Sustained Biochemical Signaling and Contact Guidance by Electrospun Bicomponents as Promising Scaffolds for Nerve Tissue Regeneration.

Electrospun fibers are excellent delivery vehicles enabling a sustained release of growth factors to elicit favorable cell responses and are increasingly used in tissue engineering. Scaffolds with specific physical/topographical features can also guide cell migration and maturation. Therefore, growth factor-loaded electrospun scaffolds with a designed topography are promising for tissue regeneration. In this investigation, aligned-fiber scaffolds composed of poly(lactic-co-glycolic acid) nanofibers incorporating a glial cell line-derived growth factor and poly (d,l-lactic acid) nanofibers incorporating a nerve growth factor were produced by electrospinning. The scaffolds provided an aligned fibrous topography and a dual release of growth factors. The rat pheochromocytoma cell (PC12 cell) response to produced non-woven and aligned-fiber scaffolds with/without growth factors was studied. The dual release of growth factors and topographical cues provided by aligned-fiber bicomponent scaffolds induced significant neurite extension, neuronal differentiation, and neurite alignment in a synergistic manner. The scaffolds with predesigned biochemical/topographical cues demonstrated in this study might be promising for nerve tissue repair.

Introduction

Tissue engineering scaffolds are designed to recapitulate cell niche and/or provide required biochemical signals. To date, the investigations on the effects of topographical cues, biological cues, and electrochemical cues have gained increasing attention, suggesting that scaffolds with both biosignal capability and appropriate topographical features are promising in inducing the regeneration of targeted tissues.

In peripheral nerve tissue engineering, local and sustained delivery of growth factors from biodegradable scaffolds has promoted neural differentiation.[1−5] Biocompatible nerve conduits with designed topographical features have been used to provide native neuronal cells with contact guidance, promoting neuronal growth and axonal extension in different ways.[6−16] Among various fabrication techniques, electrospinning is advantageous in making extracellular matrix-like scaffolds with a designed pore size, porosity, fiber diameter, and alignment. Meanwhile, electrospun scaffolds are also efficient delivery vehicles of labile biomolecules.[17−22]

In this study, electrospun scaffolds with both aligned topographical guidance and dual release of growth factors were made, aiming at eliciting enhanced neurite outgrowth, neural differentiation, and neurite alignment. Fibrous bicomponent scaffolds with aligned fibers were fabricated through high-speed dual-source dual-power electrospinning (HS-DSDP-ES). Bicomponent scaffolds with a randomly orientated fibrous structure (non-woven scaffolds) were made as a control. The structure and morphology of scaffolds and the in vitro release of both growth factors were investigated. The rat pheochromocytoma cell line (PC12) was used for in vitro experiments. Neurite outgrowth, neural differentiation, and cell alignment on scaffolds were investigated and quantified using immunofluorescence staining.

Results

Characterization of Scaffolds

Aligned-fiber scaffolds were made through HS-DSDP-ES, whereas non-woven scaffolds were made through normal DSDP-ES. Both types of fibers were uniformly distributed in aligned-fiber scaffolds, showing a distinct fiber alignment (Figure a). It can be seen that fibers had similar average fiber diameters (around 500 nm) in all scaffolds. The fiber alignment was clearly visible in aligned-fiber scaffolds (Figure d,e), while a non-woven fibrous structure was seen in non-woven scaffolds (Figure g,h). The distribution of fiber orientations in aligned-fiber scaffolds and non-woven scaffolds is shown in Figure c,f. A narrow distribution of fiber orientations was observed in aligned-fiber scaffolds, indicating a high fiber alignment. By contrast, a wide distribution of fiber orientations was found in non-woven scaffolds, which indicated a low fiber alignment.

Figure 1

Confocal image of aligned-fiber (a) and non-woven (b) bicomponent scaffolds containing rhodamine-B-labeled nerve growth factor (NGF)/poly (d,l-lactic acid) (PDLLA) fibers. SEM image of aligned-fiber + GFs (d), aligned-fiber only (e), non-aligned fiber + GFs (g), and non-aligned fiber only (h) scaffolds. Distribution of fiber orientations in aligned-fiber (c) and non-woven (f) bicomponent scaffolds. Water contact angles (i) and GF release profile (j) of the fibrous bicomponent scaffold.

