Synthetic or natural biocompatible polymers are commonly considered as candidates to develop scaffolds for tissue engineering [1
]. Effective scaffolds for tissue engineering need to consist of materials that are highly porous, fibrous, biocompatible, biodegradable, and cause no harm to the immune system. It is also important that the complex function of the scaffold mimics the extracellular matrix (ECM), the vital model in providing structural and biochemical support to human cells.
Using compatible biomaterials such as poly(ε-caprolactone) (PCL), chitosan (CS), or gelatin (GEL) is a common approach to engineer a variety of tissue types. The biodegradable polyester PCL is among the most studied scaffolds in tissue engineering and is approved by the United States Food and Drug Administration (FDA, Silver Spring, MD, USA) [2
]. PCL has high plasticity, ductility, and ester linkages, allowing for a slow degradation rate from hydrolysis, which are all useful characterization tools for these scaffolds [3
]. Biocompatible nanofibers (NFs) produced by electrospinning have been a popular scaffold material due to their similar characteristics to the ECM. PCL NFs have been shown to form suitable interwoven porous scaffolds [4
], which assist in the connection of tissues and vessels. These NFs have also been shown to be appropriate structures to mimic the ECM, due to their ability to promote the adhesion and proliferation of cells [8
]. Additionally, PCL has previously been shown to support a wide range of range of cell types, and its biodegradable features render it an excellent candidate for carrying therapeutic molecules [5
Porous silicon (pSi) is another example of a biomaterial, and has several unique properties making it attractive for both in vivo and ex vivo applications [10
]. pSi is formed by the anodization of crystalline silicon wafers in hydrofluoric acid (HF). pSi nanostructures are easily tailored by altering the wafer resistivity, HF concentration, and current density [11
]. Furthermore, pore sizes may be tuned in diameter from a few nanometers to several microns, achieving surface areas of up to 800 m2
]. The biocompatibility of pSi has been demonstrated in several animal models [13
]. Bimbo et al. showed that the biodistribution of pSi nanoparticles (pSi NPs) does not induce toxicity or inflammatory responses while displaying excellent in vivo stability in rats [14
]. Ex vivo applications for pSi include implantable biosensors [15
], cancer diagnostics [16
], and wound healing [17
]. Previously investigated in vitro functionalized pSi NPs and immune cell interactions revealed that even at concentrations of 250 mg/mL, there were no significant cytotoxicity effects [18
]. The surface chemistry of pSi is important to load different drugs where a range of in vivo and in vitro studies have demonstrated the biocompatibility of surface-modified pSi [19
]. We have reported on composite pSi and PCL membranes implanted into the subconjunctival space of rats [20
]. These membranes did not erode or cause inflammatory responses in the tissue surrounding the implant and there was also no evidence of vascularization. Our more recent work investigated the utility of surface-modified pSi membranes as a scaffold for the transfer of oral mucosal cells to the eye. We found that pSi scaffolds supported and retained transplanted rat oral mucosal epithelial cells both in vitro and in vivo [21
]. Furthermore, pSi is not limited to biological applications whereby our previous work has also shown its adaptability in solar energy conversion. This verifies pSi as a versatile nanomaterial which can be used in a variety of applications [22
Electrospinning uses an electrical charge to draw fine fibers from a liquid polymer solution to a charged collector plate. These fibers are typically one-dimensional (1-D) porous structures, which range from submicron to several nanometers in size [23
]. Electrospun fibers and in particular NFs offer promising properties such as large surface area to volume ratio, surface flexibility, and superior tensile strength [24
]. NFs are exciting candidates for a range of applications such as drug delivery [26
], biosensing [27
], photovoltaic devices [28
], and energy storage [29
]. An imperative requirement for drug delivery systems is the generation of porous scaffolds to accommodate cells in guiding their growth and regeneration in three dimensions (3-D). This can be achieved using multiple layers of electrospun materials, which produce 3-D fibrillar NF mats [30
]. NFs that are used in drug delivery typically follow one of two designs: firstly, for NFs with homogenous structures, the drug or target molecule may be dispersed throughout the fibrous polymer matrix. Secondly, core-shell or coaxial NFs may be fabricated whereby a polymer covers the matrix carrying the drug [31
]. The drug diffusion mechanism and drug release kinetics are different for the two designs. For example, in homogenous NFs, the drug must travel progressively long distances in order to diffuse throughout the outer edge of the NFs, meaning that the rate of release typically decreases with time. In contrast, core-shell NF systems enable stable diffusion rates of the drugs due to the structure of the system; however, their preparation is much more complex. A core advantage of using NFs is their structure (diameter, density, and thickness), which can be easily tailored by varying the process parameters or combining other nanomaterials within the fibrous structures [33
]. Furthermore, fiber diameter is tunable and dependent on polymer molecular weight, sol-gel concentration, flow rate, applied voltage, needle tip size, and hydrolytic degradation of NFs [35
Herein, we combine the favorable properties of both pSi and PCL to manufacture drug eluting composite electrospun PCL fibers containing pSi NPs. We investigate the release kinetics of the small molecular drug camptothecin (CPT). In an effort to control the release kinetics, we investigate the difference of release properties of materials loaded directly with CPT and those with the equivalent CPT amount preloaded into pSi NPs. We also investigate the effect of pretreating the PCL fibers with sodium hydroxide (NaOH) to enhance their wettability and degradation rate [36
2. Materials and Methods
Hydrofluoric acid (HF) 48% (Merck), dichloromethane (CH2
, Labserv, analytical grade, 99.5%), and polycarolactone (80,000 average m wt) were purchased from Sigma-Aldrich and used as received. Methanol (Merck, analytical grade, 99.5%), DCM (Chemsupply, analytical grade, 99.5%), and ethanol (Ajax, absolute, 100%) were used without further purification. N
-dimethylformamide (DMF, EMD Chemicals, Overijse, Belgium) was purified via standard laboratory protocols including drying over MgSO4
followed by distillation at reduced pressure [36
]. Milli-Q water was obtained from an Advantage A10 water purification system provided by Merck Millipore (water resistivity of 18.2 MΩcm at 25 °C, Total Organic Carbon (TOC) < 5 ppb). Dulbecco’s phosphate buffered saline (PBS) solution and fluorescein isothiocyanate (97.5%, FITC) were purchased from Sigma Aldrich and used as received.
