Number for Group 1 via MTT Assay

Data were collected after 24, 48 and 72 h. Sutures were incubated using the direct method. F1, F2, F3 contained PCL, PEG, chitosan, keratin and drug. The drug content of F1, F2 and F3 were 5, 15, 30% respectively. C1contained PCL only; C2 contained PCL and PEG; C3 contained PCL, PEG, chitosan and keratin. Cell represents cell only (i.e. does not contain polymer or drug). Asterisk (*) denotes significant difference. * denotes p ≤ 0.05, ** denotes p ≤ 0.01, *** denotes p ≤ 0.001. (n=3, data groups were compared with the one-way analysis of variance ANOVA via a post hoc Tukey’s test. p=0.0002, 0.0001, 0.0002 respectively).

Figure 51. Cell Number for Group 2 via MTT Assay

Data were collected after 24, 48 and 72 h. Sutures were incubated using the direct method. F4, F5, F6 contained PCL, PEG, chitosan, keratin and drug, the drug content of F4, F5 and F6 were 5, 15, 30% respectively. C1contained PCL only; C2 contained PCL and PEG; C3 contained PCL, PEG, chitosan and keratin. Cell represents cell only (i.e. does not contain polymer or drug). Asterisk (*) denotes significant difference. * denotes p ≤ 0.05, ** denotes p ≤ 0.01, *** denotes p ≤ 0.001. (n=3, data groups were compared with the one-way analysis of variance ANOVA via a post hoc Tukey’s test. p< 0.0001, p=0.0002, p=0.0021, respectively).

Figure 52. Cell Number for Group 3 via MTT Assay

Data were collected after 24, 48 and 72 h. Sutures were incubated using the direct method. F7, F8, F9 contained PCL, PEG, chitosan, keratin and drug, the drug content of F7, F8 and F9 were 5, 15, 30% respectively. C1contained PCL only; C2 contained PCL and PEG; C3 contained PCL, PEG, chitosan and keratin. Cell represents cell only (i.e. does not contain polymer or drug). Asterisk (*) denotes significant difference. * denotes p ≤ 0.05, ** denotes p ≤ 0.01, *** denotes p ≤ 0.001. (n=3, data groups were compared with the one-way analysis of variance ANOVA via a post hoc Tukey’s test. p=0.0141, 0.0039, 0.0002 respectively).

4.10.2 Live/Dead Assay

Live/ dead assays were conducted on group 1, group 3, control group, and cell-only samples. The results were reported at 24, 48 and 72 h with representative images (Figure 53) The results for 24, 48, 72 h culturing showed that cell fluorescence was stronger on the surface of the sutures which contain drug, possibly indicating that more cells were attached but this needs further investigation (Figure 54).

The live/dead cell ratios were obtained from fluorescence images and counted using Image J software. The cells in each sample were treated with Calcein blue for live cell staining of the intact cell membrane and propidium iodide (PI) for dead cell staining of the DNA. One-way ANOVA analysis showed that at 24 h, suture materials F (13, 27) had an extremely significant increase (p < 0.0001) on the live percentage of cells. However, there were no significant differences between any of the sutures at 48 and 72 h. At 24 h, control group C2 and C3 demonstrated relatively higher cell viability at 97% and 99%, respectively. Whereas, the cell- only sample had the lowest viability at 54%. At 48h, the viability of group 1 and group 3 samples became stable, and maintained above 90%. When it comes to 72 h, the ratios of live cells were 97%, 97%, 98%, 96%, 93%, 96%, 90%, 96%, 92%, 84 % for F1, F2, F3, F7, F8, F9, C1, C2, C3, cell only, respectively (Figure 55). Therefore, Group 1 generated relatively higher cell viability compared to the other groups. Especially, F3 demonstrated the highest cell viability at 98%.

100

(a) 24 h Live Dead

Cell only

101

24 h Live Dead

102

(b) 48 h Live Dead

Cell only

103

48 h Live Dead

104

Cell only

105

72 h Live Dead

106

Figure 53. Live/Dead Assay with Fluorescence Microscopy.

(a) cell fluorescence at 24 h; (b) cell fluorescence at 48 h; (c) cell fluorescence at 72 h. (n=3. Scale bar= 200 µm). Green fluorescence represents live cells, red fluorescence indicates dead cells.

Figure 54. Fluorescence of Live/Dead Cells on Sutures

(a) Live Cells and (b) Dead Cells Adhered to Sutures (Scale bar= 200 µm)

107

108

Figure 55. Ratio of Live/Dead Cells in Response to Different Sutures at 24, 48 and 72h

Red bars represent the percentage of the dead cells; blue bars represent the percentage of live cells. F1, F2, F3 contained PCL, PEG, chitosan, keratin and drug, the drug content of F1, F2 and F3 were 5, 15, 30% respectively. F7, F8, F9 contained PCL, PEG, chitosan, keratin and drug, the drug content of F7, F8 and F9 were 5, 15, 30% respectively. C1 contained PCL only; C2 contained PCL and PEG; C3 contained PCL, PEG, chitosan and keratin. Cell represents cell only (i.e. does not contain polymer or drug). Graphs were obtained by using Prism software (at 24 h, p< 0.0001).

109

4.10.3 Scratch Assay

The wound healing assay was used to measure the effect the of polymer/drug on the migratory capacity (wound closure rate) of HaCat cells. Pictures obtained through light microscopy on the scratched cell monolayer are reported at time 24, 48 and 72 h (Figure 56). Different dimensions of the scratch were observed by comparing the polymer sutures and drug-treated sutures. The results showed that the combination of polymer and drug (F1, F2, F3) significantly increased the gap closure rate compared to polymer only and cell-only samples, since some of the samples from formulation F1, F2, F3 were able to close the wound completely by 48 h (Figure 56).

The percentage of migration has been calculated and reported in the graph shown in Figure 57. A two-way ANOVA analysis showed that time F (1.659, 23.22) had an extremely significant effect (p < 0.0001). Whereas the different sample materials F (6, 14) had a very significant effect (p = 0.0033). Overall both time and sample materials F (12, 28) significantly increased wound closure (p = 0.0038) when compared to the cell-only sample. Before 24 h, C1, C2, C3, F1, F2 and cell-only samples demonstrated similar wound closure rates. After the 48-h point, cell migration rate largely increased in F1 and F2 samples. For sample F3, it exhibited the highest cell migration rate at all times, and it achieved a 99% wound healing at 48h (Figure 57). Therefore, the F3 material, which contained polymer PCL/PEG400/chitosan/keratin as well as 30 wt% diclofenac potassium was the ideal candidate for wound healing application.

110

Time 0 h 24 h 48 h

Cell only

Figure 56. Images of Gap Closure in Wound Healing Assay in Response to Sutures

Wounds were measured at 0, 24 and 48 h after wound was made. Control cells were treated with media only. Suture treated cells were treated through direct contact with sutures (with/without drug) Images shown here are representatives. n=6, yellow line outlines the border of the wound. Three measurements were taken per image. Scale bar = 20 µm.

111

Figure 57. Percentage of Wound Healing

F1, F2, F3 contained PCL, PEG, chitosan, keratin and drug, the drug content of F1, F2 and F3 were 5, 15, 30% respectively. C1 contained PCL only; C2 contained PCL and PEG; C3 contained PCL, PEG, chitosan and keratin. Cell only represents no polymer or drug-containing.

112

4.11 Summary of Biocompatibility Tests

The proliferation assay, viability assay and scratch assay showed that:

⚫ Group 1, 2, 3 exhibited larger numbers of cells compared to polymer suture samples at 48 h. However, only Group 1 and Group 3 showed higher cell viability at 72 h. Especially F1, which showed the highest cell number of 40,000 cells at 72 h.

⚫ There was an extremely significant effect (p < 0.0001) on the percentage of viable cells at 24 h. However, no difference was exhibited at 48 and 72 h.

⚫ Group 1(PCL/PEG400/chitosan/keratin/drug) generated relatively higher cell viability at 72 h compared to other groups. The live cell percentage of F1, F2 and F3 were 97%, 97%, 98% respectively.

⚫ F3 demonstrated the highest wound healing rate compared with the other samples. Furthermore, it achieved 99% wound closure at 48 h.

113

5 DISCUSSION

Suture implantation is prone to an inflammation response and requires postoperative surgery to extract the suture from the patient’s body (Champeau et al. 2017). In order to cope with this problem, there is a trend towards developing synthetic biodegradable suture materials that combine with anti-inflammatory drugs. Incorporation of the drug with polymer materials to produce suture products is important especially when the suture is used for internal organs and/or tissues, which also provides the possibility to predict the degradability of the sutures in a biological environment (Joseph et al. 2017). For internal targets, direct drug delivery is difficult. In some cases, a high drug content of up to 30% is required. However, a higher amount of drug leads to lower mechanical properties, as sutures tend to be degraded easily (Zhang et al. 2016). Therefore, a robust suture material or material combinations for carrying a higher amount of drug are desired.

In this study, the overall aim was to use the HME technique to generate a surgical suture from a novel biomaterial consisting of PCL/PEG400/chitosan/keratin and the drug DP, which has excellent mechanical properties, a sustained drug release rate and biocompatibility. To complete this aim, the HME technique was optimized, the sutures were characterised chemically and physically, and their thermal and mechanical properties were estimated. Additionally, DP as the target drug was studied to determine its release profile. Moreover, the in vitro biocompatibility of the sutures was assessed.

The four critical conclusions of this project originate from 1) the optimization of the hot- melt extrusion process 2) the characterisation of biopolymer sutures 3) the drug dissolution of drug embedded sutures 4) the biological studies of the drug embedded sutures in vitro.

The conclusions were as follows:

1) The optimal extrusion temperature for polymer and polymer/drug sutures preparation was 63±1 °C. Sutures that contained chitosan/keratin tended to be thicker (0.6 ~ 0.9 mm) than sutures fabricated with PCL only and PCL/PEG blend (0.4 ~ 0.5 mm). Drug embedded sutures had thickness between 0.8 and 0.9 mm. Additionally, higher amounts of PEG400 exhibited a grey colour which appeared on the suture surfaces during fabrication process. 114

2) C3 (PCL/PEG400/chitosan/keratin blend) had the best performance in characterisation studies. This blend exhibited a relatively slow and steady decomposition rate in the TGA test, and had a lower melting point and glass transition temperature in the DSC test. Moreover, in the mechanical test, C3 was relatively strong and stiff, with lower elongation at break than pure PCL and PCL/PEG blend, but had significantly higher elongation than the other drug embedded suture samples.

3) Group 3 (PCL/PEG/keratin/drug) demonstrated a rapid release profile. Whereas, Group 1 (PCL/PEG/chitosan/keratin/drug) and Group 2 (PCL/PEG/chitosan/drug) generated a controlled drug release profile. The release rate of Group 1 was even slower than Group 2 after 1 h 45 m.

4) Group 1 (PCL/PEG/chitosan/keratin/drug) showed the highest cell viability at 72 h. Furthermore, F3, which had the highest drug content (30 wt%) in Group 1, achieved 99% wound closure at 48 h.

5.1 Analysis of Suture Fabrication and Characteristics

5.1.1 Hot- melt Extrusion Parameters

During HME, process temperature was set at 63±1 °C, since the main polymer carrier selected in this study was PCL, which operates optimally at this temperature. PCL has a relatively low glass transition temperature (-60 °C) and melting point (60 °C), and it occupied 80% of the polymer matrix. In polymer production, the reduction of viscosity is essential, and to achieve this, the temperature must be maintained higher (within the range of 30-60 °C) than the glass transition temperature (Palazi et al. 2018). The process temperature also needs to be slightly higher than the polymer’s melting point to enable material melting and minimize viscosity (Palazi et al. 2018).

PEG400 used in this study, has a melting point at 4 °C and glass transition temperature at - 20 °C. Fibres extruded from these two polymers demonstrated homogenous properties, as PEG acts as a plasticizer to increase the plasticity of the PCL/PEG fibres (Barmpalexis et al.