The wettability of produced scaffolds was measured by water contact angle (WCA) tests. The WCAs of aligned-fiber bicomponent scaffolds in different directions were examined. As shown in Figure l, the WCA was 122.3 ± 1.3 and 127.9 ± 3.0° in the aligned direction and the perpendicular direction, respectively.

In Vitro Release of Growth Factors

The release behavior of both growth factors from scaffolds was investigated. A fast glial cell line-derived growth factor (GDNF) release and a slower NGF release are shown in Figure j. 16.7% of NGF was released in initial 24 h and up to 34.8% of NGF was released after 42 days. During the initial release period (24 h), 20.3% of GDNF was released, and 62.5% GDNF release was achieved after 42 days. The results showed that sustained and dual release of growth factors with different release behaviors was achieved.

Cell Morphology

PC12 cells were cultured on scaffolds and their morphology is shown in Figure . In general, cell adhesion, spreading, and migration on all types of scaffolds were improved with increasing culture time.

Figure 2

Confocal micrographs of PC12 cells grown on scaffolds: non-aligned fiber + GFs (a,e,i), non-aligned fiber only (b,f,j), aligned fiber + GFs (c,g,k), and aligned fiber only (d,h,l) at days 1, 4, and 7, respectively. Scale bar: 30 μm.

At day 1, cells on non-woven scaffolds without growth factors (Figure b) were in spherical shape with little spreading. Cells on growth factor-containing non-woven scaffolds (Figure a) exhibited improved cell attachment, spreading, and migration. After 4 days of culture, more cells were elongated to an elliptical shape, and improved neurite extensions were observed on scaffolds with GFs (Figure e) as compared to those without GFs. The neurite sprouts on both non-woven scaffolds protruded in random orientations. Most cells showed an elongated elliptical shape on aligned-fiber scaffolds in spite of no growth factor stimulation (Figure h). Much improved neurite sprouting and elongation was also seen on the aligned-fiber scaffolds as compared to the cell morphology on non-woven fibrous scaffolds without growth factors. Some of the neurite protrusions showed an alignment in the direction of the red-dotted line (Figure g). After 7 days of culture, cells on growth factor-containing non-woven scaffolds (Figure i) showed improved neurite branching and neurite outgrowth. Neurite branching and outgrowth protruded in random orientations on both types of non-woven scaffolds. Cells on aligned-fiber scaffolds appeared to spread in a well-organized manner. Much increased neurite sprouting and elongation was also seen on the aligned-fiber scaffolds as compared to the cell morphology on non-woven fibrous scaffolds without growth factors. The neurite alignment was obviously seen as revealed by the read-dotted line.

Analysis of Neurite Differentiation

Neurite differentiation and alignment were characterized as shown in Figure . At day 1, cell differentiation in the GF-containing group was higher than that in the control group. More cell differentiation was obtained in the aligned-fiber group compared with the non-woven group. The results showed limited neurite differentiation was induced at day 1. At day 4, cell differentiation in the GF-containing group was much higher than that in the control group. More cell differentiation was obtained in the aligned-fiber group compared with the non-woven group. The longest neurite outgrowth was achieved on GF-containing aligned-fiber scaffolds. At day 7, much higher cell differentiation percentage was obtained in the GF-containing group compared with the control group. More cell differentiation was achieved in the aligned-fiber group compared with the non-woven group. Significant neurite extensions were obtained on both aligned-fiber scaffolds without GFs and GF-containing non-woven scaffolds. A random distribution of neurite orientations was found in non-woven fibrous scaffolds (Figure d). However, a highly organized arrangement of neurite orientations was noticed in both types of aligned-fiber scaffolds (Figure c), indicating that a good neurite alignment was achieved attributed to the fiber alignment.

Figure 3

Differentiation percentages of PC12 cells grown on scaffolds (a). Neurite length (dotted line: 28 μm) (b). Distribution of neurite orientations on aligned-fiber scaffolds (c) and non-woven scaffolds (d).