2.2. Fabrication of pSi NPs
pSi NPs were fabricated from p-type Si wafers (boron-doped, resistivity 0.0008–0.0012 Ω cm, <100>) supplied by Siltronix (Archamps, France). The wafer was anodized in an 18-cm2 etching cell in 3:1 HF:ethanol (v/v) solution with a square wave form comprising an initial current density of 50 mA/cm2 for 7.3 s and a second current density of 400 50 mA/cm2 for 0.4 s. This two-step cycle was repeated continuously for 1 h, generating a pSi film with alternating low and high porosity layers. The etched layer was removed from the Si substrate via electropolishing in 1:20 HF:EtOH at 4 mA/cm2 for 4 min and 10 s. Subsequently, the pSi membrane was sonicated for 16 h in DMSO to generate chemically oxidized pSi NPs. These nanoparticles were sized by passing through a 220-nm polytetrafluoroethylene (PTFE) syringe filter, followed by the collection of the pellet after centrifugation at 22,000× g. This filtration and centrifugation allowed for the removal of large and small nanoparticles and facilitated the harvest of reasonably uniformly sized NPs that permanently remained in solution.
2.3. CPT Loading of pSi NPs and PCL Composite Materials
pSi NPs were placed into an Eppendorf tube, to which a known volume (50–200 µL) of CPT solution of approximately 4.6 mg/mL CPT in dry distilled DMF was added. This mixture was allowed to incubate for 2 h before the particles were dried under vacuum (10 mm Hg) in a desiccator. The total loadings were calculated based on the mass of pSi placed into the tube originally; for example, a typical loading used 200 µL of 4.6 mg/mL CPT and 15.0 mg of pSi, resulting in a loading of 61.3 µg of CPT per mg of pSi. Each individual batch loading was used to convert the release amounts into percentages.
To load the PCL NFs, a predetermined mass of pSi was added to the spinning solution based on the loading of the pSi NPs. The final loading of CPT in the spinning solution was normalized to contain 0.63 mg of CPT for both the PCL + CPT materials and the PCL + pSi-CPT materials. In the above example, 10.3 mg of CPT loaded pSi NPs would be added to the spinning solution. To ensure a homogeneous distribution of the pSi NPs in the spinning solution, the particles were briefly sonicated in a small amount of acetone before injection into the spinning solution. After injection, the solutions were stirred as best as possible and then placed into the syringe for spinning.
2.4. Water Contact Angle (WCA) Measurements
The WCA was measured by placing a 1-µL drop of water on the sample surface and capturing a digital image using a Panasonic Super Dynamic wv-BP550 Closed Circuit TV camera. The contact angle measurements were analyzed by Scion Image for Windows Framegrabber software (Beta version 4.0.2).
2.5. Electrospinning of PCL and NaOH Treatments
In a typical synthesis, a 5-mL solution of 10% PCL in acetone with pSi NPs was electrospun from a 23 G stainless steel needle. The mass of pSi added to the electrospinning solution was dependent on the loading of the particular batch of pSi; however, for all of this work the mass of pSi used was balanced to facilitate a loading equivalent to 0.63 mg of CPT in the pSi + CPT composite material. Hence, every material spun and tested for drug release contained the same initial loading of CPT. The needle was then connected to a high-voltage supply (Gamma High Voltage Research, Ormand Beach, FL, USA). The solution was fed at a rate of 0.5 mL/h using a syringe pump (PHD 2000, Harvard Apparatus, Holliston, MA, USA). A piece of flat aluminum foil was placed 10 cm below the tip of the needle as a collector plate. The voltage for the electrospinning was set to 15 kV and was conducted in a controlled temperature environment (25 °C).