115

2013). However, due to the low melting point of PEG400, the colour of the extrudates was grey, which can be ascribed to the decomposition of PEG400 (Palazi et al. 2018). To date, research on the effect of the decomposition of PEG400 on the efficacy of drug-eluting sutures is limited. Therefore, if the change in colour affects drug-eluting properties requiring future investigation.

Chitosan and keratin have a melting temperature of 290 °C and 205 °C, respectively, and their glass transition temperature were 203 °C and 155 °C, respectively. Due to the high melting points of chitosan and keratin, the extrudates of C3, C4 and C5 were slightly thicker than Control groups C1 and C2, owning to the solid granules of chitosan and keratin. Comparing extrusion of C4 with C5 under 63±1 °C, C5 required less force to extrude than C4, which is attributable to the higher ratio of chitosan/keratin mixture in the polymer content of C5. Interestingly, higher amounts of chitosan/keratin in C5 were responsible for the extrudates to turn back to white again in colour. This may be due to the lower viscosity achieved from the addition of higher amount of chitosan/keratin, which shortened the remaining time of materials inside the extruder (Ma et al. 2017). In addition, the higher polymer content of chitosan/keratin potentially increased the melting point of the whole blend powder, which enhanced the cooling rate of the extrudates and enabled to extrude longer fibres (Flores‐Hernandez et al. 2018).

HME has been widely used as a drug delivery technology and also was chosen to manufacture suture products, due to its solvent free and environmentally friendly characteristics, as well as its potential to poorly disperse water-soluble drugs in polymers for enhanced delivery (Campbell, Craig, and McNally 2010). In the procedure, the API is embedded in a carrier system with other functional excipients, which can be melted under the required temperature (Patil, Tiwari, and Repka 2016). In addition, this drug can be homogeneously distributed into polymer matrix in a molecular level with this method (Liet al. 2013, Champeau et al. 2017).

In current study, the melting point of DP is about 302-310°C. However, since PCL (55°C) and PEG400 (4-8°C) have very low Tm. A heating temperature of 63 ±1 °C was employed for melt extrusion process and successfully produced fine extrudates. Other studies also found that the HME process temperature can be reduced by mixing drugs with polymers that have low Tm, in that not only can the melting point be reduced, the viscosity of polymers can also be decreased (Shah et al. 2013). This phenomenon may be explained by intermolecular

116

interactions lowering the chemical potential of the drug, and therefore the drug with higher

Tm can be extruded below its original melting temperature (Li et al. 2014). This concept has been demonstrated by the study conducted by Papadimitriou et al. (2012), in which they applied Felodipine as a model drug, and PVP/PEG as the drug carrier. The results indicated an increase in the miscibility of the PVP-drug system, owing to the presence of PEG which acted as plasticizer (Papadimitriou et al. 2012).

Other processing parameters, such as feeding rate and screw rotating speed were not investigated in depth, because the extruder used in this study was not able to control the rotating speed. However, other studies have looked at the effects of processing parameters with PEG as plasticiser in detail, in particular Barmpalexis et al. (2013) carried out a thorough investigation on the effects of the different temperatures, mixing times and rotating speeds on the solid dispersions of PVP/PEG/drug mixtures (Barmpalexis et al. 2013). Importantly, different room temperatures and compositions of polymer mixtures may alter the specific parameters identified in the current project for melt extrusion of drug-eluting sutures. Therefore, the parameters used here should be considered as a guideline for future studies and it is highly recommended that an exhaustive optimisation must be carried out for each new experiment.

5.1.2 Characterisation of Polymer Sutures

The FTIR spectra for neat PCL was similar to spectra reported elsewhere, showing characteristic peaks around 2943 cm-1, 2866 cm-1, 1721 cm-1, 1238 cm-1 and 1163 cm-1 corresponding to –C–H asymmetric and symmetric stretching, –C=O stretching of the ester carbonyl group and C–O–C asymmetric and symmetric stretching (Elzein et al. 2004, Edwards et al. 2015).There was a slightly rounder peak in PCL/PEG blend (1161 cm−1), which could be ascribed to the C-O and C-C stretching in the amorphous phase(Yu et al. 2014).

For PCL/PEG400 blend, PCL provided free carbonyl groups that acted as potential proton receptors, and free hydroxyl groups in PEG400 that acted as potential proton donors. Hydrogen bond interactions may occur in PCL-PEG400 blend (Zehnder et al. 2016, Elzein et al. 2004). Additionally, there was a peak at 3441 cm-1 in the PCL/PEG blend, which also confirmed the presence of –OH bonding in the mixture (Yu et al. 2014). However, there were

117

no shifts or new peaks visible in PCL-PEG400 blend (Figure 34, C2), probably due to the intermolecular interactions (Yu et al. 2014).

These spectral characteristics did not change with the addition of chitosan and keratin (Figure 34). Generally, Fourier Transform Infrared (FTIR) spectral analysis is used to investigate the chemical structure and the different types of bonding within a composite through recording the unique vibrational characteristics of specific molecules in response to a certain length of IR waveband (Steele 1971). However, FTIR was not able to detect the difference between PCL/PEG400 blend (C2) and PCL/PEG400/chitosan/keratin blend (C3, C4, C5). Possibly because the chitosan/keratin blend in the polymer matrix was too low (1, 2, 3 wt%, respectively) to be detected by FTIR. However, the peaks around 1720 cm-1 showed different intensities in chitosan-keratin contained fibres. The control group that contains the highest PEG concentration (C3, PEG=19 w/w) showed a higher intensity compared to the other two control groups with lower PEG concentration, corresponding to the higher amount of –C=O bonds (Tran and Mututuvari 2015). The peak at 1106 cm-1 in chitosan-keratin containing samples showed a higher intensity with the increase in the concentration of the chitosan/keratin blend. The slightly shifted and wider peak at 1160 cm-1 in the spectrum of C5 indicated the miscibility of the polymers and the presence of the hydrogen bonding interaction (Liu et al. 2006).

The overlay FTIR spectra of all the samples including control group C1, C2, C3, C4, C5 and chitosan, keratin alone as well as the chitosan/ keratin blend may be found in Figure 34. For the chitosan and keratin 50:50 blend, the bands appeared at 1700-1600 cm-1 and 1550 cm-1, representing amide C=O stretch (amide I) and C-N stretch (amide II) vibrations, respectively (Tran and Mututuvari 2015). The N-H stretch vibration (amide A) was exhibited at the 3280 cm-1 band. A band at 1300-1200 cm-1 can be attributed to the combination of the N-H bending between chitosan and keratin, as well as the C-N stretch vibrations (amide III) (Tran and Mututuvari 2015).

The varying structures of polymer networks and mechanical characteristics of the five control groups were also evident in the thermal behaviour and tensile testing. TGA showed that the 5 control groups all had a one-step degradation. Pure PCL started to degrade through ester pyrolysis and unzipping depolymerization reactions, resulting in the rupture of its chains (Ghorbani et al. 2016). Consistent with literature findings, PCL/chitosan blend was found to lower the degradation temperature due to the dehydration of saccharide ring of chitosan 118

(Ghorbani et al. 2016). Additionally, another study compared pure PCL with PCL/vital wheat gluten mixture showed that neat PCL had higher thermal stability (Mohamed et al. 2008). Therefore, PCL itself already has a very high degradation temperature. So, when adding PEG400 into PCL (C2), the degradation started at a much lower temperature (326 °C ). Nevertheless, the thermal stability of fibre samples was maintained the same as neat PCL, by adding the chitosan/keratin blend by 4 wt% into PCL/PEG400 blend (C5). For example, the degradation temperature of C5 (PCL: PEG400: chitosan/keratin= 80:16:4 w/w) was 378 °C, and C1 (pure PCL) started to degrade at 375 °C.

In this study, DSC measurements used the second heating and cooling DSC cycles, in order to eliminate the thermal history of PCL (Mohamed et al. 2008). DSC data was used to determine the degree of miscibility of polymer composites, the degree of intermolecular interactions between polymers, and the degree of crystallization. The presence of the single

Tg and single Tm for all the composites indicated the miscibility or interaction of molecules (Mohamed et al. 2008). Since PEG400 is a plasticizer, adding PEG400 in the PCL reduced the melting point and glass transition temperature. Moreover, the addition of chitosan/keratin blend decreased Tg even further, which can be ascribed to the formation of intermolecular hydrogen bonds between the PCL/PEG and chitosan/keratin in the amorphous phase, leading to a suppression of the crystallization of PCL/PEG (Ghorbani et al. 2016).

Tensile testing aptly reflected the mechanical properties of the fibres with five different formulations. The pure PCL (C1) has relatively low stiffness, but the addition of PEG400 (C2) (acts as plasticiser) was intended to improve ductility (Barmpalexis et al. 2013). However, the mixture of PCL/PEG400 had lower elongation at break (110%) than C1 (123%). This might be due to the blending ratio of PCL with PEG400, which requires future work. Chitosan has limitations in that it has very low tensile strength, which does not meet the requirements for many biomedical applications. For instance, in tissue engineering, chitosan alone cannot be used to mimic articular cartilage, since the tensile strength of a cartilage required is to be around 27 MP, whereas the tensile strength of chitosan is only 5-7 MPa when wet (Cheng et al. 2003). Keratin alone is also mechanically weak and is not able to maintain its structural integrity (Thompson et al. 2016). However, the composite of PCL/PEG400 (C2) showed lower stiffness compared to samples contained chitosan/keratin (C3, C4, C5). Adding chitosan/keratin (50: 50 w/w) (having high melting points) increased the fibre stiffness, resulting in the improvement of elastic modulus (Catanzano et al. 2014).

119

The elastic modulus was improved by 20% in samples that contained 1% chitosan/keratin (C3). The addition of chitosan/keratin also had a dramatic impact on the elongation of fibres. The elongation at break was reduced concomitantly with the addition of chitosan and keratin (67%, 21%, 20% for C3, C4, C5 respectively), potentially due to the formation of lumps on the fibre surfaces.

In summary, the analysis of the physical and mechanical characteristics of polymer sutures successfully identified the molecular interaction and hydrogen bonding (using FTIR) in the polymer matrix, as well as the plasticity effect contributed by PEG400. It appears that the control group C3 (PCL: PEG400: chitosan/keratin= 80:19:1 w/w) has the best potential for suture applications because it showed the miscibility of polymers, and had relatively stable thermal behaviour up to 348 °C . Most importantly, it formed strong and stiff fibres with proper elongation, which would be an ideal suture for soft tissue applications, such as muscles, tendons, ligaments and fascia (Guo and Ma 2014).

5.1.3 Polymer Composites as Drug Carriers

In this study, the drug content used for the polymer matrix was 5, 15 and 30 wt%, as this successfully allowed the extrusion of fine fibres at 63±1 °C. Even though there are many advantages of using the HME technique to produce drugs, the drug load needs careful consideration. Under most circumstances, the drug load is limited to 30% of the whole dosage (Claeys et al. 2015) since high drug load or the drug release would be uncontrollable with lower proportion of polymer incorporated in the formulation (Palazi et al. 2018). However, there are ways to optimize the drug load amount. For example, by using PCL (synthetic polymer) which has a relatively low melting point of 59-64°C, combining it with polymer PEG400 (which acts as a plasticizer) and chitosan/keratin (have high melting points), it will reduce manufacturing temperature to create a safe environment to produce fibres that contain drug DP (Barmpalexis et al. 2013).

According to previous studies, PCL is compatible with a wide range of pharmaceutical ingredients, which allows drug to distribute uniformly throughout the polymer matrix and to modify its physical, chemical and mechanical properties (Dash and Konkimalla 2012a). In this study, successful concentrations of chitosan/keratin loaded into PCL/PEG400 that could be extruded were ≤ 1%, and the concentration of drug DP in the polymer matrix was ≤

120

30%. Higher loading concentration of DP (≥ 30%) caused the cessation of any material being extruded from the nozzle. Our results were similar to those found by Vithani et al. (2014) who applied the hot melt extrusion method to produce lipid matrices which contain drug DP to achieve controlled release. They designed a drug content that varied from 30 to 40%. The extrudates produced from extruder were fine strands with 1 mm thickness. Our product produced fibres of a similar size of about 1mm, and also observed that a higher drug content led to low viscosity and improved flow behaviour under the hot melt extrusion process (Vithani et al. 2014).

