Coaxial electrospun poly(lactic acid)/silk fibroin nanofibers incorporated with nerve growth factor support the differentiation of neuronal stem cells

Abstract

Coaxial electrospinning is an explicit method for encapsulation of protein drugs, and the process could preserve the bioactivity of the molecules. In this study, coaxial electrospinning was used to fabricate Poly(Lactic Acid)/Silk Fibroin/Nerve Growth Factor (PS/N) by encapsulating nerve growth factor (NGF) along with Silk Fibroin (SF) as the core of the scaffold. Air plasma treatment was applied to PS/N scaffold to improve the surface hydrophilicity without causing any damage to the nanofibers. Surface characterization of the plasma treated PS/N scaffold (p-PS/N) was carried out by Atomic Force Microscopy, X-ray Photoelectron Spectroscopy and water contact angle test. PC12 cells cultured on both PS/N and p-PS/N scaffolds using Differentiation Medium devoid of NGF expressed neurofilament 200 protein on day 8, suggesting the differentiation potential of PC12 on both the scaffolds. By day 11, the cells cultured on p-PS/N scaffolds using the Differentiation Medium devoid of NGF showed elongated neurites with the length up to 95 μm. Our results suggested the sustained release of NGF, thus demonstrating the fact that the bioactivity of NGF was retained. The p-PS/N scaffolds were able to support the attachment and differentiation of PC12 cells, with ability to function as suitable substrates for nerve tissue engineering.

---

1 Introduction

Peripheral nerve injuries (PNI) cause a huge burden to society with over 200 thousand procedures, in the US alone, performed annually to repair PNI.¹ Economically speaking, PNI cost % 150 billion annually in the US.² After the occurrence of PNI, surgical reconnection is possible for bridging small PNI gaps, but if the nerve gap is longer, meaning those that cannot be closed without tension require an autografts. Challenges, such as donor site morbidity, insufficient donor site availability, and complex surgical procedures, exist and motivate the development of novel tissue engineering strategies for peripheral nerve tissue regeneration.

PNI result in the disruption of the natural architecture of the nerve that directs and guides the developing axons towards the targets within the normal tissue. Therefore, the key is to design a bridge that could span the lesion with adequate structural and biofunctional support to restore the gap. Fibers made from biocompatible materials are capable of guiding the regenerating axons and might function as a bridging neuronal scaffold. Additionally, the scaffold should have a level of porosity³ that allows for the inward migration and incorporation of supportive cells as well as the traversing axons. Finally, the nerve graft should have the capacity to deliver signalling factors in a controlled manner to aid in the attraction, directional growth and viability of traversing regenerating axon.⁴

Electrospun nanofibers, with morphological properties similar to neural extracellular matrix (ECM), have been applied as a nerve graft, and successes with some improvements in the functionality of the tissue have been demonstrated. Electrospun polycaprolactone/gelatin nanofibrous scaffolds with a polymer ratio of 70 : 30 was suggested to enhance the nerve cell proliferation and differentiation.⁵ In vitro studies have proven that electrospun nanofibers can support the adhesion of neural stem cells (NSCs), promote their differentiation, and stimulate neurite outgrowth.⁶ On the other hand, coaxial electrospinning has frequently been selected as a carrier for drug release applications, mainly due to the ability of the method to produce core–shell structured nanofibers.⁷ With coaxial electrospinning, two polymers can be coaxially electrospun by ejecting them through two different needles to generate core–shell structured composite nanofibers.⁷ The release kinetics of the encapsulated drugs can be controlled by tuning the core–sheath compositions, while retaining their bioactivity. Poly lactic acid (PLA) is a biodegradable thermoplastic aliphatic polyester derived from renewable resources, such as starch, and has been wildly used in biomedical applications.⁸ On the other hand, Silk Fibroin (SF) is a protein excreted by the Bombyx mori silkworms during cocoon production, and it has received significant attention as a versatile natural polymer.⁹ Nerve Growth Factors (NGFs) are essential neurotrophic factors that play an important role in the growth, maintenance and survival of target neurons, supporting peripheral nerve regeneration.¹⁰

In this study, we fabricated electrospun PLA–Silk Fibroin/NGF (PS/N) scaffolds using coaxial electrospinning; further air plasma treatment was carried out to improve the hydrophilicity of the scaffolds. The change in surface roughness, surface chemistry and hydrophilicity of the scaffolds were characterized by Atomic Force Microscopy (AFM), X-ray photoelectron spectroscopy (XPS) and water contact angle measurement, respectively. Furthermore, the influence of plasma treatment and release of NGF from the scaffolds on PC12 cells behaviours were studied to investigate the potential of these scaffolds for PNS regeneration.

2 Materials and methods

2.1 Materials

PLA was purchased from PURAC biochem BV (Netherlands). Silk fibroin was purchased from Xi'an Yuensun Biological Technology Co., Ltd, China. 1,1,1,3,3,3-Hexafluoro-2-propanol (HFP), glutaraldehyde, Dulbecco's modified eagle's medium (DMEM/F12), were purchased from Sigma, Singapore. Nerve growth factor (NGF), ChemiKine Nerve Growth Factor Sandwich ELISA kit were purchased from Millipore, Singapore. Rat PC12 cells were obtained from ATCC, USA, while fetal bovine serum (FBS), horse serum (HS) and trypsin/EDTA were purchased from GIBCO Invitrogen, USA. Cell Titer 96 AQueous One solution was purchased from Promega, Madison, WI, USA.

