Electrospun SF/PLCL nanofibrous membrane: a potential scaffold for retinal progenitor cell proliferation and differentiation

Biocompatible polymer scaffolds are promising as potential carriers for the delivery of retinal progenitor cells (RPCs) in cell replacement therapy for the repair of damaged or diseased retinas. The primary goal of the present study was to investigate the effects of blended electrospun nanofibrous membranes of silk fibroin (SF) and poly(L-lactic acid-co-ε-caprolactone) (PLCL), a novel scaffold, on the biological behaviour of RPCs in vitro. To assess the cell-scaffold interaction, RPCs were cultured on SF/PLCL scaffolds for indicated durations. Our data revealed that all the SF/PLCL scaffolds were thoroughly cytocompatible, and the SF:PLCL (1:1) scaffolds yielded the best RPC growth. The in vitro proliferation assays showed that RPCs proliferated more quickly on the SF:PLCL (1:1) than on the other scaffolds and the control. Quantitative polymerase chain reaction (qPCR) and immunocytochemistry analyses demonstrated that RPCs grown on the SF:PLCL (1:1) scaffolds preferentially differentiated toward retinal neurons, including, most interestingly, photoreceptors. In summary, we demonstrated that the SF:PLCL (1:1) scaffolds can not only markedly promote RPC proliferation with cytocompatibility for RPC growth but also robustly enhance RPCs’ differentiation toward specific retinal neurons of interest in vitro, suggesting that SF:PLCL (1:1) scaffolds may have potential applications in retinal cell replacement therapy in the future.


Results
Morphology of electrospun SF/PLCL nanofibrous scaffolds. In this study, thin scaffolds (with a thickness of approximately 60-100 um) of pure SF, SF/PLCL blends at different weight ratios, and pure PLCL were successfully produced by electrospinning, and the resultant nanofibrous scaffolds appeared to be homogeneous, as can be observed in Fig. 1A. The nanofibrous scaffolds presented demonstrable differences in transparency before and after immersion in PBS (Fig. 1A), and the blended scaffolds appeared to be more transparent than pure SF or pure PLCL. SEM images depicting the micromorphology of the electrospun nanofibrous scaffolds are shown in Fig. 1B-F, and the average fibre diameter of the scaffolds gradually decreased from 432.7 nm to 137.8 nm with increasing SF content ( Fig. 1G-K), demonstrating that all five types of scaffolds were constructed of randomly displayed fibres and completely interconnected pore structures. The equilibrium swelling ratio (ESR) results are shown in Supplementary Fig. S1, and the results showed that the swelling ratio of the scaffolds decreased with an increase in the PLCL content. The swelling ratio of SF:PLCL (1:1) was 0.80 ± 0.53, indicating small changes in the fibre diameter from before to after immersion.
Mechanical and pore size measurements. The mechanical properties of SF:PLCL (3:1), SF:PLCL (1:1), SF:PLCL (1:3) and pure PLCL were reflected by typical tensile stress-strain curves, which are shown in Fig. 2A-D. The quantitative analyses are exhibited in Fig. 2E-G. The pure SF scaffolds were brittle, and thus, their mechanical properties could not be tested. With an increase in the ratio of PLCL, the scaffolds transformed from brittle to flexible, and obvious increases in the average tensile strength and elongation at break were obtained. In addition, the Young's modulus (SF:PLCL (3:1) 117.62 ± 46.2 MPa vs. SF:PLCL (1:1) 105.34 ± 17.37 MPa vs. SF:PLCL (1:3) 46.95 ± 16.20 MPa vs. pure PLCL 13.562 ± 2.89 MPa) was clearly decreased. The blending of PLCL with SF can markedly improve the mechanical properties of SF to yield employable blended biomaterials. The pore diameter was measured, and the results are shown in Supplementary Figure S2. All the scaffolds were demonstrated to consist of compact pores, with a pore diameter less than 2 μ m, indicating that RPCs can grow on the surface of all the scaffolds.