5.2 Analysis of Drug Dissolution with Products Produced by Solid Dispersion Technique

5.2.1 Solid Dispersion

Over several decades, hot- melt extrusion has been used as a novel technology for manufacturing solid dispersions of active pharmaceutical ingredients into polymer matrices (Maniruzzaman et al. 2013). The high shear mixing of the molten mass allows the drug and polymers to be mixed at the molecular level to achieve drug-polymer interactions (Sarode et al. 2013). Solid dispersion form has already been reported to have the ability to control and extend the drug releasing time, resulting in improved bioavailability (Lakshman et al. 2008, Maniruzzaman, Rana, et al. 2013).

In this study, PCL, PEG400, chitosan, keratin and drug DP were melt-extruded into fibre form. The drug was dispersed in the polymer matrix with reduced particle size and agglomeration, forming an amorphous state to increase the available surface area and therefore, enhance the dissolution of the drug (Craig 2002, Leuner and Dressman 2000, Vasconcelos, Sarmento, and Costa 2007).There are two known, basic pathways for drug release from a polymer matrix, including drug diffusion and matrix degradation (Song, Labhasetwar, and Levy 1997). For example, PCL has good characteristics in both drug diffusion and matrix degradation, as it has been shown that it can provide sustained drug release due to its high permeability to many drugs (Dordunoo et al. 1997). The solid dispersion of melt-extruded PCL/PEG400/chitosan/keratin formed a biocompatible polymer

121

matrix, allowing for tailoring the degradation kinetics and drug release rates (Azimi et al. 2014).

PEG400 is a water-soluble polymer, which is widely used as a drug carrier in the preparation of solid dispersion systems was thus selected for this study. Previous studies employed paclitaxel as a drug carrier, but this has too poor a water solubility to be the model drug. The drug release rate between blends prepared by solid dispersion and blends prepared by physical mixtures showed that the dispersion of paclitaxel in PEG can eﬀectively modify drug release from PCL films (Shen et al. 2013). Formulation groups containing varying proportions of polymers (G1, G2 and G3) all observed drug release within 15 minutes (Figure 42-44), which can be ascribed to the brilliant effect of PEG400 on improving drug solubility.

These findings demonstrated that the solid dispersion significantly improved the solubility of drug inside the polymer matrix. However, in order to obtain the optimal formulation for drug release, the dissolution behaviour was also investigated.

5.2.2 Dissolution Behaviour

In the current project, drug content of 5, 15 and 30 wt% were blended into polymer composites of PCL/PEG400/chitosan/keratin, PCL/PEG400/chitosan and PCL/PEG400/keratin, forming formulation F1 to F9. According to the different polymer compounds, samples were classified into 3 groups. When the drug contents were at 5 wt%, the drug dissolved within the polymer medium first, followed by diffusing to the surface of the fibres (F1, F4, F7). Slightly higher drug loadings (10 wt%) allowed for the formation of cavities near the device surface during drug delivery. Fluid from the external environment may fill these cavities, which provides pathways for the remaining material and drug to be released from the device (F2, F5, F8). When the drug content reached 30 wt%, the formation of the cavities is enough to build a continuous channel to the surface of the device allowing drug diffusion through those channels (F3, F6, F9) (Douglas et al. 2010). Figure 45 demonstrates the release profiles of the F3, F6 and F9, with highest drug content (30 wt%). F9 (PCL/PEG/keratin/drug) exhibited a rapid release within the 1.75h, followed by a controlled release. Moreover, F3 (PCL/PEG/chitosan/keratin/drug) and F6 (PCL/PEG/chitosan/drug) showed a sustained drug release, but F3 showed more stable release rate than F6.

122

Typical drug delivery systems require three factors. The correct matrix structure to hold the drug delivery system together but not contribute to the drug diffusion during drug delivery. The aqueous solution, which is whatever aqueous material is that is coming inside the matrix structure from the external environment. Thirdly, the drug, which is diffuses from the inner matrix into the external environment (Douglas et al. 2010). Once inside the polymer matrix, the drug dissolution rate depends on the solubility and permeability of the drug within the polymer, as well as to the biodegradability of the polymer (Douglas et al. 2010). Once inside the polymer matrix, the drug dissolution rate depends on the solubility and permeability of the drug within the polymer, as well as to the biodegradability of the polymer (Douglas et al. 2010). For instance, hydrophilic active ingredients have better solubility in a hydrophilic polymer. Similarly, lipophilic active ingredients reveal better solubility in a hydrophobic polymer. Thus, adding hydrophilic/phobic polymers can effectively improve the solubility of a drug in an incompatible polymer (Fini et al. 1996).

In drug delivery studies, PCL showed a slow degradation rate attributed to its high crystallinity and hydrophobicity (Sahoo et al. 2010). However, because PEG400, a liquid polymer, was added in all the formulations and it has very short degradation time of 72 h (Browning et al. 2014). Therefore, besides the weight loss that was expected from the drug release, the weight loss observed in the suture can also be attributed to the addition of PEG400. This was confirmed by a previous study which investigated the effect that PEG has on the release profile of the drug diclofenac sodium after manufacturing with hot melt extrusion method. Researchers found that the weight loss of PLA/ PEG blends was mainly due to the dissolution of PEG (Chen et al. 2016). Another study reported that PEG-based fibres manufactured by hot melt extrusion method have greater release rate with the increased PEG content. Up to 60% of PEG was released within 24 h (Cheng, Lei, and Guo 2010).

Despite the fast degradation of PEG400 and its effect on the integrity of the fibres, the cavities formed by the release of PEG have the potential to improve drug release from the matrix (Cheng, Lei, and Guo 2010). Group G1 (including F1, F2, F3, contains PCL/PEG/chitosan/keratin) exhibited an increased release rate after 1.75 h. And the increase of dissolution speed was observed between 1.75 and 3.75 h for group G2 (includes F4, F5, F6, contains PCL/PEG/chitosan). For the highest drug content group G3 (includes F7, F8, F9, contains PCL/PEG/kratin), relatively fast release speed was demonstrated at time between 0.75h and 1.75 h. These periods may also be the time when PEG stared to degrade.

123

Zhang et al. (2016) compared the drug release from pure PCL with the PCL/PEG blends, which prepared by hot-melt extrusion method. They found that the pure PCL matrix released only 4% of drug in 7 days. In contrast, more than 80% of drug was released in 72 h for the polymer blends with 30% of hydrophilic components (Zhang et al. 2016). However, the current study did not look at the PEG degradation kinetics in depth, which requires future work.

In summary, PCL and PEG were used to serve as the main drug carriers, while chitosan and keratin served as additives. Generally, all the samples demonstrated slow drug release patterns and the addition of chitosan/keratin or chitosan, keratin alone significantly affect drug release rate. In the group 4 (F3, F6, F9) which contained highest drug content (30 wt%), F9 achieved sustained release after 3.75 h, but F3 exhibited slower dissolution behaviour in the whole experimental period (Figure 45). In some cases, when the acceleration of drug release rate is required, higher PEG content can be considered (Cheng, Lei, and Guo 2010). Since in the formulation of this study, when drug content up to 30% was employed, 19% of PEG could cause the drug to start to release within 0.25 h.

5.3 Analysis of Biocompatibility of the Sutures

An ideal material for a drug delivery matrix or implantable device should not release toxic products or trigger adverse reactions (Grehan et al. 2014). When using novel polymer blends with drugs, one must always be aware of potential unwanted side effects. Therefore, an important step in the development of a suture product is the assessment of any safety risk. Of particular interest are parameters of cytotoxicity and biocompatibility of the sutures. PCL, PEG and PCL-PEG products have been approved by the Food and Drug Administration (Grossen et al. 2017). However, chitosan and keratin have not been approved by the FDA for drug delivery, and the addition of drugs has often unknown side effects on human cells.

To investigate the biocompatibility of the suture products, in vitro toxicity testing is used. This test aims to investigate cytotoxicity, cytocompatibility, cell proliferation, cell adhesion and cell migration (Lendlein et al. 2010). Unfortunately, a single method to cover all the general toxicity in humans is not yet known. As such, several in vitro tests along with different biological properties were used in order to determine the potential toxicity of melt-

124

extruded sutures (Pizzoferrato et al. 1994, Kirkpatrick et al. 1998, Hanks, Wataha, and Sun 1996).

5.3.1 Effect of Drug-Eluting Sutures on Cell Proliferation

Suture biocompatibility tests are commonly performed done on tissues and cells in the lab as a starting point. For example, skin is composed of three main types of cells, including keratinocytes, melanocytes, and fibroblasts (Ji, Li, and Chen 2008). Human immortalized keratinocyte (HaCaT) cells are model cell lines which play an important role in epidermal tissue regeneration (Muniandy et al. 2018). HatCaT cells may turn into human keratinocytes with the same properties of basal epidermal keratinocytes (Meineke et al. 2004, Maas- Szabowski, Stärker, and Fusenig 2003).

HaCat cells were chosen for this study because of their availability, and their immortalized nature which allows them to be used for an extended amount of time compared to primary cell lines. In addition, HaCat cells grew faster than human dermal fibroblasts, such as Nhdf, which are located in the dermis of skin (Muniandy et al. 2018). In the current project, HaCat cells became confluent within one week during the cell culture process.

It was possible to conduct direct contact for MTT assay owing to the insoluble nature of the main polymer, PCL (Grehan et al. 2014). Three groups of drug-treated sutures were immersed in the solution containing HaCat cells. Amongst these three groups, Group 1 which contained chitosan/keratin (1:1 w/w) demonstrated a significant increase on cell numbers at 72 h (p = 0.0002). The result is consistent with the former findings when chitosan: keratin was 1: 1 w/w, the cell viability was significantly increased at day 3, compared to the other two ratios (chitosan: keratin= 1: 0.5 and 1:0.25 w/w, respectively) (Flores‐Hernandez et al. 2018).

However, experimental testing revealed that the different drug content combined with the polymer types promoted cell growth. For instance, in group 1 (containing PCL/PEG/chitosan/keratin), the lowest drug content (5 wt%) showed highest number of cells (~ 40,000) in the entire treated samples at day 3. In contrast, group 3 (containing PCL/PEG/ keratin and highest drug amount (30%) had around 30,000 cells.

125

HaCaT cells tend to grow on the hydrophilic surfaces as they are mainly attached to carbonyl and carboxyl groups (Lehocký et al. 2009, Lehocký et al. 2006, Peschel et al. 2008). Therefore, polar surface entities play an important role in cell growth factors. In addition, hydrogen bonding and van der Waals forces also contribute to improved cell attachment, and this reinforcement provides a bridge between samples and cells (Paleos, Tsiourvas, and Sideratou 2004) which may promote cell adhesion and cell proliferation (García et al. 2010).

In current study, the total formulation amount was consistent, and PCL content was fixed (80 wt%), which means that lower drug content was indicative of the higher amount of PEG400 in the formulations, which can be used to explain the highest cell number (~ 40,000) observed in F1 at 72 h. The relatively high cell number that was observed in F9 at 72 h was mainly attributable to varying surface properties of matrix after melt extruding with different drug content (Grehan et al. 2014), since melt extrusion created a product with a hydrophobic surface. Higher drug content with keratin may have modified the surface properties of fibres and thus improved cell attachment and proliferation as well (Edwards et al. 2015).

In summary, group 1 and group 3 demonstrated better performance than group 2 in the cell proliferation study. Therefore, Group 1 (F1, F2, F3) and Group 3 (F7, F8, F9) were selected for the investigation of HaCat cell viability.