2.6. Scanning Electron Microscopy (SEM)
SEM was performed on an FEI Quanta 450 FEG environmental SEM fitted with a Secondary Electron Detector (SED) detector and operated at 30 keV with a spot number of 2. To help facilitate the dissipation of charge build-up, samples were coated with a 5-nm thick layer of Pt prior to analysis, according to our standard laboratory protocol.
2.7. X-ray Photoelectron Spectroscopy (XPS)
XPS measurements were recorded on a Thermo Scientific K-alpha spectrometer with monochromatic Al-Kα radiation at University College London. This involved obtaining a monatomic depth profile of the NFs using an ion beam to etch layers of the surface revealing subsurface information. The etching was performed for 200 s, which was calibrated to penetrate ~50 nm into the surface. Peak positions were calibrated to carbon (285 eV) and plotted using the CasaXPS and qtiplot software.
2.8. Energy-Dispersive X-ray Spectroscopy (EDX)
EDX was obtained using an Oxford Instruments UTW Energy Dispersive Spectroscopy (EDS) detector running ISIS software. The EDS detector was ran through a JEOL JSM-6301F Field Emission SEM.
2.9. Fluorescence Microscopy
Fluorescence microscopy was performed on an Eclipse 50i microscope equipped with a D-FL universal epi-fluorescence attachment and a 100-W mercury lamp (Nikon Instruments, Tokyo, Japan). Fluorescence images were captured with a Charged-Coupled Device (CCD) camera (Nikon Instruments, Tokyo Japan), using the following fluorescence filters. Blue channel (violet excitation, blue emission): excitation: 385–400 nm (bandpass, 393 CWL), dichromatic mirror: 435–470 nm (bandpass), and barrier filter wavelength: 450–465 nm (bandpass, 458 CWL). Green channel (blue excitation, green emission): excitation: 475–490 nm (bandpass, 483 CWL), dichromatic mirror: 500–540 nm (bandpass), and barrier filter wavelength: 505–535 nm (bandpass, 520 CWL). Red channel (green excitation, orange/red emission): excitation: 545–565 nm (bandpass, 555 CWL), dichromatic mirror: 570–645 nm (bandpass), and barrier filter wavelength: 580–620 nm (bandpass, 600 CWL). Images were analyzed using NIS-elements v3.07 software (Nikon Instruments, Tokyo, Japan).
2.10. Drug Release
CPT release was monitored via fluorimetry and performed on a Perkin Elmer Instruments LS55 luminescence spectrometer with an excitation wavelength of 340 nm and emission wavelength of 434 nm. The slit width was set to 3 nm and the photomultiplier voltage was set to 775 V. The cumulative release data of CPT into 3 mL of PBS was monitored over a 17-h period. Release rates were calculated from the slope of release curves obtained. The actual amount of CPT released was calculated with reference to a calibration curve and normalized to the surface area of the sample to give the amount of CPT released per cm2
. This allowed the CPT release data to be directly compared between each of the samples. A minimum of three release curves was averaged to produce the release curves. The release of CPT was performed in PBS at pH 7.4 and pH 1.8. Despite the low solubility of CPT in aqueous solutions (14.2 ± 2.9 μM) [37
], sink conditions were maintained for all release experiments (maximum release concentrations were below 1.4 μM).
2.11. Cytotoxicity Assay
SH-SY5Y cells (human neuroblastoma cells), were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 10% FBS, 100 U/mL penicillin, and 100 µg/mL streptomycin (Invitrogen) as previously described. SH-SY5Y cells were placed in wells of a 96-well plate at 15,000 cells per well. After one day, the cultured cells were incubated with the prepared CPT-loaded and CPT-free PCL NFs (the following set of samples was used: PCl only, PCl + CPT, PCl + pSi-CPT). PCl NFs in DMEM (10% FBS, 100 U/mL penicillin, and 100 µg/mL streptomycin) were incubated for 24, 48, and 96 h at 37 °C and 5% CO2 prior to cell viability measurements. Controls were generated by incubating SH-SY5Y cells in DMEM without a PCl NF for an identical incubation period.
To determine the effect of the microparticles treatment on cell viability, the percentage of live and dead cells, lactate dehydrogenase (LDH) released in culture supernatants was measured using an established assay (Abcam LDH-Cytotoxicity Assay Kit II) according to the manufacturer’s instructions. After 24, 48, and 96 h of incubation with microparticles, 100 µL of the cell suspension was removed and centrifuged at 600× g for 10 min, and the supernatant was transferred to a 96-well plate. To each well, 100 μL LDH reaction mix (Abchem) was added. After 30 min of incubation at room temperature, the absorbance at 450 nm was measured. All experiments were repeated at least three times.