Water contact angles of different SF/PLCL scaffolds. To clarify the effect of the SF content on the surface properties of the fibrous scaffolds, the wettability was measured through a water contact angle analysis (Fig. 2). Pure SF appeared to be completely hydrophilic, with a water contact angle of 0° at 15 s, whereas pure PLCL exhibited an angle of approximately 120° at 30 s (Fig. 2H). The addition of SF to the surface of the PLCL scaffolds resulted in a smaller water contact angles than those of pure PLCL, which suggested that these scaffolds presented better hydrophilicity in comparison to pure PLCL (Fig. 2H). Cell proliferation morphology on electrospun scaffolds. Figure 3A-F shows fluorescent micrographs of GFP + RPCs on different scaffolds 3 days after their seeding in proliferation medium, and Fig. 3G-L presents the DAPI-stained cell nuclei morphology. In general, the cells showed a healthy morphology on all the scaffolds and appeared in the form of cell clusters, which may indicate that the RPCs maintained an undifferentiated state on the nanofibrous scaffolds. After 3 days of culture, the cell number and diameter of the clusters on all of the blended scaffolds were higher than those found for the control and pure PLCL, and SF:PLCL (1:1) presented the cell clusters with the largest diameter. To visualise the morphology of the RPCs seeded on electrospun scaffolds, SEM images were taken after 3 days in culture ( Fig. 3M-Q). RPCs grown on blended SF/PLCL scaffolds adopted a cell-cluster morphology, and the diameters of the cell clusters were larger than those obtained on the pure PLCL scaffolds. In addition, as shown in the SEM micrograph, SF:PLCL (1:1) and SF:PLCL (1:3) scaffolds appear much more attractive for RPC attachment than pure SF or PLCL scaffolds.
Cytocompatibility detection of SF/PLCL nanofibrous scaffolds on RPC growth. To assess cell adhesion, the expression levels of the cell adhesion molecule cadherin 4 were tested. The qPCR results show that the expression levels of cadherin 4 after 3 days in culture were significantly higher in the SF:PLCL (1:1) group than in the control groups, suggesting that SF:PLCL (1:1) scaffolds may be more attractive for RPC adhesion (Fig. 4A). In addition, the attachment of RPCs on the scaffolds after 12 hours was analysed using a CCK8 test. WST-8-formazan (2-(2-4-nitrophenyl methoxy-)3-(4-nitrophenyl)-5-(2,4-disulfonic acid benzene)-2H-tetrazolium monosodium salt formazan, an orange final product in the CCK8 assay) serves as an intermediate to reflect the number of living cells left behind in the cell attachment test. Our results show that the SF:PLCL (1:1) scaffolds had the highest optical density (O.D.), which indicated that a markedly higher number of cells remained on the SF:PLCL (1:1) scaffold than on the other scaffolds ( Fig. 4B and Supplementary Fig. S3). DAPI staining of the cells remaining before and after PBS washing further confirmed the above-mentioned results (Fig. 4C). The acute cytotoxicity of nanofibrous scaffolds of different SF/PLCL weight ratios on the health of the cultures after 24 hours was also assessed using the cytosolic enzyme lactate dehydrogenase (LDH) assay, which clearly showed that the SF/PLCL nanofibrous scaffolds exerted no cytotoxicity on RPC cultures (Fig. 4D). In addition, the qPCR results show that RPCs cultured on electrospun membranes for 7 days showed comparable or lower expression levels of the inflammation factors IL-6 and MCP-1 and the apoptotic factor caspase 3 than those in the control group, indicating a marked down-regulation of the expression of IL-6 and caspase 3 in the RPC cultures grown on SF:PLCL (1:1) nanofibrous scaffolds (Fig. 4E). All of these data indicate that all the SF/PLCL scaffolds, particularly the SF:PLCL (1:1) nanofibrous scaffolds, present cytocompatibility for RPC growth.

Effect of SF/PLCL nanofibrous scaffolds on RPC proliferation.