5.3.2 Effect of Drug-Eluting Sutures on Cell Viability

Hacat cell viability of the melt-extruded fibres was analysed using the live/dead assay. Fluorescence staining was employed to visualise the cells that were under direct contact with 6 formulations respectively. At 24 h, suture materials F (13, 27) had an extremely significant increase (p < 0.0001) on the live percentage of cells. However, at 48 and 72 h, there were no significant differences between cell only, control group and drug embedded samples for HaCaT cell lines (p > 0.5). This indicated that the fibres containing of drug is not only non-toxic to the HaCat cells, but may also enhance the cell proliferation, which in turn may aid wound healing

Polymer sutures (C1, C2, C3) also showed high cell viability at 72 h, probably due to the main polymer content, PCL, which occupied 80 wt% of each formulation. PCL was also used as a main polymer by a previous study which incorporated PCL with milk protein by melt extrusion method. The biological testing showed that the materials had no cytotoxicity towards the keratinocytes (Ghosh et al. 2010). Another study employed chitosan/keratin at a

126

ratio of 1: 1 w/w, which is similar as this study, the results showing that the combination of chitosan/keratin significantly improved cell viability percentage (Flores‐Hernandez et al. 2018).

In summary, PCL/PEG400/chitosan/keratin (Group 1) and PCL/PEG400/keratin (Group 3) were investigated for their cytotoxicity with live/dead assay. Both groups showed very high cell viability at 48 and 72 h. Group 1 demonstrated slightly higher percentage of viable cells compared to group 3 at 72 h. In light of the literature findings, many studies have reported that PCL is non-toxic (Grossen et al. 2017, Ghosh et al. 2010). Additionally, hydrophilic property of PEG400 is preferable for the attachment of HaCat cells (Lehocký et al. 2009). Moreover, the biocompatibility of chitosan and keratin is well studied, suggesting their potential for enhancing cell proliferation (Flores‐Hernandez et al. 2018). Following these findings, the scratch assay was conducted to further investigate the migration speed of HaCats under the effect of drug embedded polymer sutures.

5.3.3 Effect of Drug-Eluting Sutures on Cell Migration

Cells migration is essential for many physiological processes, including tissue regeneration and wound healing (Balekar et al. 2012). As mentioned above, the scratch assay was carried on via direct method in order to study the possibility of interaction of the polymers and drug with the wound healing process. Following this rationale, a scratch was created in the keratinocyte monolayer, cell to cell contacts thus were broken down, resulting in the increase of the number of growth factors and cytokines at the edge of the wound (Liang, Park, and Guan 2007). Afterwards, keratinocytes (HaCats) started to proliferate and migrate to the wound site, which is the second phase of wound healing (Gurtner et al. 2008).

The cell only sample indicated the natural rate of migration of the cells, without the influence of the polymer materials and drug. And the control group represents the polymer only sutures. The entire treated group 1 even with the lowest concentration showed some significant increase in migration rate compared to cell only and control group.

Interestingly, the results showed that the suture contained 30 wt% of drug (F3) enhanced cell proliferation and migration ability, and also improved the capability of HaCat cells to close the wound by increasing the number of keratinocytes in the scratched area (Lendlein et al.

127

2010). It achieved 99% wound closure at 48 h. Therefore, the scratch assay further confirmed that F3 is the optimal formulation with the ideal biocompatibility.

5.4 Comparison of Suture Sample with Commercial Product

The comparation of the best performing suture sample F3 (PCL/PEG400/chitosan-keratin= 80/19/1 w/w), with commercially available suture Monocryl (poliglecaprone 25) which consists of both PGA and PCL (Bezwada et al. 1995), is shown in table 20. Suture F3 and Monocryl both demonstrated low tissue reaction. However, F3 exhibited much lower mechanical properties then Monocryl, which can be ascribed to the high drug loading rate. Interestingly, the elongation of F3 is much higher than commercial product Monocryl. Therefore, even though the tensile strength of the formulation suture in this study is lower than commercial product Monocryl, the drug-embedded properties and the high elongation characteristics carried by the suture F3 has more promising applications in clincal settings.

Table 20. The Comparison of Sample F3 with Commercial Product of Monocryl Properties F3 Monocryl Drug loading (wt%) 30 0 Diameter (mm) 0.6 0.38 Elongation at break (%) 66.0 39 Failure load (N) 5.5 71.82 Tissue reaction Low Low

5.5 Critical Analysis of Methodology and Limitations

The in vitro skin model was used for cytotoxicity analysis of the biomaterials and drug as a whole. In vitro models are predictable, they are easy to build and maintain, as well as a wide range of results can be provided. However, some limitations within in vitro models can inhibit the results. For example, the use of immortal cell lines may affect the assumption of the

128

independency of the cells in each well, when cells were plated at the same time and originated from the same cell culture flask. Due to the correlation among the wells on each plate, which would consider the data as pseudoreplicates (Ranstam 2012). Although in this study, even though the cells came from the same origin, the flasks were treated individually.

The reason that MTT assay was selected to investigate cell number on the drug embedded sutures was due to this assay being able to detect live cells via production of formazan crystals accurately. The formazan crystals are produced by mitochondrial reductases, which does not exist in dead cells. Even though this assay is relatively reliable, its limitations need to be considered before interpreting the results. The most significant limitation is that if the materials or active ingredients used in the experiment alter the mitochondrial signalling pathways in the cells, it may lead to the change of the amount of formazan crystals formation, which can cause the inaccurate results due to the false number of cells (Van Meerloo, Kaspers, and Cloos 2011). This could be why F2, F4 and F6 showed much higher cell number in the MTT assay at 24 h, however the incubation of the sutures alone without cells negated that possibility.

The wound healing assay was easy to conduct and inexpensive, and the results were rapidly produced. During this experiment, cells were moving in one direction to close the wound, which makes the results relatively predictable. Additionally, the velocity of the movement of the cells may be determined by recording the morphology of cells in real-time. However, one limitation is that the scratch tool used and the well sizes used may impact cell migration (Staton, Reed, and Brown 2009). However, this limitation was controlled for by using the same method across all different groups. Furthermore, because the scratches were made by hands, the size, shape, and the gap width between scratches can be variable every time once a new experiment starts. Similarly, this limitation was controlled for by having only one researcher create the scratches.

The limitation of the live/dead assay is that the calcein blue reacts to live cells and produces green fluorescent products that are trapped inside the cells. Propidium iodide can penetrate dead cell membranes and bind with nucleic acids to generate red fluorescence products. Live cells are assumed to be equipped with completely intact cell membranes, and thus PI will not react to them at all (Pollack and Ciancio 1990). If the cell membranes of live cells become accessible to PI dye, the dead cell number, therefore, will not be accurately calculated. However, PI dye has been widely used for HaCat cells studies, and the margin of error is 129

small enough that studies do not encounter issues, so this may be a problem for studies that manipulate the nature of cell walls rather than studies that test cell survivability (Hajem et al. 2013).

The limitations of this study are offset by having multiple assays that may cross reference each other. For instance, the combination of MTT and live/dead assay, not only could live cells number be calculated, the dead cell percentage could also be obtained. Furthermore, the combination of wound healing assay and the MTT assay could detect the drug effect on cell migration rate.

5.6 Future Directions

The results from this study can contribute a wealth of new knowledge to the field of bioengineering, particularly for surgical suture application – for example this suture type has great potential for wound healing, as was demonstrated by the increase of cell survival when the suture blend with 30% drug was used. This study also encourages the recruitment of chitosan/keratin as part of the suture material. However, there is still much left to explore. Specifically, the complete drug release time and the suture degradation time are unknown, and indeed PEG400 dissolution profile is not investigated. If PEG dissolution and suture degradation times (and drug release) are satisfactory, then one may try in vivo test to observe wound healing inside an animal model, this, more closely approximating the behaviour of the suture in a human body. Therefore, future studies should examine the PEG release pattern within an extrudate, and the time that a suture needs to degrade as well as the embedded drug needs to completely release, in order to back up and extend the findings of this current study.

Other research has also demonstrated the antimicrobial activity of chitosan and the anti- inflammation effect of diclofenac potassium (Catanzano et al. 2014, Kim et al. 2017, Leaper et al. 2017, Onesti, Carella, and Scuderi 2018). This effect was not investigated here because it was a study designed to see if the incorporation of PEG400/chitosan/keratin into PCL could improve suture performance and if the drug release profiles differ in drug/polymer ratios. Moreover, this study was only designed to investigate immediate cell effects, leaving future work to elucidate the method of action of the sutures over time. Therefore, future studies should seek to determine if the chitosan-containing sutures show any antimicrobial effect, and also to investigate if there is any anti-inflammatory effect of the diclofenac potassium

130

embedded sutures, using methods such as agar diffusion test (Reinbold et al. 2017), and in vivo experiments on mice (Catanzano et al. 2014).

5.7 Conclusion

In summary, this study sought to investigate and standardise a drug-eluting anti-inflammatory suture fabricated by the hot- melt extrusion method. This suture contained biomaterials including PCL, PEG400, chitosan and keratin, as well as diclofenac potassium. PCL was found to perform better if blended with other materials in powder form first, and then the addition of the aqueous polymer of PEG400, to avoid the formation of bulk lumps. Various formulations with and without drug could be extruded with fine fibres at 63±1 °C, longer fibres could be obtained with higher chitosan/keratin content, probably due to their high melting points.

All the extrudates showed a homogenous distribution of filler, which confirmed findings from FTIR and DSC studies. Hydrogen bond interactions existed in all the PEG400 containing samples, which is the main power allowing polymers to be mixed in the molecular level. From the TGA study, it was observed that the addition of chitosan/keratin reduced and stabilised decomposition rate. DSC results showed that all the samples presented a single Tg, indicating the miscibility or interaction of molecules. Moreover, the presence of PEG400 in the extrudates reduced the melting points of the polymer composite, accounting for the low melting point and glass transition temperature of PEG400. Mechanical testing indicated that the suture containing 1% chitosan/keratin blend had relatively good mechanical properties. The product was strong and stiff, with proper elongation, required a slightly higher strength to break.

Drug dissolution testing demonstrated that the addition of chitosan in the polymer-drug composites can tune the drug release profile in a controlled manner. Chitosan/keratin blend improved drug dissolution stability. When it comes to the biocompatibility of sutures, no sample showed cytotoxicity to human keratinocytes. Additionally, the suture which contained the polymer blend (PCL/PEG400/chitosan/keratin) with 30% of drug content exhibited the highest wound healing rate achieving 99% wound closure at 48 h.

Overall, the strategy used in this study has great potential since the hot melt extrusion method is relatively cheap, simple, and versatile It also can be scaled-up easily at an industrial level.

131

The product developed in this study would be the most suitable for use as a suture in the clinical field. Nevertheless, in vivo study should be further applied to the future work to back up the findings of this study.

132

REFERENCES

Adomaviciute, E., J. Baltusnikaite-Guzaitiene, V. Juskaite, M. Zilius, V. Briedis, and S. Stanys. 2018. "Formation and characterization of melt-spun polypropylene fibers with propolis for medical applications." Journal of the Textile Institute 109 (2):278-284. doi: 10.1080/00405000.2017.1341295.

Alvarez-Paino, M., A. Munoz-Bonilla, and M. Fernandez-Garcia. 2017. "Antimicrobial polymers in the nano-world." Nanomaterials (Basel) 7 (2):48-52. doi: 10.3390/nano7020048.

Arora, Anureet, Geeta Aggarwal, Janita Chander, Paramjot Maman, and Manju Nagpal. 2019. "Drug eluting sutures: a recent update." Journal of Applied Pharmaceutical Science 9 (07):111-123.

Augustine, R, and K Rajarathinam. 2012. "Synthesis and characterization of silver nanoparticles and its immobilization on alginate coated sutures for the prevention of surgical wound infections and the in vitro release studies." International Journal of Nano Dimension:205-212.

Azimi, Bahareh, Parviz Nourpanah, Mohammad Rabiee, Shahram Arbab, and Fibers. 2014. "Poly (ε-caprolactone) fiber: An overview." Journal of Engineered Fabrics 9 (3):455- 482.

