The SAP scaffolds, RADA16 (AcN-RADARADARADARADA-CONH2), were custom-synthesized (Shanghai Bootech Bioscience and Technology Corp., Ltd., Shanghai, China) through direct solid phase synthesis, purified by high-performance liquid chromatography (HPLC) (Thermo Electron Corp., Waltham, MA, USA), and characterized by mass spectroscopy (Waters Corp. Milford, MA, USA). The purity of D-RADA16 and L-RADA16 was 95.90 and 95.54%, respectively. The peptide D-RADA16 sequence contained all D-amino acids, and the L-RADA16 sequence contained all L-amino acids. Solutions of the peptides were prepared at concentrations of 1.0–10.0 mg/ml (0.1–1.0%, w/v) in water (18 MΩ•cm; Millipore Milli-Q System, Billerica, MA, USA) and stored at 4°C before use.

Dulbecco's modified Eagle's medium/F12 (DMEM/F12), fetal bovine serum (FBS) and 0.25% trypsin/EDTA were purchased from HyClone (Logan, UT, USA). Penicillin/streptomycin aqueous solution, MTT assay, β-glycerophosphate, dexamethasone, ascorbic acid and proteinase K were all obtained from Sigma Chemical Co. (St. Louis, MO, USA). The antibodies used for western blot analysis were all purchased from Abcam PLC (Cambridge, UK). Calcein-AM was purchased from Enzo Life Sciences, Inc. (Farmingdale, NY, USA).

Molecular models and chemical structures of the chiral peptides were constructed using free modeling software (Hyperchem professional version 7.5, http://www.hyper.com) and chemical structure drawing software (ChemDraw Ultra 14.0) respectively.

The samples consisted of 1.0 mg/ml aqueous stock solutions of peptide and were adjusted to 0.17 mg/ml in solution in water. A sample of 400 µl was added in a CD cuvette with a 2 mm path-length. Measurements were carried out on an J-810 CD spectrometer (JASCO International Co., Tokyo, Japan). The samples were incubated at 20°C, equilibrated for 30 sec, measured from 190 to 260 nm, averaged over 3 sec through the entire wavelength range, and took the 190–250 nm analysis.

TEM samples (2.0 mg/ml) were prepared at 25°C. A micropipet was used to load an aliquot of 5 µl of peptide solution to a carbon coated copper grid. The excess solution was removed by a piece of filter paper. The samples were dyed by 10 µl uranyl acetate for 30 sec and dried overnight in a desiccator and then conducted on a Tecnai G2 F20 system (FEI Company, Hillsboro, OR, USA) operating at 200 kV.

A total of 18 male Sprague-Dawley rats, 4 weeks of age, were used in this study. The rats were obtained from the Laboratory Animal Center of Chongqing Medical University, Chongqing, China. BMSCs were isolated from the bone shaft of the rats according to the technique reported by Lennon et al (33) and the protocol was approved by the Ethics Committee of the First Affiliated Hospital of Chongqing Medical University (permit no. 2014-201058). Briefly, the rats were sacrificed by an overdose of isoflurane. The bone marrow was flushed out from the femurs by a syringe (21-gauge needle) with 5 ml of DMEM/F12 containing 10% FBS and 1% penicillin/streptomycin (200 U/ml). The cell suspension was placed into two T-25 flasks (Nest Biotechnology Co., Jiangsu, China) and cultured at 37°C in an atmosphere with 95% humidity and 5% CO₂. The medium was changed on the second day of culture and every 3 days thereafter. When the cells became subconfluent, they were detached from the flask by treatment with an aqueous solution of 0.25% trypsin/EDTA for 3 min at 37°C. The cells were normally passaged at a density of 2×10⁴ cells/cm². Cells at the third passage at subconfluence were used in all the experiments.

In the case of cell viability assay, the chiral scaffolds at various concentrations (1.25, 2.5, 5.0 and 10.0 mg/ml) were prepared as L-RADA16 and D-RADA16. Each of the solution was sonicated for 30 min and loaded (5 µl) in the bottom of 96-well culture plates (Nest Biotechnology Co.). Subsequently, 50 µl medium were slowly added to the solution to induce gelation. The hydrogel formed in several minutes, and was rinsed twice with medium to equilibrate the gel to physiological pH. The hydrogels were thyen incubated overnight at 37°C with 5% CO₂ until cell seeding. Lastly, 50 µl of BMSCs cells (4×10⁴ cells/ml) in DMEM/F12 mixture were seeded in each well on top of the various hydrogels in 96-well culture plates, where they were able to settle into the nanofiber scaffolds. The 3D cell cultures were maintained in an incubator at 37°C with 5% CO₂ for 3, 5 or 7 days and the medium was changed every 2–3 days if necessary.