2.2 Scaffolds preparation

2.2.1 Optimization of coaxial electrospinning parameters. To identify the optimal electrospinning parameters for coaxial electrospinning, three groups of experiments have been performed by adjusting the parameters, namely high voltage, core flow rate and core solution concentration. The parameters and the details of the different variations in these three parameters are explained in Table 1.

Table 1 Coaxial electrospinning parameters studied for fabrication of core–shell PLA/Silk Fibroin (PLA/SF) nanofibers^a

a Shell flow rate was set as 1 mL h⁻¹ for all experiments.

---

2.2.2 Fabrication of the scaffolds. The optimal parameters were identified from the above experiments to fabricate the core–shell PLA nanofibers containing silk fibroin as the core, termed as PS throughout this manuscript. Further, 10 μg of NGF was added to 200 μL of 10% w/v SF solution, and the NGF containing PLA–Silk Fibroin scaffolds (PS/N) were obtained. Pure PLA nanofibers were also fabricated as a control.

2.2.3 Optimization of plasma treatment conditions. Since PLA served as the main component of the PS/N nanofibers, and knowing the hydrophobic properties of PLA, it was necessary to improve the surface properties of the developed scaffolds. In order to improve the hydrophilicity, air plasma treatment was initially conducted using pure PLA scaffolds by plasma cleaner (Model: PDC-001, Harrick Scientific Corporation, USA). Plasma discharge was applied to samples for various duration (60 s, 90 s, 120 s, 150 s and 180 s) with radiofrequency power set as 30 W under vacuum, and the samples appeared stable throughout the treatment period. The optimized duration for plasma treatment was chosen by evaluation of the plasma treated nanofibers by SEM and water contact angle measurement. Further, the PS/N nanofibers was plasma treated using the ‘optimized’ treatment time, and the obtained samples were labelled as p-PS/N.

2.3 Scaffold characterization

2.3.1 Scanning electron microscopy (SEM). The electrospun nanofibers were sputter-coated with gold (JEOL JFC-1200 Fine Coater, Tokyo, Japan) and visualized using a field emission scanning electron microscope (SEM; FEI-QUANTA 200F, Eindhoven, The Netherlands). The average diameter of the nanofibers was calculated from 50–60 random points chosen from the SEM images, using image analysis software (Image J, National Institute of Health, Bethesda, MD).

2.3.2 Transmission electron microscopy (TEM). The core–shell structure of the nanofibers were evaluated using transmission electronic microscopy (JEOL JEM-2010F) at 200 kV, and to do so, the fibers were collected on carbon-coated copper grids.

2.3.3 Water contact angle test. Hydrophilic/hydrophobic nature of the electrospun nanofibrous scaffolds was measured by water contact angle measurement using VCA Optima Surface Analysis System (AST products, Billerica, MA). Deionized water was used for drop formation. During the measurements, electrospun nanofibrous membranes on coverslips were positioned directly on a testing plate. Subsequently, deionized water was dropped on the scaffolds and the data were recorded. Three different positions of the same sample was measured and the experiment was performed in triplicate.

2.3.4 Atomic force microscopy (AFM). The surface morphology of the PS/N and p-PS/N nanofibers were analyzed using the Atomic Force Microscope (AFM) in tapping mode (JPK Instruments AG, Singapore) using a silicon tip (NanoWizard®II Atomic Force Microscope) under ambient conditions with scanned areas of 20 × 20 μm. The images were recorded at a scan rate between 1 Hz. The cantilever tips had a tip radius of 5–7 nm, with a resonance frequency of 315–369 kHz. The images were scanned at five predetermined areas for each material (n = 5). AFM images were processed with JPK data processing software and the surface roughness was measured for the nanofibers.

2.3.5 X-ray photoelectron spectroscopy (XPS). X-ray photoelectron spectroscopy is a surface chemical analysis technique that was elected to examine the surface chemistry of the developed scaffold. XPS spectra was obtained by irradiating the material with a beam of X-rays while simultaneously measuring the kinetic energy and number of electrons that escape from the top 1 to 10 nm of the material being analysed. The method allows for the determination of the chemical composition of the scaffold surface for all elements except hydrogen with a typical sampling thickness <10 nm. XPS analysis of plasma and non-plasma treated samples was carried using AXIS Ultra DLD (Kratos Analytical Ltd, UK).

2.4 Release of NGF from core–shell nanofibers

For evaluating the drug release behaviour, each scaffold weighing 20 mg was soaked in a 15 mL centrifuge tube with 10 mL of phosphate-buffered saline (PBS). The fibrous mats were incubated at 37 °C in a continuous horizontal shaker at a speed of 150 rpm. At predetermined time points, 1 mL of the supernatant was retrieved from the tube and an equal volume of fresh PBS was replaced. Chemikine Nerve Growth Factor, Sandwich ELISA was used to determine the quantity of NGF released at different time intervals (day 9 and day 14) following the instructions of the manufacturer. The optical density for each sample was determined by microplate reader (Varioskan Flash Multimode Reader, Thermo SCIENTIFIC), and the results were expressed by cumulative release as a function of the release time:

Cumulative amount of release (%) = M_t/M_x × 100

where M_t is the weight of protein released at time t and M_x is the total amount of protein incorporated with the nanofibers theoretically. All experiments were tested in triplicate.