Balekar, Neelam, Nadpi Gangadhar Katkam, Titpawan Nakpheng, Kholeeyoh Jehtae, and Teerapol Srichana. 2012. "Evaluation of the wound healing potential of Wedelia trilobata (L.) leaves." Journal of Ethnopharmacology 141 (3):817-824.

Barmpalexis, Panagiotis, Ioannis Koutsidis, Evangelos Karavas, Dimitra Louka, Sofia A Papadimitriou, and Dimitrios N Bikiaris. 2013. "Development of PVP/PEG mixtures as appropriate carriers for the preparation of drug solid dispersions by melt mixing technique and optimization of dissolution using artificial neural networks." European Journal of Pharmaceutics and Biopharmaceutics 85 (3):1219-1231.

Baygar, T., N. Sarac, A. Ugur, and I. R. Karaca. 2019. "Antimicrobial characteristics and biocompatibility of the surgical sutures coated with biosynthesized silver nanoparticles." Bioorganic Chemistry 86:254-258. doi: 10.1016/j.bioorg.2018.12.034.

Bezwada, Rao S, Dennis D Jamiolkowski, In-Young Lee, Vishvaroop Agarwal, Joseph Persivale, Susan Trenka-Benthin, Modesto Erneta, Jogendra Suryadevara, Alan Yang, and Sylvia Liu. 1995. "Monocryl® suture, a new ultra-pliable absorbable monofilament suture." Biomaterials 16 (15):1141-1148.

133

Bichon, Daniel, William Borloz, and Ana Lia Cassano-Zoppi. 1984. "In vivo evaluation of a new polyurethane-coated catgut suture." Biomaterials 5 (5):255-263.

Brianezi, Samira Faleiros Silva, Karine Cappuccio Castro, Rodolfo Debone Piazza, Maria do Socorro Fernandes Melo, Rafael Matsumoto Pereira, Rodrigo Fernando Costa Marques, and Maria Gabriela Nogueira Campos. 2018. "Preparation and characterization of chitosan/mPEG-PCL blended membranes for wound dressing and controlled gentamicin release." Materials Research 21 (6).

Browning, M. B., S. N. Cereceres, P. T. Luong, and E. M. Cosgriff-Hernandez. 2014. "Determination of the in vivo degradation mechanism of PEGDA hydrogels." Journal of Biomedical Materials Research Part A 102 (12):4244-4251. doi: 10.1002/jbm.a.35096.

Busse, Brittany. 2016. "Selecting materials for wound closure." In Wound Management in Urgent Care, 19-23. Springer.

Campbell, Kayleen T., Duncan Q. M. Craig, and Tony McNally. 2010. "Ibuprofen-loaded poly(ε-caprolactone) layered silicate nanocomposites prepared by hot melt extrusion." Journal of Materials Science: Materials in Medicine 21 (8):2307-2316. doi: 10.1007/s10856-009-3963-2.

Carey, Daniel E, Daniel H Zitomer, Anthony D Kappell, Melinda J Choi, Krassimira R Hristova, and Patrick J McNamara. 2016. "Chronic exposure to triclosan sustains microbial community shifts and alters antibiotic resistance gene levels in anaerobic digesters." Environmental Science: Processes & Impacts 18 (8):1060-1067.

Catanzano, Ovidio, Stefano Acierno, Pietro Russo, Mariarosaria Cervasio, Marialaura Del Basso De Caro, Adele Bolognese, Gilberto Sammartino, Luigi Califano, Gaetano Marenzi, and Antonio Calignano. 2014. "Melt-spun bioactive sutures containing nanohybrids for local delivery of anti-inflammatory drugs." Materials Science and Engineering: C 43:300-309.

Champeau, M., J. M. Thomassin, T. Tassaing, and C. Jerome. 2017. "Current manufacturing processes of drug-eluting sutures." Expert Opinion on Drug Delivery 14 (11):1293- 1303. doi: 10.1080/17425247.2017.1289173.

Chellamani, KP, D Veerasubramanian, and RV Balaji. 2013. "Surgical sutures: An overview." Journal of Academia Industrial Research 1 (12):778-782.

Chen, R., G. Li, A. Han, H. Wu, and S. Guo. 2016. "Controlled release of diclofenac sodium from polylactide acid-based solid dispersions prepared by hot-melt extrusion." Journal of Biomaterials Science, Polymer Edition 27 (6):529-543. doi: 10.1080/09205063.2016.1141273.

Chen, X., D. Hou, X. Tang, and L. Wang. 2015. "Quantitative physical and handling characteristics of novel antibacterial braided silk suture materials." Journal of the 134

Mechanical Behavior of Biomedical Materials 50:160-170. doi: 10.1016/j.jmbbm.2015.06.013.

Chen, X. J., D. D. Hou, L. Wang, Q. Zhang, J. H. Zou, and G. Sun. 2015. "Antibacterial surgical silk sutures using a high-performance slow-release carrier coating system." ACS Applied Materials & Interfaces 7 (40):22394-22403. doi: 10.1021/acsami.5b06239.

Chen, ZG, PW Wang, B Wei, XM Mo, and FZ Cui. 2010. "Electrospun collagen–chitosan nanofiber: A biomimetic extracellular matrix for endothelial cell and smooth muscle cell." Acta Biomaterialia 6 (2):372-382.

Cheng, L., L. Lei, and S. Guo. 2010. "In vitro and in vivo evaluation of praziquantel loaded implants based on PEG/PCL blends." International Journal of Pharmaceutics 387 (1-2):129-138. doi: 10.1016/j.ijpharm.2009.12.010.

Cheng, Mingyu, Jinguang Deng, Fei Yang, Yandao Gong, Nanming Zhao, and Xiufang Zhang. 2003. "Study on physical properties and nerve cell affinity of composite films from chitosan and gelatin solutions." Biomaterials 24 (17):2871-2880.

Cherif, Zine Eddine, Christophe Poilâne, Thierry Falher, Alexandre Vivet, Nabil Ouail, Bessem Ben Doudou, and Jun Chen. 2013. "Influence of textile treatment on mechanical and sorption properties of flax/epoxy composites." Polymer Composites 34 (10):1761-1773.

Chu, CC. 2013. "Materials for absorbable and nonabsorbable surgical sutures." In Biotextiles as Medical Implants, 275-334. Elsevier.

Chu, Chih-Chang, J Anthony von Fraunhofer, and Howard P Greisler. 1996. Wound Closure Biomaterials and Devices: CRC Press.

Claeys, Bart, Anouk Vervaeck, Xander KD Hillewaere, Sam Possemiers, Laurent Hansen, Thomas De Beer, Jean Paul Remon, and Chris Vervaet. 2015. "Thermoplastic polyurethanes for the manufacturing of highly dosed oral sustained release matrices via hot melt extrusion and injection molding." European Journal of Pharmaceutics and Biopharmaceutics 90:44-52.

Claude, O., T. Gregory, S. Montemagno, P. Bruneval, and E. H. Masmejean. 2007. "Vascular microanastomosis in rat femoral arteries: Experimental study comparing nonabsorbable and absorbable sutures." Journal of Reconstructive Microsurgery 23 (2):87-91. doi: 10.1055/s-2007-970188.

Craig, D. Q. 2002. "The mechanisms of drug release from solid dispersions in water-soluble polymers." International Journal of Pharmaceutics 231 (2):131-144. doi: 10.1016/s0378-5173(01)00891-2.

135

D’souza, Anisha A, and Ranjita Shegokar. 2016. "Polyethylene glycol (PEG): A versatile polymer for pharmaceutical applications." Expert Opinion on Drug Delivery 13 (9):1257-1275. da Silva, D., M. Kaduri, M. Poley, O. Adir, N. Krinsky, J. Shainsky-Roitman, and A. Schroeder. 2018. "Biocompatibility, biodegradation and excretion of polylactic acid (PLA) in medical implants and theranostic systems." Chemical Engineering Journal 340:9-14. doi: 10.1016/j.cej.2018.01.010.

Dash, Tapan K, and V Badireenath Konkimalla. 2012a. "Poly-є-caprolactone based formulations for drug delivery and tissue engineering: A review." Journal of Controlled Release 158 (1):15-33.

Dash, Tapan K, and V Badireenath Konkimalla. 2012b. "Polymeric modification and its implication in drug delivery: Poly-ε-caprolactone (PCL) as a model polymer." Molecular Pharmaceutics 9 (9):2365-2379.

Debbabi, Faten, and Saber Ben Abdessalem. 2015. "Effect of manufacturing conditions on structural and handling properties of braided polyamide suture." Journal of Engineered Fibers Fabrics 10 (3):155892501501000306.

Deliaert, A. E., E. Van den Kerckhove, S. Tuinder, S. Fieuws, J. H. Sawor, M. A. Meesters- Caberg, and R. R. van der Hulst. 2009. "The effect of triclosan-coated sutures in wound healing. A double blind randomised prospective pilot study." Journal of Plastic, Reconstructive & Aesthetic Surgery 62 (6):771-773. doi: 10.1016/j.bjps.2007.10.075.

Dennis, Christopher, Swaminathan Sethu, Sunita Nayak, Loganathan Mohan, Yosry Morsi, and Geetha Manivasagam. 2016. "Suture materials—Current and emerging trends." Journal of Biomedical Materials Research Part A 104 (6):1544-1559.

Dhom, J., D. A. Bloes, A. Peschel, and U. K. Hofmann. 2017. "Bacterial adhesion to suture material in a contaminated wound model: Comparison of monofilament, braided, and barbed sutures." Journal of Orthopaedic Research 35 (4):925-933. doi: 10.1002/jor.23305.

Dordunoo, SK, AMC Oktaba, W Hunter, W Min, T Cruz, and HM Burt. 1997. "Release of taxol from poly (ϵ-caprolactone) pastes: effect of water-soluble additives." Journal of Controlled Release 44 (1):87-94.

Douglas, P, G Andrews, D Jones, and G Walker. 2010. "Analysis of in vitro drug dissolution from PCL melt extrusion." Chemical Engineering Journal 164 (2-3):359-370.

Duarah, Rituparna, Yogendra P Singh, Prerak Gupta, Biman B Mandal, and Niranjan Karak. 2018. "Smart self-tightening surgical suture from a tough bio-based hyperbranched polyurethane/reduced carbon dot nanocomposite." Biomedical Materials 13 (4):045004. 136

Dubas, Stephan Thierry, Sirorat Wacharanad, and Pranut Potiyaraj. 2011. "Tunning of the antimicrobial activity of surgical sutures coated with silver nanoparticles." Colloids Surfaces A: Physicochemical Engineering Aspects 380 (1-3):25-28.

Edwards, Angela, David Jarvis, Tracy Hopkins, Sarah Pixley, and Narayan Bhattarai. 2015. "Poly (ε‐caprolactone)/keratin‐based composite nanofibers for biomedical applications." Journal of Biomedical Materials Research Part B: Applied Biomaterials 103 (1):21-30.

Elzein, Tamara, Mohamad Nasser-Eddine, Christelle Delaite, Sophie Bistac, and Philippe Dumas. 2004. "FTIR study of polycaprolactone chain organization at interfaces." Journal of Colloid and Interface Science 273 (2):381-387.

Ercan, U. K., F. Ibis, C. Dikyol, N. Horzum, O. Karaman, C. Yildirim, E. Cukur, and E. A. Demirci. 2018. "Prevention of bacterial colonization on non-thermal atmospheric plasma treated surgical sutures for control and prevention of surgical site infections." PLoS One 13 (9):e0202703. doi: 10.1371/journal.pone.0202703.

Erçin, Ersin, and Mustafa Karahan. 2018. "Literature review of suture materials." In Knots in Orthopedic Surgery, 177-180. Springer.

Fini, A., G. Fazio, M. J. F. Hervas, M. A. Holgado, and A. M. Rabasco. 1996. "Factors governing the dissolution of diclofenac salts." European Journal of Pharmaceutical Sciences 4 (4):231-238. doi: Doi 10.1016/0928-0987(95)00051-8.