To evaluate the effect of nanofibrous scaffolds on RPC proliferation, qPCR was performed (samples were normalised for number of cells), and the results showed that the expression levels of Ki-67, a cellular marker for proliferation, were significantly higher on the pure SF and blended SF/PLCL scaffolds, particularly SF:PLCL (1:1), than in the other groups (Fig. 5A), suggesting that RPCs grown on the SF:PLCL (1:1) nanofibrous scaffolds sustained a more active proliferation state. Nestin expression was indicative of neural stem or progenitor cells and is used as a maker for undifferentiated retinal progenitor cells. Its expression in the cultures grown in the SF:PLCL (1:1) scaffold was comparable to that observed in the control group, indicating that the SF:PLCL (1:1) nanofibrous scaffolds are suitable for RPC self-renewal (Fig. 4B). Immunocytochemistry analysis showed that most cells on the SF:PLCL (1:1) nanofibrous scaffolds stained positively for Ki-67 (75.3 ± 5.77%), and this percentage was markedly higher than that obtained on the pure PLCL scaffolds or the control (glass coverslips) (51 ± 6.67% and 49 ± 3.33%, respectively) ( Fig. 4C-G). In addition, the CCK8 analysis, as shown in Fig. 4H, demonstrated that all of the nanofibrous scaffolds were suitable for RPC proliferation. No obvious differences in proliferation capacity were observed in the first 24 h of culture among the different groups. Thereafter, a significantly promoted expansion ability was recorded for the RPC cultures treated with pure SF as well as all those cultured in the blended SF/PLCL nanofibrous scaffolds, which presented O.D. 450 values higher than those obtained for the control group, and the SF:PLCL (1:1) scaffold presented the highest O.D. 450 values. These results indicate that the blended SF/PLCL scaffolds, particularly the SF:PLCL (1:1) scaffolds, could provide a significant advantage as a nanofibrous scaffold material for RPC proliferation, which is important for obtaining a large number of cells for further RPC research as well as for future applications in retinal cell replacement therapy.

Effects of electrospun SF/PLCL nanofibrous scaffolds on RPC differentiation.
Under differentiation conditions, the cells grown on the SF:PLCL (1:1) scaffolds exhibited normal cell shapes with healthy neurite outgrowth in the fluorescent and SEM micrographs, as shown in Fig. 6A,B. In addition, the effects of SF/PLCL nanofibrous scaffolds on RPC differentiation were also investigated through qPCR and immunocytochemistry. The present study investigated the expression of three key genes involved in retinal development: rhodopsin (a marker for rod photoreceptor cells), MAP2 (a marker for neuronal cells) and glial fibrillary acidic protein (GFAP, a glial marker). The qPCR results, as shown in Fig. 6C-E, show that RPC cultures grown on SF:PLCL (1:1) nanofibrous scaffolds exhibited a marked up-regulation of rhodopsin and MAP2 expression (3.1-fold and 2.9-fold, respectively) in comparison to the control group. By contrast, the expression level of GFAP was markedly lower in the RPC cultures grown on SF:PLCL (1:1) scaffolds than in the control cells. The immunocytochemistry analysis showed that in the RPC cultures grown on the SF:PLCL (1:1) scaffolds, the percentage of cells expressing rhodopsin or MAP2 was significantly higher, whereas the percentage of GFAP-positive cells was clearly lower compared with the other groups ( Fig. 7), which is consistent with the qPCR results. These results suggest that RPCs grown on SF:PLCL (1:1) scaffolds under differentiation conditions are more likely to differentiate toward retinal neuronal lineages, including, most interestingly, photoreceptor cells.
Taken together, our data demonstrate that the SF:PLCL (1:1) nanofibrous scaffolds present cytocompatibility for RPC growth. Moreover, the SF:PLCL (1:1) scaffolds can markedly promote RPC proliferation and can robustly accelerate the differentiation of RPCs into retinal neuronal cells, including photoreceptors.