2.5 PC12 cell culture

PC12 cells were purchased from ATCC (Manassas, VA) and cultured in PC12 cell Growth Medium (DMEM/F12 supplemented with 10% HS, 5% FBS, and 1% antibiotic/antimycotic solution) in a 75 cm² cell culture flask. Cells were incubated at 37 °C in a humidified atmosphere containing 5% CO₂, and the culture medium was changed once in every 2 days. The 15 mm cover slips with electrospun nanofibers were placed in 24-well plate and pressed with a stainless steel ring to ensure complete contact of the scaffolds with wells. The samples were sterilized under UV light, washed thrice with PBS, and subsequently immersed in medium before cell seeding. Cells were grown to confluency, detached by trypsin/EDTA, counted by hemocytometer, and seeded on the scaffolds at a density of 10 000 cells per well.

2.6 Cell proliferation study

The cells were cultured in Growth Medium, and the proliferation of cells on different electrospun scaffolds was conducted after 2, 4, 6, and 8 days by MTS assay (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt). The metabolically active cells are known to react with tetrazolium salt in MTS reagent to produce a soluble formazan dye and absorbance was measured at 490 nm. After desired time points, the cell–scaffold constructs were rinsed with PBS, and incubated with 20% MTS reagent in serum-free medium for 3 h. The mixture was aliquoted into 96-well plates and read using a spectrophotometric plate reader (Varioskan Flash Multimode Reader, Thermo SCIENTIFIC).

2.7 PC12 cell differentiation

For PC12 cell differentiation, cells were grown in Differentiation Media which is composed of DMEM/F12 supplemented with 1% HS, 0.5% FBS, 1% antibiotic/antimycotic solution and 50 ng mL⁻¹ NGF. For certain experiments, cells were also cultured using the same above media devoid of NGF and is referred as Differentiation Medium^-NGF.

2.7.1 Cell morphology by SEM observation. To evaluate the morphological changes of in vitro differentiated PC12 cells on different scaffolds, SEM analysis was performed after 8 days of cell culture. The cell–scaffold constructs were rinsed with PBS, fixed in 3% glutaraldehyde for 3 h, rinsed with DI water, followed by washings with 50–90% of ethanol. After final washing with 100% ethanol, the cell–scaffold constructs were soaked in hexamethyldisilazane, air-dried and observed under SEM.

2.7.2 NF200 protein expression. To evaluation the differentiation of PC12 cells, the cells were grown in Differentiation Medium on TCP (tissue culture plate), P (PLA), PS (PLA/Silk Fibroin), PS/N (PLA/Silk Fibroin/NGF) and p-PS/N (plasma treated PLA/Silk Fibroin/NGF). As a comparison, PC12 cells were also grown in Differentiation Medium devoid of NGF on PS/N and p-PS/N scaffolds. After 8 and 11 days of cell culture on the different scaffolds, the immunostaining of neuronal specific protein, namely neurofilament 200 (NF200) was carried out to observe the cell phenotype and neurite extension of differentiated PC12 cells on the electrospun nanofibers. For this, the cell–scaffold constructs were rinsed with PBS, fixed in formalin for 15 min and permeabilized with 0.1% Triton-X100 for 3 min. The nonspecific binding was blocked by incubating with 3% BSA for 90 min. Subsequently the samples were stained with primary antibody, anti-NF200 produced in rabbit at a dilution of 1 : 100 (Sigma) for 2 h at room temperature. Secondary antibody staining was carried out using FITC conjugated goat anti rabbit (Sigma), and nuclei was stained with DAPI. The immunostained samples were mounted onto a glass slide and visualized under laser scanning confocal microscope (Zeiss LSM700). For negative controls of FITC alone without primary antibody incubations was performed and no specific staining was detected.

3 Results

During this study, coaxial electrospinning was utilized to fabricate p-PS/N scaffold and the process of fabrication is schematically demonstrated in Fig. 1.

Fig. 1 Schematic illustration of the electrospinning and air plasma treatment towards the formation of p-PS/N scaffolds.

3.1 Optimization of coaxial electrospinning process and period of air plasma treatment

3.1.1 Optimization of coaxial electrospinning. Electrospinning of 5% (w/v) PLA solution could produce smooth nanofibers at a flow rate of 1.0 mL h⁻¹ and under high voltage in the range of 13 to 15 kV. In this study, coaxial electrospinning of PS nanofibers was optimized by studying three different parameters (i) high voltage (ii) core flow rate and (iii) core solution concentration.

With increase in the high voltage from 11 kV to 13 kV and further to 15 kV, PS nanofibers with decreasing diameters of 268 ± 53 nm, 222 ± 44 nm and 197 ± 54 nm were respectively produced (Fig. 2A–C). The highest voltage (15 kV) utilized during this study produced the thinnest fibers, but with worsened morphology (Fig. 2C). Changing the flow rate from 0.2 mL h⁻¹ to 0.3 mL h⁻¹ or even 0.4 mL h⁻¹, did not cause any significant change in fiber diameter (Fig. 2B, D and E). It was observed that the core solution accumulated at the tip of the needle with an excess supply of core solution, and even the accumulated core solution needed manual clearance, with copious wastage of the core solution. High flow rate of the core solution was thus not recommended, and instead the core flow rate was fixed to 0.2 mL h⁻¹. For electrospinning of PLA/SF fibers, two different concentrations (10% and 30% w/v) of the core solution (aqueous silk fibroin) was used, and nanofibers with similar diameters were obtained. However, the nanofibers obtained using 30% SF concentration was found less uniform (Fig. 2F), while the electrospinnability of PS system worsened due to the higher concentration of SF solution within the core of the fibers. The reason for such effect is partially due to the poor electrospinnability of pure aqueous solution of silk fibroin solution itself.¹¹

Fig. 2 SEM images of nanofibers as well as the distribution of their diameters for the co-axial electrospun nanofibers using different parameters. (A) 11 kV, 0.2 mL h⁻¹, 10% SF; (B) 13 kV, 0.2 mL h⁻¹, 10% SF; (C) 15 kV, 0.2 mL h⁻¹, 10% SF; (D) 13 kV, 0.3 mL h⁻¹, 10% SF; (E) 13 kV, 0.4 mL h⁻¹, 10% SF; (F) 13 kV, 0.2 mL h⁻¹, 30% SF; the scale bar is 10 μm.

