Cocoons of B. mori were purchased from Tongxiang, China. Semipermeable cellulose membranes with a molecular weight cutoff of 14,000 ± 2,000 Da were purchased from Yuanju Co., Ltd. (Shanghai, China). All other chemicals were of analytical grade and were purchased from Sinopharm Chemical Reagent Co. Ltd. (Shanghai, China).

In order to remove the sericin, the cocoons were degummed in 0.5 wt% Na₂CO₃ aqueous solution at 100°C for 30 min and rinsed repeatedly with deionized water. The degummed silk fibers were then dried at 4°C overnight and immersed the next day in 9.0 M LiBr aqueous solution at 40°C for 2 h to completely dissolve them. The solution was centrifuged at 1,200 × g for 10 min at 10°C to remove undissolved impurities. Next, the supernatant was placed in a semipermeable cellulose membrane and dialyzed against deionized water for 3 days to remove LiBr ions. The obtained RSF solution was concentrated to 33 wt% by forced air flow for further experiments. According to the non-gel sieving capillary electrophoresis (NGSCE) results, it was found that the RSF had an molecular weight (MW) around 83 kDa with a wide MW distribution (MWD). The RSF molecules are mainly the segments of the heavy chain of natural B. mori silk fibroin (Wei et al., 2010).

The traditional method to electrospinning RSF mats was performed as described by Li (2015), while the preparation of electrospun RSF mats with aligned fibers was performed as described by Liu (2016). For both methods, a syringe containing 33 wt% RSF dope was run through a stainless-steel needle with an inner diameter of 0.6 mm using a syringe pump with a flow rate of 1.2 mL/h. A voltage of 20 kV was applied between the spinneret and collector, with a distance of 12 cm. The spinning temperature was kept at room temperature, and the relative humidity was 50 ± 5%. For conventional electrospun RSF mats with random fibers, the collector was an aluminum foil-coated steel plate placed perpendicular to the needle. For electrospun RSF mats with aligned fibers, the collector was an aluminum foil-coated roller rotating at a speed of 2,000 rpm. The as-spun RSF mats were peeled off the collectors and placed in a test chamber at a constant 90% relative humidity and 37°C for 36 h to increase the crystallinity of the mats, and the thickness of the samples were measured by a thickness gauge. Before cell seeding, the post-treated RSF mats were soaked in 75 vol% ethanol for 2 h, then washed with sterilized PBS and cell culture medium three times and coated with laminin solution (5 mg/mL, Sigma-Aldrich) in PBS for at least 4 h at 37°C prior to soaking in proliferation medium overnight.

Tensile properties of the post-treated RSF mats (35 × 5 mm) were measured using an Instron 5,969 material testing machine at 20 ± 5°C and 50 ± 5% relative humidity. Samples were tested at an extension rate of 3 mm/min at a gauge length of 20 mm. The thickness of each mat sample was measured 10 times by a CH-1-S thickness gauge (Shanghai Liuling Instruments Co., Shanghai, China) with a resolution of 1 μm (Li, 2015). In the case of the RSF mats with aligned fibers, the tensile strength was measured along the direction of the fiber alignment.

Two different media were used in this study. The proliferation medium consisted of 2% B27 (Life Technologies), 20 ng/mL epidermal growth factor (Thermo Fisher Scientific, Rockford, IL, USA), 10 ng/mL basic fibroblast growth factor (Invitrogen), and 1% penicillin-streptomycin (BBI Life Sciences) in DMEM/F12 (Gibco). The differentiation medium consisted of 2% B27, 1% fetal bovine serum (Life Technologies), 1 μM retinoic acid (Sigma-Aldrich, USA), and 1% penicillin-streptomycin in DMEM/F12.

As previously described (Guo et al., 2016), the NPCs were extracted and purified from the hippocampus of embryonic day 16 ICR mice (adult mice for producing embryos were purchased from Shanghai Jiesijie Animal Experiment Co., Ltd.) and then seeded at a concentration of ~5 × 10⁴ cells/mL in a 25 cm³ culture flask (Corning) in proliferation medium to form NPC neurospheres. For proliferation studies, the NPCs were seeded at a concentration of ~8 × 10⁴ cells/mL in proliferation medium. For differentiation studies, the NPCs were seeded at a concentration of ~8 × 10⁴ cells/mL in differentiation medium. For cell seeding, the neurospheres were gathered and enzymatically digested with Accutase (Life Technologies) to obtain a suspension of single cells. After seeding, the RSF mats together with seeded NPCs were moved to 24-well cell culture plates (Corning). Cells in 25 cm³ culture flask were routinely passaged 1:2, and for propagation, the NPC neurospheres were collected after 7 days in culture and dissociated with Accutase at 37°C for 7 min followed by the addition of 0.01 M PBS to block the reaction. The cells were then re-plated into 25 cm³ culture flasks, and the propagation was repeated at 7-day intervals. All cells were used at low passage numbers (N = 5–10). All animal procedures were performed according to protocols approved by the Animal Care and Use Committee of Fudan University and were consistent with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. All efforts were made to minimize the number of animals used and to prevent their suffering.