In the case of differentiation assay, 500 ml of peptide solution (5 mg/ml) were directly loaded into a 6-well culture plate (Nest Biotechnology Co.) and then stimulated to self-assemble following the addition of 1 ml culture medium in each well. During the following 30 min, the medium was changed twice to equilibrate the growth environment to physiological pH and then incubated for 2 h at 37°C with 5% CO₂ for gelation. Subsequently, 200 µl of BMSCs (1×10⁶ cells/ml) in DMEM/F12 mixture were seeded on top of the hydrogel. The cells were cultured in the differentiation media consisting of 10 mM β-glycerophosphate, 10 nM dexamethasone, 50 µg/ml ascorbic acid. The media were changed every 3 days.

In the case of the migration assay, cell culture Transwell inserts (Corning Inc., Corning, NY, USA) were used for peptide hydrogelation as previously reported (20). Briefly, the inserts were placed in a 24-well culture plate (Nest Biotechnology Co.) with 400 µl culture medium in the lower chambers of each well. One hundred microliters of peptide solutions solution (5 mg/ml) were directly loaded into the upper chamber of each well. The medium was changed twice and incubated as described in the differentiation assay. To prevent the drying of the surfaces of the formed hydrogels, 400 ml of culture medium were gently layered onto the hydrogels and then incubated overnight at 37°C with 5% CO₂. Two hundred microliters of BMSCs (1×10⁵ cells/ml) in DMEM/F12 mixture were seeded on top of the peptide gel in the inserts. The cells were cultured in the maintenance medium (DMEM/F12 containing 10% FBS and 1% penicillin/streptomycin) and this was changed every 3 days by removing 400 µl of medium from the lower chambers and adding 400 µl of fresh medium inside the upper chambers. The cells were harvested at planned time points for analysis.

In addition, a conventional 2D cell culture method (tissue culture plate) was used to culture the BMSCs to examine their proliferation and migration as a control.

To determine the extent of the proliferation of the cells seeded on the chiral scaffolds, a well-characterized quantitative assat, 3-(4,5-dimethylthiazol-2yl)-2,5-diphenyl tetrazolium bromide test (MTT) assay was performed. Briefly, after the cells were incubated for 3, 5 or 7 days, the medium was removed and the cells were treated with 20 µl of 5 mg/ml MTT solution. Follwoing incubation at 37°C for 4 h, the MTT solution was removed and 150 µl of dimethyl sulfoxide (DMSO) were added to dissolve the insoluble formanzan crystals. After 20 min, the absorbance was read at 490 nm using a Multiskan Spectrum spectrophotometer (Thermo Fisher Scientific, Vantaa, Finland). Since some hydrogels may dissolve partially when removing the formanzans, we tested the hydrogels without cells to confirm any possible bias in the absorbance measurements, and found no significant differences between any of the tested scaffolds and the scaffolds without cells (data not shown). In order eliminate dye absorbance derived from medium, a background group was added to the analysis groups. Three independent experiments comprising 4 replicates each were performed.

At the end of the culture period, the cells were lysed for 30 min in ice-cold RIPA buffer (BCA; Beyotime, Jiangsu, China). The cell lysate was centrifuged at 12,000 rpm for 20 min at 4°C. The supernatants were heated at 100°C for 5 min in 5× loading buffer (Beyotime). Equal aliquots of protein (40 µg) were loaded onto 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto PVDF membranes (0.45 µm). The membranes were blocked for 1 h at room temperature by 5% non-fat milk, washed 3 times and incubated with primary antibodies [anti-osteopontin (OPN; ab8448, 1:1,000), anti-runt-related transcription factor 2 (RUNX2) (ab23981, 1:1,000) and anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (ab8245, 1:2,000) (all from Abcam PLC)] at 4°C overnight. The membranes were then washed in TBST 3 times and incubated with corresponding horseradish peroxidase conjugated secondary antibody (goat anti-rabbit) for 1 h at room temperature. The blots were developed by a chemiluminescence kit (Beyotime) on a Bio-Rad imaging system.