Discussion
Increasing lines of evidence suggest that the transplantation of RPC with a carrier, instead of the injection of cell suspensions, may be more feasible for RPC survival, differentiation and further migration into an injured retina 17 . An ideal carrier for RPC transplantation should exhibit the following properties: a porous ultrastructure to allow the transport of nutrients and metabolic wastes and good cytocompatibility for cell adhesion, growth and maintenance of multipotency. In the present study, SF and PLCL were chosen for the fabrication of blended scaffolds (SF/PLCL) for two reasons: First, PCL and PLLA have appeared to be advantageous in RPC transplantation experiments 16,27 , suggesting that their copolymer PLCL may also perform well. Second, bioactiveSF can provide biological functional groups on the surface of PLCL scaffolds, which may have positive effects on RPC growth, as supported by a previous study that demonstrated that SF scaffolds can support the survival, migration and differentiation of embryonic stem   cell-derived neural progenitors 28 ; thus, because retinal progenitors are a type of neural progenitor, the above-mentioned lines of evidence indicate that SF may have a similar effect on RPC growth behaviour.
The good mechanical properties of a scaffold are helpful for successful cell transplantation. Retinal tissue has been reported to have an elastic modulus of 0.1 MPa, which indicates that it is a soft and flexible tissue 29 . PLGA has been reported to have an elastic modulus of 1.4-2.8 GPa 30 . In comparison to PLGA, our results showed that SF:PLCL (1:1) has an elastic modulus of 105.3 ± 17.4 MPa, which is closer to that of retinal tissue. However, further study was needed to improve the mechanical properties of the scaffolds in order to better match those of retinal tissue.
In the present study, we show that the SF/PLCL nanofibrous scaffolds exhibited good cytocompatibility, which is a requirement for any biomaterial used for clinical application. Few previous studies have addressed the effects of artificial scaffolds on the expression of inflammation and apoptosis factors by RPCs. In the present study, the expression levels of IL-6, MCP1 and caspase 3 in the RPC cultures grown on the SF/PLCL scaffolds were assessed. IL-6, a pro-inflammatory cytokine, is relevant for intraocular inflammatory diseases 31 . MCP-1 is considered the key gene in the migration of immune-competent cells as well as monocyte infiltration during retinal inflammatory diseases 32 . Caspase 3 is a major terminal cleavage ribozyme in the apoptosis process 33 . In this study, the expression levels of IL-6, MCP-1 and caspase 3 in RPCs cultured on SF:PLCL (1:1) nanofibrous scaffolds were not obviously up-regulated. Moreover, the expression of the cell adhesion molecule cadherin 4 was markedly up-regulated in the cell cultures grown on the SF:PLCL (1:1) scaffolds. All of these data suggest that the electrospun SF:PLCL (1:1) nanofibrous scaffolds present cytocompatibility with RPC growth.
Our data demonstrate that SF:PLCL (1:1) markedly enhanced RPC proliferation, possibly through the following potential mechanisms. First, the resultant nanofibrous scaffolds demonstrate interconnected porous structures with high porosity, as determined by SEM micrography, which would allow an exchange of nutrients and provide a high surface area, thereby mimicking the topographic features of the ECM. A previous study has shown that mimicking the ECM is important for organising cells and for providing signals for cellular responses 34 . In our study, all the scaffolds consisted of compact pores (with a pore diameter less than 2 μ m), which would provide larger surface areas for protein adsorption as well as enhance the nanoscale roughness, thus contributing to increases in cell attachment and proliferation 35,36 . Second, electro-spinning of SF (with its powerful hydrophilicity) with PLCL will enhance the hydrophilicity of PLCL. The balance between hydrophilicity and hydrophobicity of the scaffold materials may exert positive effects on cell growth 37 . Third, the introduction of biological functional groups, such as -NH 2 and -COOH, via SF in SF/PLCL may enhance the proliferation of RPCs. Faucheux and colleagues have demonstrated that chemical functional groups, such as -NH 2 and -COOH, can promote cell-scaffold recognition and subsequent proliferation 38 . In this study, the SF:PLCL (1:1) nanofibrous scaffolds were the most helpful for RPC proliferation, and increasing the amount of SF in the SF/PLCL scaffolds did not yield better effects. A previous study conducted by our group showed that the process of introducing bioactive amino groups on the surface of polymers exerted only a limited positive effect on RPC growth 14 ; although the reasons may be various, the strong negative charge on the surface of SF may partly explain this phenomenon 39 .