After optimization, the PS fibers were obtained using the following parameters and fibers with diameter in the range of 222 ± 44 nm was obtained: high voltage: 13 kV; shell solution: 5% wt. PLA with a flow rate at 1 mL h⁻¹; core solution: 10% wt. SF aqueous solution with a flow rate at 0.2 mL h⁻¹.

For the electrospinning of PS/N, the core SF solution was incorporated with an additional 50 μg mL⁻¹ of NGF.

3.1.2 Optimization of plasma treatment. The hydrophilicity of the scaffold is important since it can influence the initial cell adhesion and migration.^12,13 Plasma treatment is a convenient and cost-effective method for altering the surface properties of polymeric materials by introducing desired functionalities, such as surface adhesive property and permeability, of the scaffolds while it also improves the hydrophilicity of the scaffolds.^14,15 Plasma treatment was carried out for pure PLA scaffold, and this was performed for varying time periods (60 s, 90 s, 120 s, 150 s and 180 s). Following the plasma treatment the scaffolds were evaluated by SEM and water contact angle was measured (Fig. 3). With longer treatment time, the water contact angle was found to decrease; and for the treatment time longer than 120 s, the scaffolds became highly hydrophilic, with a water contact angle of 0°. From the SEM images, we could see that a longer treatment time (180 s) caused distortions and breakage in fiber morphology, such that it caused the damage of the fibers (indicated by arrows in Fig. 3). Therefore the optimized time for plasma treatment was chosen as 120 s and thus the PS/N scaffold was also treated with air plasma for 120 s, and the scaffold was labelled as p-PS/N.

Fig. 3 Fiber morphology of plasma treated PLA samples after a treatment time of: (A) 0 s; (B) 60 s; (C) 90 s; (D) 120 s; (E) 150 s and (F) 180 s; (G) water contact angle of the nanofibrous scaffolds at different time scale of plasma treatment; the scale bar is 10 μm; the arrows in (F) show the fiber breakage.

3.2 Influence of air plasma treatment on PS/N nanofibers

PS/N and p-PS/N nanofibrous scaffolds were obtained by the methods we described in the earlier sections, while the morphology and the core–shell structure were characterized by SEM and TEM, respectively (Fig. 4). The influence of air plasma treatment on PS/N nanofibers was evaluated for: fiber morphology (SEM), hydrophilicity (water contact angle), surface roughness (AFM), availability of surface functional groups (XPS), and NGF release. The fiber diameters did not change significantly by plasma treatment, and the diameter of PS/N and p-PS/N were 221 ± 49 nm and 228 ± 39 nm, respectively. The water contact angle changed from 133.60° to 0°, which shows that the plasma treatment for an appropriate amount of time could improve the hydrophilicity of the scaffolds significantly (p ≤ 0.05), by introducing functional groups on the surfaces without damaging the fibers.

Fig. 4 Fiber morphologies, surface roughness and hydrophilicity of PS/N and p-PS/N scaffolds. SEM images of (A) PS/N and (B) p-PS/N scaffolds. The in situ picture shown within (A) is the TEM image which shows the core–shell structure of the nanofibers. The AFM images of (C) PS/N and (D) p-PS/N scaffolds; and the water contact angle of (E) PS/N and (F) p-PS/N scaffolds; the scale bar for (A) and (B) is 10 μm.

AFM images revealed the three dimensional surface morphology of the nanofibrous scaffolds. The average roughness (Ra) for PS/N and p-PS/N were respectively obtained as 217.97 ± 38.53 and 171.63 ± 32.50 nm. It was therefore obvious that the surface of nanofibers became smoother after air plasma treatment.

Results of the XPS studies (Table 2) showed higher % of oxygen and nitrogen on the surface of p-PS/N scaffolds compared to the amounts of oxygen and nitrogen on the surface of PS/N scaffolds. The subtle nitrogen on the PS/N surface might have come from silk fibroin, which was encapsulated in the core part of the nanofibers. But as coaxial electrospinning is a dynamic process to some extent, the core materials might also appear on the surface of the scaffold accidently, but in very low amounts. More nitrogen was however observed on the surface of p-PS/N scaffolds, which demonstrated that the air plasma treatment introduced surface functional groups, such as amino groups and carboxylic groups, on the surface of p-PS/N nanofibers.

Table 2 Atomic ratios (%) of plasma and non-plasma treated scaffolds determined by XPS

Atomic elements	Scaffold
	PS/N	p-PS/N
Oxygen	35.96	36.85
Nitrogen	0.07	0.24
Carbon	63.97	62.91

---

NGF induces neuronal differentiation and survival processes and has high potential for nerve regeneration. However, NGF released from nanofibers produced by conventional electrospinning has limited effectiveness due to the burst release¹⁶ and the poor bioactivity⁷ after release. In our study, coaxial electrospinning, which could, by introducing the sheath barrier, alleviate the burst release and preserve the bioactivity of the encapsulated growth factor,¹⁷ and we used this procedure to encapsulate NGF inside the SF core. And the release results showed that, the release of NGF from PS/N and p-PS/N were respectively obtained as 0.13% and 3.84%, after a period of 9 days; while after a period of 14 days, the release from the PS/N and p-PS/N scaffolds were found similar as 6.74% and 6.90%, respectively.