Flores‐Hernandez, Cynthia G, Ana L Martinez‐Hernandez, Arturo Colin‐Cruz, Fernando Martinez‐Bustos, Víctor M Castaño, Imelda Olivas‐Armendariz, Armando Almendarez‐Camarillo, and Carlos Velasco‐Santos. 2018. "Starch modified with chitosan and reinforced with feather keratin materials produced by extrusion process: an alternative to starch polymers." Starch‐Stärke 70 (11-12):1700295.

Fong, TC, N Saba, CK Liew, R De Silva, M Enamul Hoque, and KL Goh. 2015. "Yarn flax fibres for polymer-coated sutures and hand layup polymer composite laminates." In Manufacturing of natural fibre reinforced polymer composites, 155-175. Springer.

Gabel, Elizabeth A, Gloria P Jimenez, William H Eaglstein, Francisco A Kerdel, and Vincent Falanga. 2000. "Performance comparison of nylon and an absorbable suture material (Polyglactin 910) in the closure of punch biopsy sites." Dermatologic surgery 26 (8):750-753.

García, Jorge López, Ahmad Asadinezhad, Jiří Pacherník, Marián Lehocký, Ita Junkar, Petr Humpolíček, Petr Sáha, and Pavel Valášek. 2010. "Cell proliferation of HaCaT keratinocytes on collagen films modified by argon plasma treatment." Molecules 15 (4):2845-2856.

Ghorbani, Fereshte Mohammad, Babak Kaffashi, Parvin Shokrollahi, Shahin Akhlaghi, and

137

Mikael S Hedenqvist. 2016. "Effect of hydroxyapatite nano-particles on morphology, rheology and thermal behavior of poly (caprolactone)/chitosan blends." Materials Science and Engineering: C 59:980-989.

Ghosh, Arun, M Azam Ali, Luxmanan Selvanesan, and George J Dias. 2010. "Structure– function characteristics of the biomaterials based on milk-derived proteins." International Journal Of Biological Macromolecules 46 (4):404-411.

Greenberg, J. A., and R. H. Goldman. 2013. "Barbed suture: A review of the technology and clinical uses in obstetrics and gynecology." Review of Obstetrics & Gynecology 6 (3- 4):107-115.

Greenberg, James A. 2010. "The use of barbed sutures in obstetrics and gynecology." Reviews in Obstetrics Gynecology 3 (3):82.

Grehan, Laura, John A Killion, Declan M Devine, Elaine K Kenny, Sinead Devery, Clement L Higginbotham, and Luke M Geever. 2014. "The development of hot melt extruded biocompatible controlled release drug delivery devices." International Journal of Polymeric Materials and Polymeric Biomaterials 63 (9):476-485.

Grossen, P., D. Witzigmann, S. Sieber, and J. Huwyler. 2017. "PEG-PCL-based nanomedicines: A biodegradable drug delivery system and its application." Journal of Controlled Release 260:46-60. doi: 10.1016/j.jconrel.2017.05.028.

Guo, BaoLin, and Peter X Ma. 2014. "Synthetic biodegradable functional polymers for tissue engineering: a brief review." Science China Chemistry 57 (4):490-500.

Gurtner, Geoffrey C, Sabine Werner, Yann Barrandon, and Michael T Longaker. 2008. "Wound repair and regeneration." Nature 453 (7193):314-321.

Hajem, N, A Chapelle, J Bignon, A Pinault, J‐M Liu, N Salah‐Mohellibi, E Lati, and J Wdzieczak‐Bakala. 2013. "The regulatory role of the tetrapeptide A c SDKP in skin and hair physiology and the prevention of ageing effects in these tissues–a potential cosmetic role." International journal of cosmetic science 35 (3):286-298.

Han, Changwoo, Youn-Hee Lim, and Yun-Chul Hong. 2016. "Ten-year trends in urinary concentrations of triclosan and benzophenone-3 in the general US population from 2003 to 2012." Environmental Pollution 208:803-810.

Hanks, Carl T, John C Wataha, and Zhilin Sun. 1996. "In vitro models of biocompatibility: a review." Dental Materials 12 (3):186-193.

Harnet, J. C., E. Le Guen, V. Ball, H. Tenenbaum, J. Ogier, Y. Haikel, and C. Vodouhe. 2009. "Antibacterial protection of suture material by chlorhexidine-functionalized polyelectrolyte multilayer films." Journal of Materials Science: Materials in Medicine 20 (1):185-193. doi: 10.1007/s10856-008-3559-2.

138

He, C. L., Z. M. Huang, and X. J. Han. 2009. "Fabrication of drug-loaded electrospun aligned fibrous threads for suture applications." Journal of Biomedical Materials Research Part A 89 (1):80-95. doi: 10.1002/jbm.a.32004.

He, Y., Z. W. Hu, M. D. Ren, C. K. Ding, P. Chen, Q. Gu, and Q. Wu. 2014. "Evaluation of PHBHHx and PHBV/PLA fibers used as medical sutures." Journal of Materials Science-Materials in Medicine 25 (2):561-571. doi: 10.1007/s10856-013-5073-4.

Henry-Stanley, Michelle J, Donavon J Hess, Aaron MT Barnes, Gary M Dunny, and Carol L Wells. 2010. "Bacterial contamination of surgical suture resembles a biofilm." Surgical infections 11 (5):433-439.

Herrmann, J. B. 1971. "Tensile strength and knot security of surgical suture materials." The American Surgeon 37 (4):209-217.

Hirano, Shigehiro, Min Zhang, Masuo Nakagawa, and Teruo %J Biomaterials Miyata. 2000. "Wet spun chitosan–collagen fibers, their chemical N-modifications, and blood compatibility." Biomaterials 21 (10):997-1003.

Hochberg, J., K. M. Meyer, and M. D. Marion. 2009. "Suture choice and other methods of skin closure." Surgical Clinics of North America 89 (3):627-641. doi: 10.1016/j.suc.2009.03.001.

Hossain, Kazi M Zakir, Andrew J Parsons, Chris D Rudd, Ifty Ahmed, and Wim Thielemans. 2014. "Mechanical, crystallisation and moisture absorption properties of melt drawn polylactic acid fibres." European Polymer Journal 53:270-281.

Hu, W., Z. M. Huang, and X. Y. Liu. 2010. "Development of braided drug-loaded nanofiber sutures." Nanotechnology 21 (31):315104. doi: 10.1088/0957-4484/21/31/315104.

Im, Jung Nam, Jeong Kyung Kim, Hyun-Kyoon Kim, Chang Hoon In, Kuen Yong Lee, and Won Ho Park. 2007. "In vitro and in vivo degradation behaviors of synthetic absorbable bicomponent monofilament suture prepared with poly (p-dioxanone) and its copolymer." Polymer degradation stability 92 (4):667-674.

Intra, J., X. Q. Zhang, R. L. Williams, X. Zhu, A. D. Sandler, and A. K. Salem. 2011. "Immunostimulatory sutures that treat local disease recurrence following primary tumor resection." Biomedical Materials 6 (1):011001. doi: 10.1088/1748- 6041/6/1/011001.

Ji, Yan, Xiao-Tao Li, and Guo-Qiang Chen. 2008. "Interactions between a poly (3- hydroxybutyrate-co-3-hydroxyvalerate-co-3-hydroxyhexanoate) terpolyester and human keratinocytes." Biomaterials 29 (28):3807-3814.

Joseph, Blessy, Anne George, Sreeraj Gopi, Nandakumar Kalarikkal, and Sabu Thomas. 2017. "Polymer sutures for simultaneous wound healing and drug delivery–A review."

139

International Journal of Pharmaceutics 524 (1-2):454-466.

Justinger, Christoph, Mohammed Reza Moussavian, Christian Schlueter, Berit Kopp, Otto Kollmar, and Martin Karl Schilling. 2009. "Antibiotic coating of abdominal closure sutures and wound infection." Surgery 145 (3):330-334.

Kashiwabuchi, F., K. S. Parikh, R. Omiadze, S. Zhang, L. Luo, H. V. Patel, Q. Xu, L. M. Ensign, H. Q. Mao, J. Hanes, and P. J. McDonnell. 2017. "Development of absorbable, antibiotic-eluting sutures for ophthalmic surgery." Translational Vision Science and Technology 6 (1):1. doi: 10.1167/tvst.6.1.1.

Kim, H., B. H. Kim, B. K. Huh, Y. C. Yoo, C. Y. Heo, Y. B. Choy, and J. H. Park. 2017. "Surgical suture releasing macrophage-targeted drug-loaded nanoparticles for an enhanced anti-inflammatory effect." Biomaterials Science 5 (8):1670-1677. doi: 10.1039/c7bm00345e.

Kirkpatrick, CJ, F Bittinger, M Wagner, H Köhler, TG Van Kooten, CL Klein, and M Otto. 1998. "Current trends in biocompatibility testing." Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine 212 (2):75-84.

Krishnamoorthy, Ganesan, Rajendran Selvakumar, Thotapalli Parvathaleswara Sastry, Sayeed Sadulla, Asit Baran Mandal, and Mukesh Doble. 2014. "Experimental and theoretical studies on Gallic acid assisted EDC/NHS initiated crosslinked collagen scaffolds." Materials Science Engineering: C 43:164-171.

Lakshman, J. P., Y. Cao, J. Kowalski, and A. T. Serajuddin. 2008. "Application of melt extrusion in the development of a physically and chemically stable high-energy amorphous solid dispersion of a poorly water-soluble drug." Molecular Pharmaceutics 5 (6):994-1002. doi: 10.1021/mp8001073.

Langley-Hobbs, SJ. 2013. "Sutures and general surgical implants." Feline Soft Tissue General Surgery E-Book:105.

Leaper, D., O. Assadian, N. O. Hubner, A. McBain, T. Barbolt, S. Rothenburger, and P. Wilson. 2011. "Antimicrobial sutures and prevention of surgical site infection: Assessment of the safety of the antiseptic triclosan." International Wound Journal 8 (6):556-566. doi: 10.1111/j.1742-481X.2011.00841.x.

Leaper, D., P. Wilson, O. Assadian, C. Edmiston, M. Kiernan, A. Miller, G. Bond-Smith, and J. Yap. 2017. "The role of antimicrobial sutures in preventing surgical site infection." Annals of the Royal College of Surgeons of England 99 (6):439-443. doi: 10.1308/rcsann.2017.0071.

Lee, Du-Hyeong, Tae-Yub Kwon, Kyo-Han Kim, Soon-Taek Kwon, Dae-Hyun Cho, Soon Ho Jang, Jun Sik Son, and Kyu-Bok Lee. 2014. "Anti-inflammatory drug releasing absorbable surgical sutures using poly (lactic-co-glycolic acid) particle carriers." Polymer Bulletin 71 (8):1933-1946. 140

Lee, E. J., B. K. Huh, S. N. Kim, J. Y. Lee, C. G. Park, A. G. Mikos, and Y. Bin Choy. 2017. "Application of materials as medical devices with localized drug delivery capabilities for enhanced wound repair." Progress in Materials Science 89:392-410. doi: 10.1016/j.pmatsci.2017.06.003.

Lee, J. E., S. Park, M. Park, M. H. Kim, C. G. Park, S. H. Lee, S. Y. Choi, B. H. Kim, H. J. Park, J. H. Park, C. Y. Heo, and Y. B. Choy. 2013. "Surgical suture assembled with polymeric drug-delivery sheet for sustained, local pain relief." Acta Biomaterialia 9 (9):8318-8327. doi: 10.1016/j.actbio.2013.06.003.

Lehocký, Marián, PFF Amaral, MAZ Coelho, P Sťahel, AM Barros-Timmons, and JAP Coutinho. 2006. "Attachment/detachment of Saccharomyces cerevisiae on plasma deposited organosilicon thin films." Czechoslovak Journal of Physics 56 (2):B1256.