Cell Morphology on Scaffolds

SEM images of cells grown on scaffolds are shown in Figure . Enhanced cell attachment, spreading, and migration over culture time were observed. Cells proliferated well on fibrous scaffolds and covered a large portion of the scaffold surfaces at day 7. Cell differentiation characterized by neurite outgrowth on fibrous scaffolds increased over culture time as more cells being in the elongated elliptical shape with neurite protrusions were seen at day 4 and day 7. The stimulation of GFs released from fibrous scaffolds significantly improved cell differentiation in both non-woven scaffolds and aligned-fiber scaffolds as more cells in the elliptical shape bearing neurite outgrowth but less cells in the spherical shape were clearly seen on GF-containing scaffolds as compared to scaffolds without GFs. The fiber alignment also contributed to the elongation of cells, neurite outgrowth, and neural differentiation. High-resolution SEM micrographs (insets in Figure g,h,k,l) revealed more details about neurite outgrowth and cell–scaffold interactions on aligned-fiber scaffolds. It was found that neurite protrusions tended to extend along fibers underneath and appreciable amounts of short filopodia and microspikes protruded out of the cell body toward adjacent fibers. The fiber alignment noticeably increased cell differentiation as much more neurite outgrowth was observed on aligned-fiber scaffolds compared with non-woven fibrous scaffolds. These results showed that the fiber alignment in fibrous scaffolds and released GFs promoted neurite outgrowth and neural differentiation independently or synergistically.

Figure 4

SEM images of PC12 cells grown on scaffolds: non-aligned fiber + GFs (a,e,i), non-aligned fiber only (b,f,j), aligned fiber + GFs (c,g,k), and aligned fiber only (d,h,l). Scale bar: 50 μm.

Discussion

Scaffold-based nerve guidance conduits mimicking the microenvironments are important for peripheral nerve regeneration. Studies have shown that plain nanostructured scaffolds can provide topographical cues to enhance neurite outgrowth. Neuronal growth and axonal extension can also be promoted by neurotrophic factors secreted from surrounding.[4] Due to the vulnerability and short half-lives of these growth factors, delivery vehicles are essential for their controlled release.[23,24] Therefore, the combination of sustained delivery of dual growth factors (GDNF and NGF) and topographical signaling to seeded cells within a single scaffold would be advantageous.

Biocompatible scaffolds resembling natural ECM can induce enhanced cell adhesion, migration, and proliferation.[25−28] Electrospun fibers have been employed to deliver various bioactive molecules including proteins and nucleic acids.[21−23,29−31] In this study, scaffolds with dual growth factor delivery and an aligned-fibrous structure were produced, attempting to provide biochemical cues and topographical cues simultaneously for enhanced neurite outgrowth, differentiation, and neurite alignment.

Topographical signaling plays important roles in cell sensing, focal adhesions, cell contractility, and other cell activities.[32,33] It was reported that neuronal differentiation and neurite alignment of PC12 cells could be achieved on various nanogratings with or without the stimulation of biochemical cues.[6,8,10] Highly aligned electrospun fibers could guide neurite outgrowth along fibers.[34] It was also claimed that the fiber diameter could influence the cellular response and determine cell fate.[35,36] Aligned fibers could be produced through a high-speed electrospinning (HS-ES) technique, which could help alleviate or eliminate bending instability of electrospun fibers and consequently facilitate fiber alignment.[37,38] The fiber alignment would decrease with the increase of the scaffold thickness.[39,40] Fibrous bicomponent scaffolds with aligned-fiber structures and an even distribution of both types of fibers were achieved here through the HS-DSDP-ES.