Inspiringly, our current study demonstrated that SF/PLCL nanofibrous scaffolds, particularly SF:PLCL (1:1) scaffolds, can markedly enhance the differentiation of RPCs toward retinal neurons, including photoreceptors, which are one of the most interesting retinal neuronal cells in retinal cell replacement therapy research. The positive effects of the SF:PLCL (1:1) scaffolds on inducing RPCs to produce more retinal neurons may be attributable to the introduction of bioactive groups, such as -NH 2 , on the surface of PLCL via SF, which may be supported by a previous report that neural stem cells in contact with glass surfaces modified by -NH 2 were more likely to differentiate into neurons 40 . Furthermore, a previous study demonstrated that a fibre diameter of approximately 200 nm was efficient for neural stem cell differentiation 41 . This may partially explain why SF:PLCL (1:1) (with a fibre diameter of approximately 200 nm) was efficient for RPC differentiation. Although the reasons regarding why SF/PLCL would cause an obvious increase in RPC differentiation towards photoreceptor phenotypes are undiscovered, the possible chemical cues in conjunction with microtopography may have an effect on promoting rod-specific differentiation 15 . The two major clinical subtypes of retinal degeneration (RD), namely retinitis pigmentosa and age-related macular degeneration (AMD), share the hallmark of photoreceptor cell degeneration, which ultimately results in vision loss 1 . The use of SF:PLCL (1:1) scaffolds in this study produced markedly increased populations of photoreceptor cells, which will improve future cell replacement therapies for retinal degenerative diseases.

Conclusion
In this study, the SF:PLCL nanofibrous membrane was used as a novel scaffold for RPC growth in vitro. Our data demonstrate that the SF:PLCL (1:1) nanofibrous scaffolds exhibited cytocompatibility for RPC growth and that these scaffolds can not only markedly promote RPC expansion but also robustly enhance RPC differentiation toward photoreceptors, the most interesting class of retinal neuronal cells for retinal cell replacement therapy. Subsequent investigations will be needed to evaluate the cytocompatibility of SF:PLCL (1:1) nanofibrous scaffolds and to measure the effects of the scaffolds on the proliferation and differentiation of RPCs in vivo.
Scientific RepoRts | 5:14326 | DOi: 10.1038/srep14326 Materials and Methods SF/PLCL nanofibrous scaffolds preparation. Raw silk (Jiaxing Silk, China) was degummed three times with 0.5% (w/w) Na 2 CO 3 solution at 100 °C for 30 min and then washed with distilled water. The degummed silk was dissolved in a ternary solvent system consisting of CaCl 2 /H 2 O/EtOH solution (1/8/2 in mole ratio) for 1 hour at 70 °C. The SF solution was then subjected to dialysis with a cellulose tubular membrane (250-7u; Sigma Aldrich; USA) 42,43 in distilled water for 3 days at room temperature. After filtration, the SF solution was lyophilised to obtain the regenerated SF sponges. The PLCL copolymer was prepared at a ratio of 50% PLLA to 50% PCL using previously reported methods 44 . For electrospinning, pure SF, SF/PLCL blends with different weight ratios (including SF:PLCL (3:1), SF:PLCL (1:1) and SF:PLCL (1:3)), and pure PLCL were dissolved in hexafluoroisopropanol (HFIP) solvents (Chinese Academy of Sciences, China) to obtain a final concentration of 10% (w/v), and the blending solutions were stirred at room temperature for 6 hours. All of the operations were conducted in a biological safety cabinet (Frontline FHC-1200A, USA). The solutions were then filled into a 2.5-ml plastic syringe with a blunt-ended needle. A syringe with an inner diameter of 0.21 mm was loaded in a syringe pump (789100C, Cole-Parmer, America) and dispensed at a speed of 1.2 ml/h. Using a high-voltage power supply (BGG6-358, BMEICO China), a voltage of 12 kV was applied across the needle and ground collector (a flat grounded steel plate covered with aluminium foil), which was placed at a distance of 12-15 cm, and the nanofibres were collected on the 13-mm glass cover slips on the plate. The resultant scaffolds were fumigated with 75% alcohol for 24 h. They were then rinsed three times with PBS and incubated in proliferation or differentiation medium for 1 hour before the cells were seeded on them. To perform the experiments in the proliferation medium, laminin was coated equally on the substrate of all the groups.