Lehocký, Marián, Pavel Sťahel, Marek Koutný, Jan Čech, Jakub Institoris, and Aleš Mráček. 2009. "Adhesion of Rhodococcus sp. S3E2 and Rhodococcus sp. S3E3 to plasma prepared Teflon-like and organosilicon surfaces." Journal of Materials Processing Technology 209 (6):2871-2875.

Lendlein, Andreas, Marc Behl, Bernhard Hiebl, and Christian Wischke. 2010. "Shape- memory polymers as a technology platform for biomedical applications." Expert Review of Medical Devices 7 (3):357-379.

Leuner, C., and J. Dressman. 2000. "Improving drug solubility for oral delivery using solid dispersions." European Journal of Pharmaceutics and Biopharmaceutics 50 (1):47- 60. doi: Doi 10.1016/S0939-6411(00)00076-X.

Li, D., G. Guo, X. Deng, R. Fan, Q. Guo, M. Fan, J. Liang, F. Luo, and Z. Qian. 2013. "PLA/PEG-PPG-PEG/dexamethasone implant prepared by hot-melt extrusion for controlled release of immunosuppressive drug to implantable medical devices, Part 2: in vivo evaluation." Drug Delivery 20 (3-4):134-142. doi: 10.3109/10717544.2013.801049.

Li, D., G. Guo, R. Fan, J. Liang, X. Deng, F. Luo, and Z. Qian. 2013. "PLA/F68/dexamethasone implants prepared by hot-melt extrusion for controlled release of anti-inflammatory drug to implantable medical devices: I. Preparation, characterization and hydrolytic degradation study." International Journal of Pharmaceutics 441 (1-2):365-372. doi: 10.1016/j.ijpharm.2012.11.019.

Li, Yongcheng, Huishi Pang, Zhefei Guo, Ling Lin, Yixuan Dong, Ge Li, Ming Lu, and Chuangbin Wu. 2014. "Interactions between drugs and polymers influencing hot melt extrusion." Journal of Pharmacy and Pharmacology 66 (2):148-166.

Liang, C. C., A. Y. Park, and J. L. Guan. 2007. "In vitro scratch assay: a convenient and inexpensive method for analysis of cell migration in vitro." Nature protocols 2 (2):329-330.

141

Liu, Guoqin, Chunlong Guan, Hesheng Xia, Fuquan Guo, Xiaobin Ding, and Yuxing Peng. 2006. "Novel shape‐memory polymer based on hydrogen bonding." Macromolecular Rapid Communications 27 (14):1100-1104.

Liu, S. Q., G. H. Wu, X. F. Zhang, J. J. Yu, M. F. Liu, Y. Zhang, P. Wang, X. L. Yin, J. Zhang, F. Li, and M. Zhang. 2019. "Preparation and properties of poly (lactic acid) (PLA) suture loaded with PLA microspheres enclosed drugs (PM-Ds)." Journal of the Textile Institute 110 (11):1-10. doi: 10.1080/00405000.2019.1610999.

Liu, Shuqiang, Juanjuan Yu, Huimin Li, Kaiwen Wang, Gaihong Wu, Bowen Wang, Mingfang Liu, Yao Zhang, Peng Wang, and Jie Zhang. 2020. "Controllable drug release behavior of polylactic acid (PLA) surgical suture coating with ciprofloxacin (CPFX)—Polycaprolactone (PCL)/Polyglycolide (PGA)." Polymers 12 (2):288-292.

Lowe, A., W. J. Deng, D. W. Smith, and K. J. Balkus. 2014. "Coated melt-spun acrylonitrile- based suture for delayed release of nitric oxide." Materials Letters 125:221-223. doi: 10.1016/j.matlet.2014.03.174.

Ma, B. M., W. X. Chen, X. Qiao, G. W. Pan, W. Jakpa, X. L. Hou, and Y. Q. Yang. 2017. "Tunable wettability and tensile strength of chitosan membranes using keratin microparticles as reinforcement." Journal of Applied Polymer Science 134 (14). doi: 10.1002/App.44667.

Maas-Szabowski, Nicole, Anja Stärker, and Norbert E Fusenig. 2003. "Epidermal tissue regeneration and stromal interaction in HaCaT cells is initiated by TGF-α." Journal of cell science 116 (14):2937-2948.

Makino, Kimiko, Masayuki Arakawa, and Tamotsu Kondo. 1985. "Preparation and in vitro degradation properties of polylactide microcapsules." Chemical Pharmaceutical Bulletin 33 (3):1195-1201.

Malikmammadov, E., T. E. Tanir, A. Kiziltay, V. Hasirci, and N. Hasirci. 2018. "PCL and PCL-based materials in biomedical applications." Journal of Biomaterials Science, Polymer Edition 29 (7-9):863-893. doi: 10.1080/09205063.2017.1394711.

Maniruzzaman, M., M. M. Rana, J. S. Boateng, J. C. Mitchell, and D. Douroumis. 2013. "Dissolution enhancement of poorly water-soluble APIs processed by hot-melt extrusion using hydrophilic polymers." Drug Development and Industrial Pharmacy 39 (2):218-227. doi: 10.3109/03639045.2012.670642.

Maniruzzaman, Mohammed, David J Morgan, Andrew P Mendham, Jiayun Pang, Martin J Snowden, and Dennis Douroumis. 2013. "Drug–polymer intermolecular interactions in hot-melt extruded solid dispersions." International Journal of Pharmaceutics 443 (1-2):199-208.

Marturello, Danielle M, Michael S McFadden, R Avery Bennett, Guillaume R Ragetly, and

142

Gavin Horn. 2014. "Knot security and tensile strength of suture materials." Veterinary Surgery 43 (1):73-79.

Matl, F. D., J. Zlotnyk, A. Obermeier, W. Friess, S. Vogt, H. Buchner, H. Schnabelrauch, A. Stemberger, and K. D. Kuhn. 2009. "New anti-infective coatings of surgical sutures based on a combination of antiseptics and fatty acids." Journal of Biomaterials Science, Polymer Edition 20 (10):1439-1449. doi: 10.1163/092050609X12457418973107.

Meineke, Viktor, Kerstin Müller, Roland Ridi, Nils Cordes, Frank-Michael Köhn, Artur Mayerhofer, Johannes Ring, and Dirk van Beuningen. 2004. "Development and evaluation of a skin organ model for the analysis of radiation effects." Strahlentherapie und Onkologie 180 (2):102-108.

Middleton, John C, and AI Tipton. 1998. "Synthetic biodegradable polymers as medical devices." Medical Plastic Biomaterials 5:30-39.

Mohamed, Abdellatif, VL Finkenstadt, Sherald H Gordon, Girma Biresaw, Debra E Palmquist, and Patricia Rayas‐Duarte. 2008. "Thermal properties of PCL/gluten bioblends characterized by TGA, DSC, SEM, and infrared‐PAS." Journal of Applied Polymer Science 110 (5):3256-3266.

Muniandy, K., S. Gothai, W. S. Tan, S. S. Kumar, N. Mohd Esa, G. Chandramohan, K. S. Al- Numair, and P. Arulselvan. 2018. "In vitro wound healing potential of stem extract of alternanthera sessilis." Evidence-Based Complementary and Alternative Medicine 2018:3142073. doi: 10.1155/2018/3142073.

Muzzarelli, RAA, and C Muzzarelli. 2005. "Chitosan chemistry: relevance to the biomedical sciences." In Polysaccharides I, 151-209. Springer.

Naleway, Steven E, William Lear, Jamie J Kruzic, and Cory B Maughan. 2015. "Mechanical properties of suture materials in general and cutaneous surgery." Journal of Biomedical Materials Research Part B: Applied Biomaterials 103 (4):735-742.

Obermeier, A., J. Schneider, P. Fohr, S. Wehner, K. D. Kuhn, A. Stemberger, M. Schieker, and R. Burgkart. 2015. "In vitro evaluation of novel antimicrobial coatings for surgical sutures using octenidine." BMC Microbiol 15 (1):186-190. doi: 10.1186/s12866-015-0523-4.

Obermeier, Andreas, Jochen Schneider, Norbert Harrasser, Jutta Tübel, Heinrich Mühlhofer, Dominik Pförringer, Constantin von Deimling, Peter Foehr, Barbara Kiefel, and Christina Krämer. 2018. "Viable adhered Staphylococcus aureus highly reduced on novel antimicrobial sutures using chlorhexidine and octenidine to avoid surgical site infection (SSI)." PLoS One 13 (1):e0190912.

Onesti, M. G., S. Carella, and N. Scuderi. 2018. "Effectiveness of antimicrobial-coated

143

sutures for the prevention of surgical site infection: a review of the literature." European Review for Medical and Pharmacological Sciences 22 (17):5729-5739. doi: 10.26355/eurrev_201809_15841.

Padmakumar, S., B. Paul-Prasanth, K. Pavithran, D. K. Vijaykumar, A. Rajanbabu, T. B. Sivanarayanan, E. Kadakia, M. M. Amiji, S. V. Nair, and D. Menon. 2019. "Long- term drug delivery using implantable electrospun woven polymeric nanotextiles." Nanomedicine 15 (1):274-284. doi: 10.1016/j.nano.2018.10.002.

Padmakumar, Smrithi, John Joseph, Madhuri Harsha Neppalli, Sumi Elizabeth Mathew, Shantikumar V Nair, Sahadev A Shankarappa, Deepthy Menon, and interfaces. 2016. "Electrospun polymeric core–sheath yarns as drug eluting surgical sutures." ACS Applied Materials & Interfaces 8 (11):6925-6934.

Palazi, E., E. Karavas, P. Barmpalexis, M. Kostoglou, S. Nanaki, E. Christodoulou, and D. N. Bikiaris. 2018. "Melt extrusion process for adjusting drug release of poorly water soluble drug felodipine using different polymer matrices." European Journal of Pharmaceutical Sciences 114:332-345. doi: 10.1016/j.ejps.2018.01.004.

Paleos, Constantinos M, Dimitris Tsiourvas, and Zili Sideratou. 2004. "Hydrogen bonding interactions of liposomes simulating cell-cell recognition. A review." Origins of Life and Evolution of the Biosphere 34 (1-2):195-213.

Papadimitriou, Sofia A, Panagiotis Barmpalexis, Evangelos Karavas, and Dimitrios N Bikiaris. 2012. "Optimizing the ability of PVP/PEG mixtures to be used as appropriate carriers for the preparation of drug solid dispersions by melt mixing technique using artificial neural networks: I." European Journal of Pharmaceutics and Biopharmaceutics 82 (1):175-186.

Patil, H., R. V. Tiwari, and M. A. Repka. 2016. "Hot-melt extrusion: From theory to application in pharmaceutical formulation." AAPS PharmSciTech 17 (1):20-42. doi: 10.1208/s12249-015-0360-7.

Perioli, L., V. Ambrogi, L. di Nauta, M. Nocchetti, and C. Rossi. 2011. "Effects of hydrotalcite-like nanostructured compounds on biopharmaceutical properties and release of BCS class II drugs: The case of flurbiprofen." Applied Clay Science 51 (4):407-413. doi: 10.1016/j.clay.2010.12.019.

Perioli, L., T. Posati, M. Nocchetti, F. Bellezza, U. Costantino, and A. Cipiciani. 2011. "Intercalation and release of antiinflammatory drug diclofenac into nanosized ZnAl hydrotalcite-like compound." Applied Clay Science 53 (3):374-378. doi: 10.1016/j.clay.2010.06.028.

Peschel, Gundela, Hans‐Martin Dahse, Anke Konrad, Gerhard Dieter Wieland, Peter‐ Juergen Mueller, David P Martin, and Martin Roth. 2008. "Growth of keratinocytes on porous films of poly (3‐hydroxybutyrate) and poly (4‐hydroxybutyrate) blended with hyaluronic acid and chitosan." Journal of Biomedical Materials Research Part 144

A 85 (4):1072-1081.