It is well known that wettability of scaffolds would influence cell spreading and adhesion. The wettability of produced aligned-fiber bicomponent scaffolds in different directions was both above 120° (Figure i), revealing that the produced fibrous scaffold was hydrophobic. The wettability of bicomponent scaffolds was predominantly influenced by the PDLLA component, which was consistent with the study reported by other groups.[41] In wettability tests, the static contact angle in the x-direction (aligned direction of fibers) was defined as θ and the static contact angle in the y-direction (perpendicular direction to the fiber alignment) was defined as θ. Wetting anisotropy was defined as Δθ = θ – θ. In this study, θ was 127.9°, θ was 122.3°, and Δθ was 5.6°. The anisotropy was caused by preferential spreading and ingression of water droplets in the x-direction.[42]

Dual release of growth factors with different release profiles was achieved in this investigation. NGF and GDNF can both stimulate neurite outgrowth and neural differentiation through a specific signaling pathway.[3,43] In our recent study, it was found that dual delivery of NGF and GDNF could promote neurite outgrowth and neural differentiation synergistically.[31]

Contact guidance provided by aligned fibers induced a much higher level of neurite outgrowth and neuronal differentiation and an obvious neurite alignment (Figure d,h,l) as compared to non-woven fibrous scaffolds (Figure b,f,j). Interestingly, cells on randomly oriented fibrous scaffolds showed partial neurite outgrowth and neuronal differentiation but no neurite alignment over the culture time, indicating that cells also responded to topographical cues provided by randomly oriented fibers. The neurite outgrowth and neuronal differentiation might be attributed to short-ranged topographical guidance. The long-ranged topographical guidance was disturbed by randomly oriented fibers and resulted in no neurite alignment. The neurite outgrowth on fibrous scaffolds and cell-scaffold interactions were revealed by SEM micrographs.

Elongated cells with an elliptical shape on fibrous scaffolds differentiated as characterized by neurite outgrowth tending to extend along fibers underneath. The observations might be explained by the involvement of focal adhesions, which coordinate cell polarity and cytoskeleton arrangement and consequently motivate neurite migration and path-finding.[44,45] In this study, the difficulty in establishing focal adhesions to adjacent fibers resulted in cytoskeleton organization and cell migration preferentially along electrospun fibers underneath, which consequently led to cell polarity and neurite path-finding. The difficulty in establishment of focal adhesions to adjacent fibers was revealed by high-resolution SEM micrographs (insets in Figure g,h,k,l). Appreciable amounts of very short filopodia and microspikes protrude out of the cell body toward adjacent fibers in the orthogonal direction of the fiber alignment but most of them failed to anchor to the adjacent fibers. Therefore, a much higher level of neurite outgrowth and neuronal differentiation and more obvious neurite alignment were observed in aligned-fiber scaffolds.

Interestingly, the degree of neurite alignment was significantly higher than that of the fiber alignment, suggesting that the neurite alignment was tolerant of topographical noise to a certain extent. The degree of neurite alignment in GF-containing aligned-fiber scaffolds was slightly higher than that in the aligned-fiber scaffolds without growth factors, indicating that growth factors contributed to the neurite alignment synergistically in the context of topographical guidance.

It can be concluded that growth factors released from scaffolds and topographical cues induced enhanced neurite outgrowth and differentiation in different manners. With the combinatory stimulation of biochemical cues and topographical cues, most neuronal differentiation and neurite alignment was achieved. Although these results highlighted the effectiveness of aligned bicomponent fibrous scaffolds in guiding neurite outgrowth, neuronal differentiation, and neurite alignment, the non-human cell line used in this study has a limited translational value for nerve conduit application. More representative human neural cells should be utilized in the future investigation of nerve tissue regeneration.

Conclusions

Fibrous bicomponent scaffolds providing both biochemical cues and contact guidance were fabricated through HS-DSDP-ES. The dual delivery of GDNF and NGF and the topographical cues in aligned bicomponent scaffolds induced significant neurite extension, neuronal differentiation, and neurite alignment in a synergistic manner. The elaborately designed scaffolds providing multiple microenvironmental signals in a controlled manner are promising for potential applications.

Experimental Section

Materials

Poly(lactic-co-glycolic acid) (PLGA, LA/GA = 50:50, Mw = 100 kDa) and PDLLA (Mw = 100 kDa) were supplied by Lakeshore Biomaterials. The human β-NGF with the ELISA kit was supplied by Peprotech Inc., and human GDNF with the ELISA kit was supplied by R&D Systems, Inc. The culture medium and other reagents were purchased from Invitrogen, Inc. All chemicals were used as purchased.