For imaging the morphology of NPCs on different substrates, cells were fixed in 4% paraformaldehyde for 2 h and then dehydrated with a series of graded ethanol solutions [30, 50, 70, 75, 80, 90, and 100% (v/v)]. The ethanol was removed and tert-butanol was added. Finally, the SF mats were frozen at −50°C for 4 h and then freeze-dried overnight. After sputter coating with platinum, the surface morphology of the prepared RSF mats and the prepared NPCs on RSF mats was imaged on a field emission scanning electron microscope (SEM) (SU8010, Hitachi, Japan) operating at an accelerating voltage of 10 kV. One hundred fibers in the SEM image were randomly selected, and the average diameter of the fibers was measured with the ImageJ software.

After 5 days of culture in proliferation medium, a cell viability test was performed using ethidium homodimer (Abcam) and calcein-AM (Abcam) according to the manufacturer's instructions. First, the NPCs cultured on the laminin-coated coverslip and on the RSF mats were rinsed three times in sterile 0.01 M PBS. Then the cells were incubated in sterile 0.01 M PBS containing 2 × 10⁻⁴ mM of calcein-AM and 2 × 10⁻⁴ mM ethidium homodimer for 30 min at 37°C. After incubation, the cells were rinsed three times in 0.01 M PBS for 10 min. The live NPCs labeled with calcein-AM showed green color and the dead NPCs stained with ethidium homodimer showed red color under a Leica SP8 confocal microscope. The percentage of living cells was calculated by dividing the number of calcein-AM–positive cells by the total cell number.

The NPCs were cultured for 1, 3, 5, and 7 days in proliferation medium and then used with the CCK-8 cell proliferation assay and cytotoxicity assay kit (DOJINDO, Japan) according to the manufacturer's instructions. In brief, NPCs were transferred to 24-well plates with coverslips or with aligned or random RSF mats and maintained in a humidified atmosphere with 5% CO₂ at 37°C. Then 10% v/v CCK-8 solution was added to the culture medium at the indicated time points and the wells were incubated for 4 h. Finally, 100 μL of cell solution was transferred to a 96-well plate and the absorbance at 450 nm was measured using a microplate reader.

After 5 days of culture in proliferation medium and 7 days of culture in differentiation medium, NPCs cultured on RSF mats and laminin-coated coverslips were washed once with 0.01 M PBS, fixed in ice cold 4% paraformaldehyde in 0.01 M PBS for 30 min, washed with 0.01 M PBS three times for 10 min, permeabilized with 1% Triton-X100 in 0.01 M PBS (1% PBS-T) for 30 min, and blocked with 10% BSA in 0.01 M PBS for 1 h at 37°C. The samples were incubated with the primary antibody overnight at 4°C and then incubated at 37°C for 1 h. The next day unbound primary antibodies were rinsed with 0.01 M PBS three times for 10 min and secondary antibodies were incubated for 1 h at room temperature. Unbound antibodies were then rinsed with 0.01 M PBS three times for 10 min followed by DAPI staining for 15 min. The Click-iT EdU Imaging Kit (Life Technologies) was used to detect cell proliferation according to the manufacturer's instructions. The primary antibodies used in this study included mouse anti-nestin IgG1 (1:1,000 dilution, Abcam), rabbit anti-neuron-specific class III beta-tubulin (Tuj-1, 1:1,000 dilution, Abcam), mouse anti-glial fibrillary acidic protein (GFAP, 1:1,000 dilution, Sigma-Aldrich), rabbit anti-Ki-67 (1:1,000 dilution, Abcam), and rabbit anti-vinculin (1:500 dilution, Abcam). The corresponding secondary antibodies were cy3 and 488 (1:500 dilution, Jackson ImmunoResearch). The nuclei were stained with DAPI (1:800 dilution, Sigma-Aldrich). All antibodies used in this study were diluted in 1% PBS-T.

The potential cytotoxicity of the RSF mats was measured by detecting apoptosis-related DNA fragmentation with a TUNEL staining kit (Click-iT® Plus TUNEL Assay for in situ Apoptosis Detection, Invitrogen) according to the manufacturer's instruction. The nuclei were labeled with DAPI (1:800 dilution), and the cells were evaluated by confocal microscopy (Leica, SP8, Germany).

