Bioprinting is a technology, which is gaining increasing attention within the field of tissue engineering. One advantage of bioprinting is the wide variety of materials suitable for bioprinting, namely biodegradable polymers. Bioprinting is capable of producing composite scaffolds, composed of both synthetic and natural polymers, to combine the benefits of each polymer type and alleviate the respective limitations.
Additive manufacturing (AM) technologies, such as FDM, and electrospinning have been used extensively in the production of scaffolds for tissue engineering purposes; however, these methods incur limitations. Electrospinning, for example, often requires the use of highly volatile solvents during processing, while FDM has been known to produce highly rigid scaffolds which are unsuitable for tissue engineering purposes . Bioprinting technologies overcome these limitations with the ability to rapidly produce scaffolds of various geometries and sizes without the need of volatile solvents. Furthermore, these scaffolds are durable and have suitable mechanical properties to provide support to the wound, while also exhibiting a degree of flexibility. Bioprinting also offers the opportunity to print bioinks containing cells which often display high viability [5, 9], notably for natural biodegradable polymers, which require low processing temperatures.
PCL has been used in a number of studies for tissue engineering purposes where it has displayed excellent mechanical properties, alongside desirable biocompatibility . Given this, it was selected for this study to fabricate the scaffold, providing the structural framework of the scaffold.
Many factors contribute to the impaired wound-healing response, such as the presence of diabetic neuropathy and peripheral arterial disease as manifestations of diabetes [42, 43]. Infection, particularly biofilm formation witnessed in the later stages of DFU progression is another major contributing factor to poor wound-healing outcomes in diabetic patients. For mild infection, antibiotics are often administered orally, which limit the bioavailability due to the first-pass metabolism, as well as requiring higher doses, which can risk inducing toxicity. While intravenous administration of antibiotics, often used for more severe infection, bypasses the first-pass metabolism experienced during oral administration, this route is invasive, requires hospital treatment and chronic administration via this route is not feasible. Therefore, topical administration of antibiotics can be used as an alternative method to provide sustained, localised drug delivery to the target site, with low potential for systemic side effects and ease of administration being additional benefits of this route.
LFX, a fluoroquinolone antibiotic, was selected for incorporation into the scaffold. Levofloxacin has been successful in many wound-healing applications [16, 20]. Previous studies have used levofloxacin concentrations at a range of 0.5–1.5% for tissue engineering applications, with similar values observed for topical administration to the eye [44, 45]. LFX has been used extensively in the treatment of ocular infection. Studies investigating the efficacy of topical LFX applied to the cornea found 3.0% LFX to cause ocular damage and therefore be unsafe for use in terms of ocular drug delivery . Despite this, the bactericidal ability of LFX up to concentrations of 3% was investigated. The main role of the skin is to provide protection given its multilayer nature; thus, it is less sensitive to higher drug concentrations that the eye, which is more sensitive to toxicity.
Lower concentrations of 0.5–1.5% have been widely investigated and proven beneficial in the topical applications of LFX in eradicating commonly exposed bacteria such as S. aureus and E. coli [19, 44, 45]; therefore, in this study, higher concentrations were also used in order to investigate any potential benefits.
During this proof of concept study, scaffolds with different designs, loaded with antibiotic were fabricated using bioprinting technologies. The temperature for printing was increased for LFX-loaded scaffolds to accommodate the drug loading. To fabricate the scaffolds, PCL loaded with LFX was successfully printed using optimised parameters for extrusion-based bioprinting before being characterised. Due to the high viscosity of PCL, slower printing speed was required to maintain shape fidelity of the scaffold. Following analysis of drug distribution confirmed that even drug distribution was achieved.
Both control and LFX-loaded scaffold were observed using SEM and presented smooth surfaces. Importantly, the LFX-loaded scaffolds did not show any signs of drug aggregation on the surface of the scaffold.