Fabrication of Electrospun Scaffolds

NGF or GDNF containing DI water (supplemented with 0.5% BSA) and 5 wt % Span-80 were mixed with PLGA/chloroform or PDLLA/chloroform polymer solutions (15%, w/v) at a volume ratio of 10:1 for 10 min at 300 rpm to prepare water-in-oil (w/o) emulsions. The applied voltage during electrospinning was set at 16 kV. A needle tip with a 0.8 mm inner diameter was used. The needle-to-collector distance was optimized at 8 cm, and the feeding rate of emulsions was set at 2 mL/h. A drum collector rotating at 3000 rpm was used to make aligned-fiber scaffolds. The emulsion formula is shown in Table . The fabricated scaffolds were freeze-dried for 24 h.

Table 1

Emulsion Formula

	GDNF/PLGA emulsions		NGF/PDLLA emulsions
scaffold	water phasea	oil phaseb	water phasea	oil phaseb
aligned-fiber scaffolds + GFs	GDNF	PLGA	NGF	PDLLA
aligned-fiber scaffolds	NA	PLGA	NA	PDLLA
non-woven scaffolds + GFs	GDNF	PLGA	NGF	PDLLA
non-woven scaffolds	NA	PLGA	NA	PDLLA

5 μg of growth factors was added to the water phase (1 mL H2O).

The oil phase was formed by dissolving 1.5 g of polymers in 10 mL CHCl3.

Characterization

Produced scaffolds were processed using a sputter coater (BEL-TACSCD005) and examined under a scanning electron microscope (Hitachi S-4800 FEG, Japan). WCAs of scaffolds at room temperature were examined using a measuring machine (Solon SL200B, China).

In Vitro Release Tests

For in vitro release tests, previous procedures were followed.[46] PBS release medium containing 0.5% BSA, 0.1% heparin, 0.05% Tween-20, and 0.02% NaN3 was prepared. 50 mg freeze-dried scaffolds were immersed in 3 mL release medium and incubated in a 37 °C water bath. About 0.4 mL release medium was collected and replaced by 0.4 mL fresh release medium at different times during the 42 day tests. The concentration of growth factors in the collected release medium was measured by ELISA.

In Vitro Experiments

PC12 cells (Biowit, China) were maintained and expanded in DMEM medium and incubated in a 5% CO2 atmosphere at 37 °C. The culture medium was replaced every 2 days. When the cultured cells reached 80% confluence, they were used for further experiments.

Cell Adhesion and Proliferation

Scaffold samples (5 mg each) in the square shape were sterilized by 60Co γ-irradiation at a dose of 15 kGy before any further experiment. Samples were immersed in the culture medium and fixed at the bottom of the culture plate. About 4 × 103 PC12 cells were seeded on each scaffold. After preset culture times, the cell-scaffold constructs were collected and washed two times with PBS, followed by 10 min fixation with 4% PFA at room temperature. After that, cells were treated with 0.1% Triton X-100 in (1% w/v) BSA block solution for 30 min. The cytoskeletons and nuclei of cells were stained with FITC phalloidin (1:40 dilution) for 30 min and DAPI (1:1000 dilution) for 5 min. Cells were observed using a confocal laser scanning microscope (CLSM 710, Carl Zeiss).

The samples with cells were rinsed with PBS and fixed with 2.5% glutaraldehyde at 4 °C for 4 h, followed by further washing with 0.1 M sucrose-containing sodium cacodylate buffer, PBS, and DI water. The cell-scaffold samples were further freeze-dried for 48 h, sputter-coated with gold, and photographed using a scanning electron microscope.

Image Analysis

Image J software (NIH, USA) was used for morphological analysis. SEM micrographs and confocal images were loaded into Image J to measure the fiber alignment, neurite length, and neural orientation. Outgrowth of cells longer than 28 μm were counted as neurites. At least 100 fibers or cells in each scaffold were randomly selected for measuring the fiber alignment, neurite length, and neural orientation.

Statistical Analysis

All of the data were achieved at least in triplicate, and the results were expressed as the mean and standard deviation.