3.3 PC12 cells proliferation

The proliferation of cells gradually increased on TCP and all the electrospun nanofibers from day 2 to day 8 (Fig. 5). After growing the cells for a period of 6 days, the cells were able to contact each other, and facilitated the growth of the neighboring cells. NGF is a factor that is supportive towards PC12 cells differentiation, and PC12 stops dividing and proliferating once they are treated with certain concentration of NGF.¹⁸ NGF facilitated axonal growth is a dose-dependent response.¹⁹ The literature shows that a dose of ‘50 ng mL⁻¹’ or more^20–22 can induce differentiation. Media was changed every two days for all the scaffolds during the proliferation test. From day 6 to day 8, very little amount of NGF was released from PS/N scaffold, which was also evident from the NGF release data on day 9 (0.13%), while relatively more NGF was released from p-PS/N (3.84%) during the same period. As we mentioned, little amounts of NGF could not be sufficient for the differentiation of PC12 cells, while higher amounts of NGF (3.84%) could trigger the neuronal differentiation of PC12 cells. The NGF released from p-PS/N scaffolds during the initial 4 days might not be sufficient to trigger the differentiation of PC12 cells. However the proliferation of more PC12 cells on p-PS/N ceased and the NGF influenced the cells to differentiate rather than proliferate after 4 days. This might be the reason for the presence of high cell numbers on the PS/N scaffold than on p-PS/N on day 8. The released NGF might be a barrier for PC12 proliferation on p-PS/N, such that the plasma treated surface compensated the proliferation of the cells due to improved hydrophilicity and enhanced surface property. Due to these reasons, the cell number on p-PS/N was no better than the cell number on PS/N scaffolds. However, the PS/N and p-PS/N scaffolds showed good biocompatibility to PC12 cells, and further we evaluated the differentiation potential of PC12 cells on these scaffolds, using the Differentiation Media devoid of NGF.

Fig. 5 PC12 cell proliferation by MTS assay.

3.4 PC12 cells differentiation

3.4.1 Morphology of differentiated PC12 cells. The morphology of the differentiated PC12 cells grown on nanofibrous scaffolds were observed using SEM (Fig. 6). After being treated with NGF, the PC12 cells grown on P, PS, PS/N and p-PS/N scaffolds showed a phenotype with elongated morphologies, which confirmed the differentiation of PC12 cells on all the nanofibrous scaffolds.

Fig. 6 Morphology of the differentiated PC12 cells on (A) P; (B) PS; (C) PS/N and (D) p-PS/N on day 8; the scale bar is 10 μm.

3.4.2 NF200 expression. The neurofilaments are the major groups of intermediate filaments and are found predominantly in cells or tissues of neuronal origin. Neurofilament proteins are synthesized in the neuronal perikarya, assembled to form filaments and then slowly transported within the axons towards the synaptic terminals. During this study we evaluated the expression of the neuronal protein (NF200) by cells cultured on different scaffolds. The differentiation of PC12 cells on the scaffolds elucidated by immunocytochemistry experiments of NF200 also enabled the comparison of the cell phenotype and neurite outgrowth on different scaffolds. Fig. 7 shows the LSCM micrographs of PC12 cells cultured on TCP and various scaffolds, grown using the Differentiation Medium on day 8. For comparison purpose, PS/N and p-PS/N were also cultured in Differentiation Medium devoid of NGF. It can be seen from Fig. 7 that the cells cultured on all the scaffolds expressed NF 200 protein, but the PS, PS/N and p-PS/N scaffolds showed more supportive for NF200 expression in the Differentiation Medium compared to cell differentiations observed on TCP and pure PLA. The NF200 expression on PS/N (Fig. 7F) and p-PS/N (Fig. 7G) in Differentiation Medium devoid of NGF was comparable to that of cells on PS/N (Fig. 7D) and p-PS/N (Fig. 7E) grown with the Differentiation Medium, highlighting the fact that the released NGF was sufficient and active enough to support the neuronal differentiation. And it was worth noticing that only cells grown on p-PS/N scaffolds showed elongated neurites with a length of up to 95 μm on day 11 (Fig. 8D). Our results, suggest that the p-PS/N nanofibrous scaffold, produced by coaxial electrospinning followed by air plasma treatment, are biocompatible to PC12 cells, and furthermore supported the neuronal differentiation of PC12 cells even in the absence of NGF within the media.

Fig. 7 NF200 expression of PC12 cells cultured in Differentiation Medium on (A) TCP; (B) P; (C) PS; (D) PS/N; (E) p-PS/N and cultured in Differentiation Medium devoid of NGF on (F) PS/N; (G) p-PS/N, on day 8; the scale bar is 100 μm.

Fig. 8 NF200 expression of the PC12 cells grown in Differentiation Medium on (A) PS/N; (B) p-PS/N and grown in Differentiation Medium devoid of NGF on (C) PS/N and (D) p-PS/N, on day 11; the scale bar is 100 μm.