Cells were harvested with Accutase and lysed in RIPA and PMSF buffer (99:1) on ice for 30 min to 1 h. Total protein was isolated and the concentration was measured using a bicinchoninic acid kit (Thermo Fisher Scientific). The collected protein samples were loaded on a 12% sodium dodecyl sulfate-polyacrylamide gel, separated by gel electrophoresis, and transferred to polyvinylidene difluoride membranes (Immobilon-P; Millipore, Bedford, MA, USA). The membranes were blocked in 5% nonfat dried milk in 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, and 0.1% Tween-20 (TBST) at room temperature for 1 h and then incubated with primary antibodies at 4°C overnight. The primary antibodies were anti-nestin (1:1,000 dilution, Abcam), anti-Tuj-1 (1:1,000 dilution, Abcam), anti-GFAP (1:1,000, dilution, Sigma-Aldrich), anti-Ki-67 (1:1,000 dilution, Abcam), and anti-GAPDH (1:1,000 dilution). Unbound primary antibodies were washed away with TBST three times for 10 min, and the membranes were incubated with goat anti-rabbit (1:1,000 dilution) and goat anti-mouse (1:1,000 dilution) secondary antibodies at room temperature for 1 h and then washed with TBST three times for 10 min. Finally, the ECL Detection Reagent (Pierce, USA) was used to measure protein expression with GAPDH as the internal control.

The NPCs were cultured in differentiation medium for 7 days, then the total RNA was extracted from the NPCs cultured on the coverslips and RSF mats using TRIzol (Invitrogen, Carlsbad, CA, USA). This was followed by complementary DNA synthesis using the GoScript Reverse Transcription System (Promega, Madison, WI, USA). qPCR reactions were performed with GoTaq® qPCR Master Mix (Promega) on a 7500HT Fast Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). Each PCR reaction was carried out in triplicate. The mRNA levels were normalized to GAPDH (internal control), and the relative quantification of gene expression was analyzed using the ΔΔCT method (Livak and Schmittgen, 2001). The primer sequences are shown in Table 1.

For every experiment in all conditions studied, six random images obtained at 40 × magnification with the Leica SP8 laser confocal microscope were picked for manual quantification. Judged by DAPI staining, each such micrograph depicted around 150 to 1,000 cells in fields with a diameter of around 500 mm. Microsoft Excel and GraphPad Prism 6.0 (GraphPad Software, San Diego, California) were used for statistical analysis of the data. Unpaired Student's t-tests were used to determine statistical significance when comparing two groups, and one-way ANOVA followed by a Dunnett's multiple comparisons test was used when comparing more than two groups. P-values < 0.05 were considered statistically significant. Data are shown as the mean ± SEM.

The fiber surfaces of the aligned and random RSF mats were smooth and uniform in diameter. The average thickness of these two types of RSF mats is about 0.12 mm. The fibers of the conventional electrospun RSF scaffolds were randomly distributed, while the fibers of the modified electrospun RSF mats were aligned along the rotating direction of the collector. The reason for this alignment was that the Taylor cone of the RSF dope was subjected to a certain traction force from the high-speed rotation of the roller (Figure 1A). The breaking strength of the RSF mats with aligned fibers were improved from 1.12 ± 0.02 MPa to 1.94 ± 0.15 MPa (Figure 1B).

As in previous literature, several methods were used to detect NPCs. After 5 days of culture under proliferation conditions, we observed the formation of free-floating neurospheres (Figure 2A), which indicated the presence of NPCs. Also, immunostaining against nestin, an NPC marker, showed that most of the cells were nestin positive (Figures 2B–G), further validating the NPCs in this study. After 3 days of culture under proliferation and differentiation conditions, the characteristics of the NPCs were carefully examined on the RSF mats. Because cell adhesion plays a vital role in regulating cell growth, migration, differentiation, proliferation, and apoptosis, we investigated the adhesion of NPCs on RSF mats. After culturing for 3 days, the NPCs seeded on the aligned or random RSF mats adhered to the surface of the mats (Figures 3A,B). SEM images showed that the NPCs extended pseudopodia and established connections with each other after 3 days of culture in proliferative medium on aligned and random RSF mats (Figures 3A,B). In addition, we observed cell adhesion on RSF mats without laminin coating according to SEM (Figure S1). Vinculin, a cytoskeleton protein, plays an important role in cell adhesion, stretching, movement, proliferation, and survival activities by combining and interacting with a variety of cytoskeleton proteins and cytoskeleton f-actin to participate in cell chemical signal transduction. Immunostaining for vinculin after 3 days of culture showed that NPCs could adhere to the RSF mats along the RSF fibers (Figures 3C–C”,D–D”).