Characterisation of SF/PLCL nanofibres. The scaffolds were sprayed with Pt by a MP-19020NCTR
NeoCoater (JEOL Ltd., Tokyo), and the morphology of the resultant scaffolds was observed with a scanning electron microscope (SEM) (JSM-6701; JEOL, Tokyo, JAPAN) at voltage of 25 kV. The mean fibre diameters were estimated using image analysis software (Image-Pro Plus, Rockville, MD, USA) by selecting 100 fibres randomly observed on the SEM images.
Equilibrium swelling ratio measurements. Different weight ratios of known scaffold dry weights were immersed in proliferation medium and maintained at 37 °C for 12 h until a swelling equilibrium was reached. The wet scaffolds were then immediately weighed using a microbalance after the excess water lying on the surfaces was absorbed by a filter paper. The equilibrium swelling ratio (ESR) was calculated using the following equation 45 : where Ws is the weight of the scaffolds at the equilibrium swelling state and W d is the weight of the scaffolds at the dry state.
Mechanical measurements. The mechanical properties of pure SF, SF:PLCL (3:1), SF:PLCL (1:1), SF:PLCL (1:3) and pure PLCL were explored by applying a tensile test to all of the specimens prepared from the scaffolds. The mechanical properties were tested using a materials testing machine (H5K-S, Hounsfield, England) at an elongation speed of 10 mm/min. The temperature was controlled at 20 °C, and the relative humidity was maintained at 65%. Each scaffold weight ratio included five samples. The Young's modulus was calculated by measuring the slope of the initial linear region of the stress-strain curves.
Pore size measurements. The pore sizes were measured using a CFP-1100-AI capillary flow porometer (PMI Porous Materials Int.). The electrospinning nanofibrous membranes were cut into dimensions of 3 cm × 3 cm and were then infiltrated into the wetting liquid with a surface tension of 21st dynes/cm (PMI Porous Materials Int.). As the pressure was increased, the largest hole was the first to open, and the smaller holes then opened. As a result, the pore size distribution of the samples was obtained.
Water contact angle analysis. Water contact angle measurements were used to explore the surface wettabilities of the electrospun scaffolds. Distilled water was automatically dropped onto the electrospun scaffolds. The images of the droplets on the scaffolds after 10 s were visualised using an image analyser (OCA40, Data Physics, Germany). Furthermore, the angle changes over time were measured automatically. The contact angle was measured three times from different positions, and an average value was calculated.

Isolation, in vitro culture and induction of differentiation of retina progenitor cells (RPCs).
RPCs were separated from the fresh retinal tissue of postnatal-day-1 GFP-transgenic C57BL/6 mice (a gift from Dr Masaru Okabe, University of Osaka, Japan). The cells were cultured in T25 flasks with proliferation medium consisting of advanced DMEM/F12 (Invitrogen, Carlsbad, CA, USA), 1% N2 neural supplement (Invitrogen), 2 mM L-glutamine (Invitrogen), 100 U/ml penicillin-streptomycin (Invitrogen) and 20 ng/ml epidermal growth factor (recombinant human EGF, Invitrogen) 46 . Half of the proliferation Scientific RepoRts | 5:14326 | DOi: 10.1038/srep14326 medium was changed every 2 days, and the clones were passaged through mechanical isolation or enzymatic trypsinisation at regular 3-day intervals. For RPC differentiation, the cells were trypsinised and seeded at a density of 1 × 10 5 cells/ml with differentiation medium consisting of 10% foetal bovine serum (FBS) (Invitrogen) without EGF. Half of the culture medium was changed every 2 days. The cells were cultured at 37 °C and 5% CO 2 . All animals were handled according to ARVO animal usage standards, following approval by the animal care and use committee of the Schepens Eye Research Institute, where the original derivation of the cells was performed.