4 Discussion

Various biological responses of cells, such as the adhesion, morphology, secretion of bioactive molecules, migration, proliferation and differentiation,²³ are influenced by the cell–matrix interactions when the cells are cultured on synthetic ECMs or scaffolds.²⁴ The fate of the cells can be manipulated by deliberately controlling the interaction between the cells and their microenvironment, i.e., by controlling the surface property and regulating the release of the encapsulated proteins. The aim of the current study was to fabricate electrospun nanofibers with good biocompatibility and sustained drug release to support nerve tissue regeneration. We hypothesized that the electrospun nanofibers loaded with NGF prepared by coaxial electrospinning could introduce the neuronal differentiation of neuronal stem cells (NSCs). The particular advantage of coaxial electrospinning is that the shell can protect the proteins within the core, avoiding the initial burst release, and this has been proved intensively by other researchers using bovine serum albumin^7,25 and other growth factors, such as platelet-derived growth factor (PDGF).²⁶ At the same time, plasma treatment was able to improve the surface hydrophilicity of the scaffolds, enabling the scaffold surface to capture more proteins or nutrients from the medium.²⁷

4.1 Formation of functionalized nanofibrous scaffold

The coaxial electrospinning process was optimized using PLA as the shell and SF as the core solution. Core–shell structured PS nanofibers were obtained by adjusting the high voltage to 13 kV, core flow rate to 0.2 mL h⁻¹ and using a core solution concentration of 10% silk fibroin (wt/v). Smooth PS nanofibers with the diameter in the range of 222 ± 44 nm was thus obtained. The following three aspects were found crucial during the coaxial spinning process: (1) the high voltage had a great impact on the fiber diameter; (2) core flow rate might not affect the morphology of the obtained fiber, but inadequate core flow rate lead to the wastage of the core solution; (3) the concentration of the core solution while using solutions with poor electrospinnability as the core should be carefully considered, mainly because an excessive concentration of the core solutions would worsen the electrospinning process, and furthermore affect the fiber morphology. These parameters were designed carefully to obtain core–shell structured nanofibers, while the parameters of the shell solution, such as the concentration and flow rate, were pre-determined by the electrospinning of the pure PLA solution itself.

The surface of the PS/N nanofibers remained highly hydrophobic with a water contact angle of 133.60°, and it lacked functional groups for cell adhesion, and this might hinder the neuronal differentiation process. In order to improve the hydrophilicity of the PS/N scaffold, air plasma treatment was carried out, after which the water contact angle of the scaffold reduced to 0°. The air plasma treatment for 120 s did not bring huge damage or differences in the fiber diameter. Surface modification via air plasma was able to induce hydroxyl (–OH), carboxyl (–COOH), carbonyl (C=O) and amino (–NH₂) functional groups,²⁸ which has been shown to affect the binding of cell-adhesive proteins, such as fibronectin and vitronectin on the scaffold surfaces, through changing the surface energy and charges.²⁹ It has also been proved that plasma treatment could enhance the neuronal differentiation of Mesenchymal Stem Cells.³⁰ In our study, air plasma treatment improved the hydrophilicity of the electrospun nanofibrous scaffolds mainly by introducing amino groups, while the average roughness factor of the scaffold demonstrated smoother surfaces by AFM analysis. Such attributes might influence the behaviour of PC12 cells on the scaffolds.

4.2 Cell–scaffold interaction

PC12 cell line, derived from the rat pheochromocytoma, was used to test the potential application of PS/N and p-PS/N scaffolds for nerve tissue applications.³¹ PC12 cells have been widely used to study the neurotrophic factor-induced signaling pathways that control differentiation, and as in vitro models to detect the effect of chemicals on neurite outgrowth. An important feature of PC12 cells is that they respond to NGF with a dramatic change in phenotype and acquire a number of properties characteristic of sympathetic neurons.³² Upon exposure to NGF, PC12 cells cease to proliferate, extend multiple neurites, and acquire the properties of sympathetic neurons.¹⁸ This response makes PC12 cells a good ‘biological sensor’ of NGF.³³ During this study, the biofunctionality of the polymeric nanofibrous scaffolds was achieved in three aspects: (i) ECM mimicking scaffolds fabricated by electrospinning could provide the structural support for attachment of PC12 cells (ii) improvement of surface property achieved by air plasma treatment helped the initial adhesion and spreading of PC12 cells on the scaffolds (iii) NGF could retain its bioactivity and function after being released from the p-PS/N scaffolds in a sustained manner, which was crucial for the neuronal differentiation of the PC12 cells.