During ATR-FTIR analysis, no peak shift was observed for the lowest concentration used in this study (0.5%), suggesting this concentration was too low to acquire any conclusive ATR-FTIR result. Thermal analysis using TGA was subsequently performed to assess any interactions occurred between PCL and LFX. It was found that PCL displayed a higher Tonset value of 380 °C, which saw a slight decline to 362 °C for both 0.5% and 3.0% LFX-loaded samples. Despite the incorporation of LFX causing a slight decrease in the observed thermal resistance for PCL, the temperatures used during the 3D bioprinting process were lower than the degradation temperatures, concluding that 3D bioprinting is suitable for the fabrication of LFX-loaded PCL scaffolds. Similar reports for PCL and LFX combinations have been reported in literature . DSC analysis was also conducted to assess thermal properties of the components of the scaffold. A strong endothermic peak was observed at approximately 235 °C for LFX, which corresponds to its melting point and is supported by literature [22, 24]. For the drug-loaded scaffolds, this sharp peak disappeared. This could result from dilution of the drug compared to the amount of polymer used in these formulations, or as a result of drug amorphisation during the printing process. The sharp endothermic peak observed for PCL occurred at 62 °C, corresponding to the expected melting point of PCL . Both LFX and PCL showed a low Tg temperature, which is desirable for ease of processing.
During this study, it was found that modifying the geometry and design of the scaffold, resulted in a change to the mechanical properties displayed. While the scaffold should provide support to the wound, it must also exhibit some degree of flexibility in order to respond to the dynamic nature of the skin in a normal range of body movements. In this study, the square design displayed the lowest tensile stiffness and therefore was selected for in vitro drug release analysis. The triangular design was not suitable for analysis due to limitations of the texture analyser, due to reasons previously outlined, and therefore was omitted from this experiment.
In clinical practice, the scaffold could be applied to the skin using an adherent dressing that would aid the retention of the scaffold in direct contact with the skin over the desired application time. Such bioadhesives have been fabricated from a number of materials, including polyethylene glycol and chitosan to aid wound healing . Other commonly used bioadhesives for wound closure, such as cyanoacrylate-based bioadhesives, have however been shown to be unsuitable for application to areas vulnerable to torsion or moist tissues ; therefore, these factors should be considered when selecting an appropriate bioadhesive material to aid scaffold retention. Furthermore, a gauze and bandage could also be used to further secure the positioning of the scaffold and could be easily removed at time points where the scaffold was required to be replaced. These options would offer greater freedom of mobility for the patient versus total contact casts commonly employed in the treatment of DFU.
Insight into the effect of drug concentration on the mechanical properties of the scaffolds was also conducted. Given that there was no significant difference in the elastic modulus values obtained for PCL and drug-loaded PCL samples, it can be concluded that the addition of LFX had no effect on the elastic modulus of the first scaffold layer. Given this, concentrations of 0.5, 1.5, and 3.0% LFX-loaded scaffold were selected for in vitro drug release to observe the release over the range of drug concentrations. The results showed significantly lower elastic modulus results than reported in similar studies using PCL and LFX to produce electrospun scaffolds ; however, values obtained are closer to the previously reported elastic modulus values (1.14 MPa) of plantar soft tissue [33, 34].
During the in vitro drug release analysis, all the concentrations (e.g. 0.5%, 1.5%, and 3.0%), demonstrated similar release profiles, with an initial burst release phase followed by steady-state release. Despite a burst release being observed, it was lower than previous studies investigating the release of LFX from PCL electrospun scaffolds, where 50% of drug had been released by 12 h . Moreover, the 3.0% formulation also showed a higher percentage release, based on initial drug-loading quantities, at day 14 (6.4%) versus the 0.5% formulation (4.0%). Given that increasing the drug-loaded concentrations results in increased drug release can be concluded that no major drug-polymer interaction occurred. Steady-state drug release was achieved following day 4, 7, and 14 for 0.5%, 1.5%, and 3.0% LFX, respectively. The 3.0% LFX provided the longest duration of release, indicating its suitability for sustained drug delivery of the antibiotic. The low drug release values obtained for these formulations can be owed to the slow hydrolytic degradation of PCL which is required to release the drug from the polymer matrix . Furthermore, this avoids potential toxicity, which can result from high burst release. If required, the rate of drug release can be increased through changing the PCL molecular weight or through structure modification using hydrophilic moieties.