Morphology of RPCs on the SF/PLCL nanofibrous scaffolds. The morphology of RPCs on the SF/PLCL nanofibrous scaffolds was assessed using a fluorescent microscope (Olympus BX51, Japan) and SEM. The cells were fixed in 4% (w/v) paraformaldehyde (PFA) (Sigma-Aldrich, Poole, UK) in PBS for 15 min at room temperature and then washed three times with PBS. The nuclei were stained with Slow-Fade Gold with DAPI (Invitrogen, Carlsbad, CA, USA), and photographs of GFP + RPCs and DAPI-stained RPCs were taken using a fluorescent microscope. For SEM, the cell-seeded scaffolds were rinsed with PBS and fixed in 4% glutaraldehyde for 2 h at 4 °C. The samples were then dehydrated through a concentration gradient of 30%, 50%, 70%, 80% and 90% ethanol. After air-drying overnight, the samples were sprayed with Pt using a MP-19020NCTR NeoCoater and observed using SEM. Cell attachment test. The RPCs were cultured in 24-well culture plates for 12 hours; the medium was then replaced with PBS, and the cells were washed three times to remove the un-attached cells. The numbers of cells remaining on both the electrospun scaffolds and the control tissue culture plate surface were detected using the CCK8 method. Briefly, the cells were again incubated with 400 μ l of proliferation medium containing 20 μ l of CCK8 solution for another 4 hours. The medium was then pipetted into the wells of a 24-well plate and assessed using a microplate spectrophotometer, and the O.D. value at 450 nm for each well was measured. The nuclei were stained with Slow-Fade Gold with DAPI before and after three washes with PBS, and photographs were taken using a fluorescent microscope.

Cytotoxicity test.
To determine the acute cytotoxicity of our scaffolds on RPCs, the amount of the LDH released into the medium was measured using a CytoTox 96 Non-Radioactive Cytotoxicity Assay kit (Promega, USA) 48 . The RPCs were seeded onto scaffolds of different weight ratios for 24 hours. Following the LDH assay kit instructions, briefly, after the culture medium was discarded, 150 μ l of the LDH releasing reagent was added, and the plate was incubated for 1 h. Then, 120 μ l of liquid was pipetted out to a new 96-well plate, and the LDH testing reagent was added. The O.D. at 450 nm was measured using an ELISA microplate reader (ELX800, BioTeK, USA). The cytotoxicity (%) was expressed based on the value of LDH released into the medium divided by the total LDH activity.
Total RNA isolation and quality controls. The RPCs were quantified, and the same numbers of RPCs were seeded on the various SF/PLCL scaffolds for 3 days (proliferation condition) or 7 days (differentiation condition). The cell sheets were then treated with 1 ml of TRIzol reagent (Invitrogen) and stirred using a Micro Tissue Grinder (PRO Scientific Inc., Oxford, CT, USA) for 30 s. The total RNA from the cultured cells was extracted according to the manufacturer's instructions. The samples with OD260/280 nm ratios between 1.8 and 2.0 were used for cDNA synthesis.
Reverse transcription and quantitative polymerase chain reaction (qPCR). Using the PrimeScript ™ RT reagent kit (Perfect Real Time, TaKaRa, Dalian, China), 1000 ng of total RNA was reverse transcribed into cDNA in a final reaction volume of 20 μ l. The produced cDNAs were diluted 10-fold in nuclease-free water (Invitrogen) and used as templates for qPCR. qPCR was conducted in 10 μ l solutions containing 5 μ l of 2 × SYBR Premix EX Taq ™ (TaKaRa), 3 μ l of nuclease-free water (Invitrogen), 1 μ l of diluted cDNA, and 1 μ l of gene-specific primers (Table 1). qPCR was conducted using a 7500 Real-Time PCR Detection System (Applied Biosystems, Foster, CA). The efficiency of the PCR was measured using serial dilutions of the cDNA (1:1, 1:5, 1:25, 1:125, 1:625 and 1:3,125). Each sample was tested in triplicate. The relative mRNA expressions were analysed using the Pfaffl method 49 . Data are expressed as fold change relative to untreated controls, after normalisation to the expression of β -actin.