Cultured on P, PS, PS/N and p-PS/N scaffolds in Differentiation Medium, the PC12 cells assumed polygonal appearance and connected with the adjacent cells (Fig. 6), suggesting their preparedness and initiation to differentiate or they even appear to have started their differentiation process. On day 8, there was no big difference between NF 200 protein expressions on PS/N for cells grown using the Differentiation Medium and Differentiation Medium devoid of NGF, realizing the fact that the activity of NGF was well retained and the release was adequate for triggering the differentiation of PC12 cells. Similar results were also gained for p-PS/N scaffolds. The advantage of p-PS/N scaffolds over PS/N scaffolds is to support the differentiation of PC12 cells, which has been proved by the results of our NF200 expression and neurite outgrowth on day 11. Under the conditions of cell density and NGF exposure used in this study, we observed a definite neurite outgrowth for cells seeded on p-PS/N scaffolds by day 11. The length of the neurite was 34.40 ± 20.04 μm, with a longest neurite of 95 μm (Fig. 8D). Dinis et al. cultured PC12 cells on NGF-functionalized fibroin nanofibers without any NGF in the culture medium, and they found the cells presented neurite-like structures confined to the NGF-functionalized fibers, and the longest neurite was around 13 μm.³⁴ Lee et al. prepared NGF–polypyrrole–PLGA nanofibers and found median neurite length of 14.7 μm for PC12 cells cultured on these scaffolds.³³ Valmikinathan et al. found that the NGF released from PCL–BSA nanofibers could significantly stimulate the increase in neurite lengths, and the average neurite length for PC12 cells cultured after 14 days were around 20 μm.²² In the above mentioned studies, NGF was introduced in the nanofibers by either immobilization or a blending procedure, and the resultant biofunctional nanofibers introduced the PC12 cells differentiation and neurite outgrowth. In our study, the release rate of NGF in the SF core was influenced by the shell degradation and the core material, namely SF used along with the NGF. The NGF released from PS/N and p-PS/N scaffolds was bioactive enough, which was proved by the release study (ELISA test) and the subsequent neuronal differentiation of PC12 cells observed from the enhanced neurite outgrowth. Intriguingly, the p-PS/N scaffolds enhanced the PC12 cells differentiation, which was evident from the lengthier neurites (95 μm), while no obvious neurite outgrowth were observed for the cells grown on PS/N scaffolds. Air plasma treatment produced several functional groups on the scaffold surface,²⁸ and this has been shown to affect the binding of cell-adhesive proteins, i.e.; the cell attachment improved through increased surface wettability²⁷ and thus the scaffold demonstrated improved biocompatibility.³⁵ It has also been proved that plasma treatment could enhance the neuronal differentiation.³⁰ Other than that, the p-PS/N scaffolds were able to provide a more sustained release of NGF to the PC12 cells than the PS/N scaffolds, which has been proved by both the NGF release result and the NF200 expression. NGF released from nanofibers has been proved to be able to enhance the neuronal differentiation³⁶ and increase neurite lengths.²² The neurite outgrowth is only achieved on the plasma treated scaffolds resulting from the synergistic effect of plasma treatment and bioactive NGF released from the scaffolds. Our results thus highlight the advantage of the core–shell structured nanofibers in preserving the bioactivity of nerve growth factor for accelerated nerve tissue regeneration.

The differentiation of PC12 cells on the nanofibrous scaffolds is schematically illustrated in Fig. 9. Firstly, NGF was released from the nanofibrous scaffolds and as the NGF contacted the PC12 cells, it bound to TrkA, which activated the autophosphorylation of the receptor,³⁷ and initiated the neuronal differentiation program,³⁸ as well as initiating a number of signaling cascades including the Raf/MEK/MAP kinase pathway^21,39 and the PLCγ/PKC pathway,^40–42 which inhibited the cell proliferation and supported the neurite growth. During development, neurons become assembled into functional networks by growing out axons and dendrites (collectively called neurites) that connect synaptically to other neurons. The outgrowth of neurons proceeds by the dynamic behavior of growth cones, specialized structures at the tip of growing neurites. To allow regeneration following PNI, the microenvironment must be permissive enough for neurite outgrowth. The neurite outgrowth was only observed on p-PS/N scaffolds, which is the synergistic effects of scaffold plasma treatment and controlled delivery of NGF from the p-PS/N scaffolds.

Fig. 9 Schematic illustration of the neuronal differentiation of PC12 cells on p-PS/N scaffolds.

5 Conclusions

To bridge the disruption caused by peripheral nerve injury, tissue engineered scaffolds with biomimetic architecture that allows for the controlled release of NGF are considered. During this study, PS/N scaffolds with satisfactory nanofibrous structure, surface hydrophilicity and biological cues were successfully fabricated by coaxial electrospinning technique, and this was followed by plasma treatment of the scaffolds to obtain p-PS/N nanofibers. The scaffolds showed good biocompatibility, with non-cytotoxicity to PC12 cells. On day 11, the differentiated PC12 cells on p-P/N nanofibers showed elongated neurites with a length of 95 μm. Results demonstrated that the plasma treated core–shell structured nanofibers encapsulated with NGF has great potential to be applied in curing peripheral nerve injuries.

Acknowledgements

This study is funded by the Danish Council for Strategic Research Project titled ‘Electrospun Biomimetic Nanofibres as Regenerative Medicines (ElectroMed)’ from Aarhus University, in collaboration with the Department of Mechanical Engineering at the National University of Singapore (WBS R-265-000-487-597) in Singapore.