Cell culture was performed as described above. After being cultured for 7 days, the cells on the hydrogels were examined using calcein-AM staining, an indicator of intracellular esterase activity which stains whole living cells, according to the manufacturer's instructions. Briefly, cells on the chiral scaffolds of the inserts were washed twice using phosphate-buffered saline (PBS). Subsequently, 4 µM calcein-AM solution were added to the insert and incubated for 30 min within the incubator at 37°C with 5% CO₂. The inserts were rinsed twice with PBS and multiple images were obtained with confocal laser scanning microscopy (Nikon A1R; Nikon, Tokyo, Japan) through z-stack scanning mode with a step size of 3 µm. The fiber diameters were examined using Image-Pro Plus software (version 6.0) analysis software. For each image, 200 fibers were measured and recorded.

We dissolved 1 mg of peptide in 5 ml 0.01 M PBS, and 0.25 mg proteinase K was added to the solution, resulting in a 0.2 and 0.05 mg/ml concentration of peptide and proteinase K, respectively. The samples, consisting of hydrogels and proteinase K, were incubated at 37°C on shaking tables for 1–24 h and 400 µl aliquots were taken after 1, 2, 4, 6, 8, 12 and 24 h. The enzymatic reaction was terminated by the addition of phenylmethanesulfonylfluoride (PMSF) and the concentration of the intact peptide was analyzed by HPLC (LCMS-2020; Shimadzu, Kyoto, Japan) using following conditions: column: Venusil XBP C18, 5 µm, 4.6×50 mm (Agela, Tianjin, China); sample injection volume: 100 µl; gradient elution: A, 0.035% trifluoroacetic acid (TFA) in water; and B, 0.035% TFA in acetonitrile/water (80:20, vol/vol). A gradient of 10–100% B in 10 min was conducted. The experiment was conducted in 3 replicates.

Data are presented as the means ± standard deviation (SD) and compared between any 2 groups using a two-tailed paired t-test. The differences between multiple group comparisons were made by one-way ANOVA and followed by multiple pairwise comparisons using Fisher's least significant difference (LSD) test. Significance levels were set to P<0.05 for all comparisons.

We observed the formation of hydrogelation after 30 min. Both of the peptide solutions were obtained by dissolving the peptide powder in water at the concentration of 10 mg/ml. As shown in Fig. 1A, we found that the introduction of D-amino acids did not effect on the gelation behavior of the SAP solution.

We present two molecular models (Fig. 2A and B) and chemical structures (Fig. 2C) of L-RADA16 and D-RADA16 peptides. These peptides have an identical sequence, but are composed of amino acids of a different chiral form: all L-amino acids in L-RADA16 and all D-amino acids in D-RADA16. The pair of RADA16 structures appears similar, but some of their properties are quite diverse.

The secondary structures of the chiral RADA16 peptides were examined by CD measurements. From a previous study, L-RADA16 is known to form a stable β-sheet structure (21). In this study, CD spectroscopy revealed that the D-RADA16 was not only able to adopt a typical β-sheet structure, but also had an inverted β-sheet spectrum with a positive peak at 216.8 nm and a negative peak at 193.1 nm (Fig. 3), which was almost a mirror image of the L-RADA16 spectrum which had an β-sheet spectrum with a positive peak at 195.7 nm and a negative peak at 216.2 nm. This suggests that D-RADA16 is the enantiomer of L-RADA16.

A previous study demonstrated that the peptide L-RADA16 possesses the ability of self-assembly into interwoven nanofibers (21). In the present study, we wished to determine whether D-RADA16 made of D-amino acids would trigger self-assembly process and form well-ordered nanofibers. TEM morphological analyses denoted that D-RADA16 indeed formed ordered nanofibers ranging in length from several hundred nanometers to a few microns, and the diameters of nanofibers assembled from L-RADA16 and D-RADA16 were 16.34±6.13 and 13.52±2.94 nm, respectively (Fig. 4). These observations are in agreement with those of previous findings (34,35).