Pillai, Chennakkattu Krishna Sadasivan, and Chandra P Sharma. 2010. "Absorbable polymeric surgical sutures: Chemistry, production, properties, biodegradability, and performance." Journal of Biomaterials Applications 25 (4):291-366.

Pizzoferrato, Arturo, Gabriela Ciapetti, Susanna Stea, Elisabetta Cenni, Carla Renata Arciola, and Donatella Granchi. 1994. "Cell culture methods for testing biocompatibility." Clinical Materials 15 (3):173-190.

Pollack, Alan, and Gaetano Ciancio. 1990. "Cell cycle phase-specific analysis of cell viability using Hoechst 33342 and propidium iodide after ethanol preservation." In Methods in Cell Biology, 19-24. Elsevier.

Ranstam, Jonas. 2012. Repeated measurements, bilateral observations and pseudoreplicates, why does it matter? : Elsevier.

Ratner, Buddy D, Allan S Hoffman, Frederick J Schoen, and Jack E Lemons. 2004. Biomaterials Science: An Introduction to Materials in Medicine: Elsevier.

Reinbold, J., A. K. Uhde, I. Muller, T. Weindl, J. Geis-Gerstorfer, C. Schlensak, H. P. Wendel, and S. Krajewski. 2017. "Preventing surgical site infections using a natural, biodegradable, antibacterial coating on surgical sutures." Molecules 22 (9):1570- 1571. doi: 10.3390/molecules22091570.

Rouse, Jillian G, and Mark E Van Dyke. 2010. "A review of keratin-based biomaterials for biomedical applications." Materials 3 (2):999-1014.

Sahoo, Sarmila, Abhisek Sasmal, Debasish Sahoo, and Padmalochan Nayak. 2010. "Synthesis and characterization of chitosan‐polycaprolactone blended with organoclay for control release of doxycycline." Journal of Applied Polymer Science 118 (6):3167-3175.

Sando, Lillian, Misook Kim, Michelle L Colgrave, John AM Ramshaw, Jerome A Werkmeister, and Christopher M Elvin. 2010. "Photochemical crosslinking of soluble wool keratins produces a mechanically stable biomaterial that supports cell adhesion and proliferation." Journal of Biomedical Materials Research Part A 95 (3):901-911.

Sarode, Ashish L, Harpreet Sandhu, Navnit Shah, Waseem Malick, and Hossein Zia. 2013. "Hot melt extrusion (HME) for amorphous solid dispersions: predictive tools for processing and impact of drug–polymer interactions on supersaturation." European Journal of Pharmaceutical Sciences 48 (3):371-384.

Scaffaro, R., L. Botta, M. Sanfilippo, G. Gallo, G. Palazzolo, and A. M. Puglia. 2013. "Combining in the melt physical and biological properties of poly(caprolactone) and chlorhexidine to obtain antimicrobial surgical monofilaments." Applied

145

Microbiology and Biotechnology 97 (1):99-109. doi: 10.1007/s00253-012-4283-x.

Serrano, C., L. Garcia-Fernandez, J. P. Fernandez-Blazquez, M. Barbeck, S. Ghanaati, R. Unger, J. Kirkpatrick, E. Arzt, L. Funk, P. Turon, and A. del Campo. 2015. "Nanostructured medical sutures with antibacterial properties." Biomaterials 52:291- 300. doi: 10.1016/j.biomaterials.2015.02.039.

Shah, S., S. Maddineni, J. Lu, and M. A. Repka. 2013. "Melt extrusion with poorly soluble drugs." International Journal of Pharmaceutics 453 (1):233-52. doi: 10.1016/j.ijpharm.2012.11.001.

Shen, Y., F. Lu, J. Hou, Y. Shen, and S. Guo. 2013. "Incorporation of paclitaxel solid dispersions with poloxamer188 or polyethylene glycol to tune drug release from poly(-caprolactone) films." Drug Development and Industrial Pharmacy 39 (8):1187-1196. doi: 10.3109/03639045.2012.704042.

Song, C. X., V. Labhasetwar, and R. J. Levy. 1997. "Controlled release of U-86983 from double-layer biodegradable matrices: Effect of additives on release mechanism and kinetics." Journal of Controlled Release 45 (2):177-192. doi: Doi 10.1016/S0168- 3659(96)01551-9.

Stanković, Milica, Hans de Waard, Rob Steendam, Christine Hiemstra, Johan Zuidema, Henderik W Frijlink, and Wouter L Hinrichs. 2013. "Low temperature extruded implants based on novel hydrophilic multiblock copolymer for long-term protein delivery." European Journal of Pharmaceutical Sciences 49 (4):578-587.

Stanković, Milica, Henderik W Frijlink, and Wouter L Hinrichs. 2015. "Polymeric formulations for drug release prepared by hot melt extrusion: application and characterization." Drug Discovery Today 20 (7):812-823.

Stanković, Milica, Jasmine Tomar, Christine Hiemstra, Rob Steendam, Henderik W Frijlink, and Wouter L Hinrichs. 2014. "Tailored protein release from biodegradable poly (ε- caprolactone-PEG)-b-poly (ε-caprolactone) multiblock-copolymer implants." European Journal of Pharmaceuticas Biopharmaceutics 87 (2):329-337.

Staton, C. A., M. W. Reed, and N. J. Brown. 2009. "A critical analysis of current in vitro and in vivo angiogenesis assays." International Journal of Experimental Pathology 90 (3):195-221. doi: 10.1111/j.1365-2613.2008.00633.x.

Steele, Derek. 1971. Theory of Vibrational Spectroscopy. Vol. 8: Saunders Limited.

Tajirian, A. L., and D. J. Goldberg. 2010. "A review of sutures and other skin closure materials." Journal of Cosmetic and Laser Therapy 12 (6):296-302. doi: 10.3109/14764172.2010.538413.

Tan, H. B., F. Y. Wang, W. Ding, Y. Zhang, J. Ding, D. X. Cai, K. F. Yu, J. Yang, L. Yang, and

146

Y. Q. Xu. 2015. "Fabrication and evaluation of porous keratin/chitosan (KCS) scaffolds for effectively accelerating wound healing." Biomedical and Environmental Sciences 28 (3):178-189. doi: 10.3967/bes2015.024.

Tanase, C. E., and I. Spiridon. 2014. "PLA/chitosan/keratin composites for biomedical applications." Materials Science and Engineering C 40:242-247. doi: 10.1016/j.msec.2014.03.054.

Thompson, Zanshe S, Nava P Rijal, David Jarvis, Angela Edwards, and Narayan Bhattarai. 2016. "Synthesis of keratin-based nanofiber for biomedical engineering." Journal of Visualized Experiments (108):e53381.

Tran, C. D., and T. M. Mututuvari. 2015. "Cellulose, chitosan, and keratin composite materials. Controlled drug release." Langmuir 31 (4):1516-26. doi: 10.1021/la5034367.

Tran, Chieu D, Simon Duri, and April L Harkins. 2013. "Recyclable synthesis, characterization, and antimicrobial activity of chitosan‐based polysaccharide composite materials." Journal of Biomedical Materials Research Part A 101 (8):2248-2257.

Trott, Alexander. 2012. "Chapter 8—Instruments, Suture Materials, and Closure Choices." In Wounds and Lacerations 4th Edition, 82-94.

Tummalapalli, M., S. Anjum, S. Kumari, and B. Gupta. 2016. "Antimicrobial surgical sutures: Recent developments and strategies." Polymer Reviews 56 (4):607-630. doi: 10.1080/15583724.2015.1119163.

Tyler, Betty, David Gullotti, Antonella Mangraviti, Tadanobu Utsuki, and Henry Brem. 2016. "Polylactic acid (PLA) controlled delivery carriers for biomedical applications." Advanced Drug Delivery Reviews 107:163-175.

Umair, Malik Muhammad, Zhiming Jiang, Waseem Safdar, Zhiwei Xie, and Xuehong Ren. 2015. "N‐halamine‐modified polyglycolide (PGA) multifilament as a potential bactericidal surgical suture: In vitro study." Journal of Applied Polymer Science 132 (35).

Umranikar, Salil A, Sarvpreet S Ubee, Masilamani Selvan, and Peter Cooke. 2017. "Barbed suture tissue closure device in urological surgery–a comprehensive review." Journal of Clinical Urology 10 (5):476-484.

Van Meerloo, Johan, Gertjan JL Kaspers, and Jacqueline Cloos. 2011. "Cell sensitivity assays: The MTT assay." In Cancer Cell Culture, 237-245. Springer.

Vasconcelos, Andreia, and Artur Cavaco-Paulo. 2013. "The use of keratin in biomedical applications." Current drug targets 14 (5):612-619.

147

Vasconcelos, T., B. Sarmento, and P. Costa. 2007. "Solid dispersions as strategy to improve oral bioavailability of poor water soluble drugs." Drug Discovery Today 12 (23- 24):1068-1075. doi: 10.1016/j.drudis.2007.09.005.

Viju, S., and G. Thilagavathi. 2013a. "Characterization of tetracycline hydrochloride drug incorporated silk sutures." Journal of the Textile Institute 104 (3):289-294. doi: 10.1080/00405000.2012.720758.

Viju, S., and G. Thilagavathi. 2013b. "Effect of chitosan coating on the characteristics of silk- braided sutures." Journal of Industrial Textiles 42 (3):256-268. doi: 10.1177/1528083711435713.

Visco, Annamaria, Cristina Scolaro, Alberto Giamporcaro, Salvatore De Caro, Elisabetta Tranquillo, and Michelina Catauro. 2019. "Threads made with blended biopolymers: Mechanical, physical and biological features." Polymers 11 (5):901.

Vithani, K, Y Cuppok, S Mostafa, IJ Slipper, MJ Snowden, and D Douroumis. 2014. "Diclofenac sodium sustained release hot melt extruded lipid matrices." Pharmaceutical development and technology 19 (5):531-538.

Wang, L., D. Chen, and J. Sun. 2009. "Layer-by-layer deposition of polymeric microgel films on surgical sutures for loading and release of ibuprofen." Langmuir 25 (14):7990- 7994. doi: 10.1021/la9004664.

Wang, Xiang, Tong Wu, Wei Wang, Chen Huang, and Xiangyu Jin. 2016. "Regenerated collagen fibers with grooved surface texture: Physicochemical characterization and cytocompatibility." Materials Science Engineering: C 58:750-756.

Weldon, C. B., J. H. Tsui, S. A. Shankarappa, V. T. Nguyen, M. Ma, D. G. Anderson, and D. S. Kohane. 2012. "Electrospun drug-eluting sutures for local anesthesia." Journal of Controlled Release 161 (3):903-909. doi: 10.1016/j.jconrel.2012.05.021.

Wolf, Bettina. 2010. "Polysaccharide functionality through extrusion processing." Current Opinion in Colloid Interface Science 15 (1-2):50-54.

Wu, D. Q., H. C. Cui, J. Zhu, X. H. Qin, and T. Xie. 2016. "Novel amino acid based nanogel conjugated suture for antibacterial application." Journal of Materials Chemistry B 4 (15):2606-2613. doi: 10.1039/c6tb00186f.

Yu, H., Y. Jia, C. Yao, and Y. Lu. 2014. "PCL/PEG core/sheath fibers with controlled drug release rate fabricated on the basis of a novel combined technique." International Journal of Pharmaceutics 469 (1):17-22. doi: 10.1016/j.ijpharm.2014.04.045.

Zehnder, T., T. Freund, M. Demir, R. Detsch, and A. R. Boccaccini. 2016. "Fabrication of cell-loaded two-phase 3D constructs for tissue engineering." Materials (Basel) 9 (11):887-891. doi: 10.3390/ma9110887.

148

Zhang, Cong, Xia Chen, Guiting Liu, Rong Chen, and Shaoyun Guo. 2016. "Mechanism and kinetics of drug release from poly (ϵ-caprolactone) based extrudates prepared by hot- melt extrusion." Journal of Macromolecular Science, Part B 55 (3):285-298.

149