References

1. C. E. Schmidt and J. B. Leach, Annu. Rev. Biomed. Eng., 2003, 5, 293–347.
2. C. A. Taylor, D. Braza, J. B. Rice and T. Dillingham, Am. J. P. M. R., 2008, 87, 381–385.
3. S. Nedjari, G. Schlatter and A. Hébraud, Mater. Lett., 2015, 142, 180–183.
4. W. N. Chow, D. G. Simpson, J. W. Bigbee and R. J. Colello, Neuron Glia Biol., 2007, 3, 119–126.
5. L. Ghasemi-Mobarakeh, M. P. Prabhakaran, M. Morshed, M.-H. Nasr-Esfahani and S. Ramakrishna, Biomaterials, 2008, 29, 4532–4539.
6. F. Yang, C. Y. Xu, M. Kotaki, S. Wang and S. Ramakrishna, J. Biomater. Sci., Polym. Ed., 2004, 15, 1483–1497.
7. Y. Z. Zhang, X. Wang, Y. Feng, J. Li, C. T. Lim and S. Ramakrishna, Biomacromolecules, 2006, 7, 1049–1057.
8. P. Simamora and W. Chern, J. Drugs Dermatol., 2006, 5, 436–440.
9. S. Sofia, M. B. McCarthy, G. Gronowicz and D. L. Kaplan, J. Biomed. Mater. Res., 2001, 54, 139–148.
10. J. W. Griffin, M. V. Hogan, A. B. Chhabra and D. N. Deal, J. Bone Jt. Surg., Am. Vol., 2013, 95A, 2144–2151.
11. Z. Li, L. Song, X. Huang, H. Wang, H. Shao, M. Xie, Y. Xu and Y. Zhang, RSC Adv., 2015, 5, 16748–16758.
12. H. S. Koh, T. Yong, C. K. Chan and S. Ramakrishna, Biomaterials, 2008, 29, 3574–3582.
13. L. Ghasemi-Mobarakeh, M. P. Prabhakaran, M. Morshed, M. H. Nasr-Esfahani and S. Ramakrishna, Mater. Sci. Eng., C, 2010, 30, 1129–1136.
14. M. P. Prabhakaran, J. Venugopal, C. K. Chan and S. Ramakrishna, Nanotechnology, 2008, 19, 455102.
15. G. Greene, G. Yao and R. Tannenbaum, Langmuir, 2003, 19, 5869–5874.
16. E. R. Kenawy, G. L. Bowlin, K. Mansfield, J. Layman, D. G. Simpson, E. H. Sanders and G. E. Wnek, J. Controlled Release, 2002, 81, 57–64.
17. J.-J. Liu, C.-Y. Wang, J.-G. Wang, H.-J. Ruan and C.-Y. Fan, J. Biomed. Mater. Res., Part A, 2011, 96, 13–20.
18. J. A. Harrill and W. R. Mundy, Methods Mol. Biol., 2011, 758, 331–348.
19. S. W. P. Kemp, A. A. Webb, S. Dhaliwal, S. Syed, S. K. Walsh and R. Midha, Exp. Neurol., 2011, 229, 460–470.
20. Y. Il Cho, J. S. Choi, S. Y. Jeong and H. S. Yoo, Acta Biomater., 2010, 6, 4725–4733.
21. S. Traverse, N. Gomez, H. Paterson, C. Marshall and P. Cohen, Biochem. J., 1992, 288, 351–355.
22. C. M. Valmikinathan, S. Defroda and X. Yu, Biomacromolecules, 2009, 10, 1084–1089.
23. A. S. Badami, M. R. Kreke, M. S. Thompson, J. S. Riffle and A. S. Goldstein, Biomaterials, 2006, 27, 596–606.
24. H. Park, J. W. Lee, K. E. Park, W. H. Park and K. Y. Lee, Colloids Surf., B, 2010, 77, 90–95.
25. X. Q. Li, Y. Su, R. Chen, C. L. He, H. S. Wang and X. M. Mo, J. Appl. Polym. Sci., 2009, 111, 1564–1570.
26. H. Li, C. G. Zhao, Z. X. Wang, H. Zhang, X. Y. Yuan and D. L. Kong, J. Biomater. Sci., Polym. Ed., 2010, 21, 803–819.
27. X. Jing, H.-Y. Mi, J. Peng, X.-F. Peng and L.-S. Turng, Carbohydr. Polym., 2015, 117, 941–949.
28. S. H. Keshel, S. N. K. Azhdadi, A. Asefnezhad, M. Sadraeian, M. Montazeri and E. Biazar, Int. J. Nanomed., 2011, 6, 641–647.
29. Y. Arima and H. Iwata, J. Mater. Chem., 2007, 17, 4079–4087.
30. H. Jahani, F. A. Jalilian, C. Y. Wu, S. Kaviani, M. Soleimani, N. Abassi, K. L. Ou and H. Hosseinkhani, J. Biomed. Mater. Res., Part A, 2014, 103, 1875–1881.
31. F. Fornai, P. Lenzi, G. Lazzeri, M. Ferrucci, F. Fulceri, F. S. Giorgi, A. Falleni, S. Ruggieri and A. Paparelli, Brain Res., 2007, 1129, 174–190.
32. K. P. Das, T. M. Freudenrich and W. R. Mundy, Neurotoxicol. Teratol., 2004, 26, 397–406.
33. J. Y. Lee, C. A. Bashur, C. A. Milroy, L. Forciniti, A. S. Goldstein and C. E. Schmidt, IEEE Trans. Nanobiosci., 2012, 11, 15–21.
34. T. M. Dinis, G. Vidal, R. R. Jose, P. Vigneron, D. Bresson, V. Fitzpatrick, F. Marin, D. L. Kaplan and C. Egles, PLoS One, 2014, 9, e109770.
35. C. Zandén, N. Hellström Erkenstam, T. Padel, J. Wittgenstein, J. Liu and H. G. Kuhn, Nanomedicine, 2014, 10, 949–958.
36. X. Q. Li, Y. Su, S. P. Liu, L. J. Tan, X. M. Mo and S. Ramakrishna, Colloids Surf., B, 2010, 75, 418–424.
37. G. M. Brodeur, Nat. Rev. Cancer, 2003, 3, 203–216.
38. D. R. Kaplan and R. M. Stephens, J. Neurobiol., 1994, 25, 1404–1417.
39. S. C. Kao, R. K. Jaiswal, W. Kolch and G. E. Landreth, J. Biol. Chem., 2001, 276, 18169–18177.
40. U. H. Kim, D. Fink, H. S. Kim, D. J. Park, M. L. Contreras, G. Guroff and S. G. Rhee, J. Biol. Chem., 1991, 266, 1359–1362.
41. D. M. Loeb, R. M. Stephens, T. Copeland, D. R. Kaplan and L. A. Greene, J. Biol. Chem., 1994, 269, 8901–8910.
42. M. Ohmichi, G. Zhu and A. R. Saltiel, Biochem. J., 1993, 295(Pt 3), 767–772.

---