BMSCs were grown on the chiral scaffolds at various concentrations from 1.25 to 10 mg/ml. The cell proliferation rate of the 3D cell culture method and the conventional 2D cell culture method as a control were performed for 3 days (Fig. 5A). We noted that there were no significant difference in the proliferation rates between the chiral scaffolds at each concentration after 3 days of culture. The peptides at low concentrations of 1.25 or 2.5 mg/ml did not lead to a statistically significant difference in the cell proliferation rate in comparison to the control. By contrast, the peptides at high concentrations of 5 or 10 mg/ml led to a statistically significant difference in the cell proliferation rate in comparison to the control. These results indicated that the chiral peptide scaffolds at relatively high concentrations promoted cell proliferation. However, there was no statistically significant difference in the cell proliferation rate when the peptide at 5 mg/ml was compared to a higher concentration of the peptide at 10 mg/ml. Therefore, the optimal concentration of the chiral peptides for cell culture was 5 mg/ml, and this concentration was selected for use in further experiments.

Subsequently, BMSCs were grown on the chiral scaffolds at 5 mg/ml and the tissue culure plate for 5 and 7 days (Fig. 5B). Again, the chiral scaffolds possessed similar cell-scaffold bioactivity at each time point. Furthermore, significant differences in the chiral scaffolds compared to the control were noted after 5 days of culture. Nonetheless, there were no significant differences in the chiral scaffolds compared to the control after 7 days of culture, and this can be explained by the fact that cells stop growing when they reach confluence. Since L-RADA16 has been proven to be non-toxic and non-immunogenic (36,37), we hypothesized that D-RADA16 displayed no apparent toxicity in vitro under our present experimental conditions.

According to the protocol of MTT assay, after the cells were incubated for a given period of time, MTT solution was added to each sample and MTT was reduced by metabolically active cells to insoluble purple formazan dye crystals. We serendipitously observed the crystals under an inverted phase contrast microscope (Fig. 6). Of note, the seeded cell planes were out of focus, overlapping the focused plane, resulting in relatively fuzzy images when they were grown on D-RADA16 scaffolds at 5 and 10 mg/ml. This fact suggests that the formazan exhibits various 3D morphologies at relatively high concentrations of the D-RADA16 scaffold. By contrast, clear images can be captured when the cells were cultured in the concentrations of 0.125 and 2.5 mg/ml, and the control, denoting that formazan retained 2D morphologies in the control and at low concentrations of the D-RADA16 scaffolds.

The BMSCs were cultured in the SAP hydrogels to evaluate the osteogenic differentiation level at day 7. As a control, the BMSCs were cultured with the conventional 2D cell culture method. The relative expression level of RUNX2, osteopontin (OPN) was examined by western blot analysis. GAPDH was used as an internal control (n=3). For all proteins, the two 3D scaffold groups possessed a significantly lower expression than the 2D culture control group (Fig. 7). The results indicated that the chiral SAP scaffolds did not promote the osteogenic differentiation of the BMSCs in vitro under our present experimental conditions.

In addition to the cell viability assay, the BMSCs were seeded in the chiral peptide hydrogels (Fig. 8A–D) and the tissue culture plate (Fig. 8E and F) to examine their 3D migration using calcein-AM staining. The confocal microscopy 3D reconstruction images revealed that cells migrated into the L-RADA16 (Fig. 8A and B) and D-RADA16 (Fig. 8C and D) scaffolds at ~285 and ~186 µm, respectively. The results demonstrated that both of the two chiral peptides promoted cell migration into the 3D nanofiber scaffolds. They are consistent with those of a previous study on human adipose stem cells that migrated into 3D peptide hydrogel scaffolds (20).

In addition, the chiral peptides were also examined for enzymatic stability against proteinase K, a potent and highly non-specific proteolytic enzyme. To avoid the effects of resistance due to hydrogelation, the concentration of each SAP was set at 0.2 mg/ml, which is markedly lower than the critical gelation concentration of the hydrogel. As shown in Fig. 9, ~75 and 60% of D-RADA16 remained after 2 and 6 h of incubation with proteinase K respectively, and >50% remained after 24 h. As for L-RADA16, <25 and 8% of SAP remained after 2 and 6 h of incubation respectively. These results illustrate that D-RADA16 exhibits stronger resistance to proteinase K digestion.