The cytotoxicity of the aligned and random RSF mats was evaluated by calcein-AM and ethidium homodimer staining assay with laminin-coated coverslips as the control. After 5 days of culture, almost 95% of the attached cells were viable on all three substrates (Figures 4A,B), and the TUNEL assay showed no significant differences in apoptosis for any of the substrates (Figures 4C,D). This indicates that RSF mats have good biocompatibility, which is consistent with previous studies (Jiang et al., 2014). Cells cultured on RSF mats were also stained with antibodies against nestin. Nearly all of cells on the aligned and random RSF mats were immunopositive for nestin after 5 days of culture (Figures 2H,I), with no obvious difference compared to the laminin-coated coverslip. Furthermore, the real-time PCR analysis showed similar results for Nestin expression in NPCs grown for 5 days on aligned and random RSF mats and laminin-coating coverslips (Figure 2J), suggesting that NPCs grew well on the aligned and random RSF mats and maintained their stemness.

NPC proliferation on the aligned and random RSF mats was assayed by immunostaining for Ki-67, which is a nucleoprotein involved in the cell cycle and is an indicator of cell proliferation activity. As we expected, after 7 days of culturing in proliferation medium 19.49 ± 1.07% (SE) of the cells on the laminin-coated coverslip were Ki-67 positive (Figure 5D), while almost 43.85 ± 0.9% and 40.6 ± 1.06% of the NPCs were Ki-67 positive on the random (Figure 5E) and aligned RSF mats (Figure 5F), respectively. Thus, cell proliferation was significantly improved on RSF mats compared to laminin-coated coverslips (Figure 5H). EdU, which is a thymine nucleoside analog, can be incorporated in place of thymine during the DNA synthesis phase of the cell cycle and thus also acts as a marker of proliferation. The percentages of EdU-positive cells grown on random (Figure 5B) and aligned RSF mats (Figure 5C) were 60.47 ± 1.1% and 57.66 ± 1.27%, respectively, compared to 24.23 ± 1.1% on the coverslip (Figure 5A). Thus, the percentages of Ki-67–positive cells were analogous to the percentages of EdU-positive cells (Figures 5G,H). Western blot analysis demonstrated that the protein expression of Ki-67 in NPCs cultured on the aligned and random RSF mats was significantly greater than on the coverslips (Figure 5I). This suggests that NPCs on the RSF mats sustained a more active NPC proliferation state compared to the coverslips (consistent with the CCK-8 cell proliferation assay after culturing for 1, 3, 5, 7 days in proliferation medium respectively, Figure 5J).

To investigate the phenotypic changes of differentiated NPCs on RSF mats and coverslips, NPCs were observed after culturing for 7 days in differentiation medium. After 7 days of differentiation, the cells exhibited elongated cell shapes with healthy neurite outgrowth, leading to a confluent neural network covering almost the whole RSF surface (Figures 3A,B). After 7 days differentiation, we used Tuj-1 as a neuron marker and GFAP as an astrocyte marker (Figure 6A). The differentiated cells were 18.89 ± 0.53%, 36.18 ± 0.61%, and 31.39 ± 1.31% Tuj-1 positive on control coverslips, aligned, and random RSF mats, respectively (Figure 6B), and were 54.20 ± 1.82%, 23.58 ± 0.96%, and 22.54 ± 0.94% GFAP positive on coverslips and aligned and random RSF mats, respectively (Figure 6C). There was a significant difference in the number of GFAP-positive cells on the aligned and random RSF mats compared to coverslips and a significant difference in the number of Tuj-1–positive cells on the random RSF mats compared to coverslips. Thus, the NPCs retained their ability to differentiate into different neural subtypes on the RSF mats. For a quantitative analysis, the total protein was extracted and subjected to western blot assay, and total RNA was extracted and subjected to real-time PCR and qPCR analysis after 7 days of culture in differentiation medium. The western blot analysis showed that compared with controls, NPCs cultured on aligned and random RSF mats exhibited significantly lower nestin expression, while the expression of Tuj-1 was enhanced in cells grown on aligned and random RSF mats (Figure 6D). The nestin mRNA expression was lower in cells grown on RSF mats compared to those grown on coverslips, while the Tuj-1 mRNA expression was significantly greater in cells grown on the RSF mats compared to those grown on coverslips (Figure 6E). Thus, both the aligned and random RSF mats greatly enhanced NPC differentiation into neurons, while the laminin-coating coverslips greatly enhanced NPC differentiation into astrocytes. We further analyzed the signaling factors that might be involved in this process of differentiation using real-time PCR and qPCR analysis after 7 days of culture. There was no significant difference in the SHH signaling pathway among the three groups (Figure 6F), while the expression of factors in the BMP4 signaling pathway was reduced in both the aligned and random RSF mats compared to coverslips (Figure 6G).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.