Strengthening 3D Printed Gelatin Scaffolds with Thread Support for Soft Tissue Repair

The SEM image shows that several micro-sized suture fibers are well contained in the gelatin biomaterial ink (Figure 1A–D).

The rheological properties of a biomaterial ink, such as viscosity and gel strength, are very important factors in determining the printability and printing accuracy of the biomaterial ink [37]. In terms of rheological properties, gelatin biomaterial inks with low viscosity are not suitable for 3D printing. Therefore, we used the thermo-responsive behavior of gelatin to increase their viscosity. As previously shown, optimal rheological properties for gelatin biomaterial inks were evaluated [23]. Herein, it was necessary to confirm the effect of suture fibers on the thermo-responsive behavior of gelatin. A rheometer was used to determine the viscosity, gel point, and complex shear modulus of the suture fibers added gelatin biomaterial ink. As shown in Figure 2A, all the biomaterial inks showed obvious shear thinning behavior, which is an essential property for hydrogel-based biomaterial ink printing [38]. The flow curve of the biomaterial ink varied depending on the suture fiber content. The biomaterial ink viscosity increased from 0.01 Pa·s to 33.3 Pa·s as the suture fiber content was increased from 0% to 0.5% (at a 1/s shear rate). Notably, the viscosity of the biomaterial ink with 0.5% suture fiber content was about 3100 times higher than that of the biomaterial ink without suture fibers. This difference was due to the added suture fibers having absorbed moisture from the gelatin aqueous solution and the suture fibers having physically interacted with each other. When the suture fiber content was more than 0.5%, the fibers did not disperse homogeneously in the gelatin aqueous solution and instead became entangled. We were thus unable to implement a further increase in the suture fiber content to achieve a higher viscosity. In order to increase the viscosity of the suture fibers added gelatin biomaterial ink, we instead used the thermo-responsive behavior of the gelatin. As shown in Figure 2B, significant changes in biomaterial ink viscosity occurred when changing the temperature. The viscosity of the gelatin biomaterial ink increased rapidly as temperature was cooled to below 20 °C. Below 20 °C, the gelatin generated phase transition from spread chain to triple helix structure and this transition was reversible [39]. Adding suture fibers did not result in any significant change in the sol-gel transition temperature (18~20 °C). This result indicated that the sol-gel transition temperature of the gelatin biomaterial ink into which suture fibers were added was highly dependent on the concentration of the gelatin [23] but not dependent on the content of the added suture fibers. Based on these data, the temperature of the gelatin biomaterial ink reservoir should be kept at 20 °C during the 3D printing process. In the gel state biomaterial ink, the gel strength is an important rheological property in determining printability. As shown in Figure 2C, the complex shear modulus increased with increasing suture fiber content. The complex shear modulus of the 0, 0.1, 0.3, and 0.5% biomaterial inks were 77.4 ± 12.9, 146.4 ± 13.6, 459.7 ± 40.0, and 1789.5 ± 163.3 Pa, respectively. These results showed that the addition of suture fibers did not affect the sol-gel transition temperature but did affect the gel strength of the gelatin biomaterial ink. These results, in particular those showing an improvement in biomaterial ink rheological properties with the addition of micro/nano fibers, were consistent with those of several studies [34,36].

G. Gillispie et al. noted that the word ‘printability’ was recently adopted in the 3D printing field to describe the capabilities of materials used in various 3D printing modalities [40]. Many conditions, such as biomaterial ink rheological properties, printing speed, pneumatic pressure, and temperature, are involved in determining printability. We have already determined certain rheological properties (sol-gel transition temperatures, gel strengths) of the gelatin biomaterial inks into which suture fibers were added (in Figure 2C), so we had to find a suitable pneumatic pressure and printing speed for each biomaterial ink. Figure 3A shows the extruded strand state for each of the various pneumatic pressures and suture fiber contents. When the pneumatic pressure was too low, the biomaterial ink was not able to come out through the nozzle (Δ: unextrudable in the figure). On the other hand, using too high of a pneumatic pressure yielded irregular strands (X: irregular). We found a suitable pneumatic pressure between 40 and 80 kPa for each biomaterial ink formulation made in the current work. As the suture fiber content in the biomaterial ink was increased, the ideal pneumatic pressure to print also increased. Note that these results were consistent with the rheological properties of the biomaterial ink (Figure 2C). As shown in Figure 3B, when the printing speed was less than 8 mm/s, irregular strands were produced and the printing time was too long (>10 min). On the other hand, when the printing speed was more than 10 mm/s, it was impossible to fabricate a 3D structure due to the break of the strands. Therefore, the appropriate printing speed, applicable to all four biomaterial inks, was determined to be 9 mm/s. Figure 3C shows the printed strand widths of the different biomaterial inks at their respective optimized pneumatic pressures. It was found that suture fibers added biomaterial ink (0.1% and 0.3%) could obviously reduce the printed strand width compared with the biomaterial ink without suture fibers. When suture fiber content was increased to 0.5%, the strand width was similar to that of the biomaterial ink without suture fibers (1.69 ± 0.29 mm). This is the second reason why the suture fiber content cannot be increased further. The higher the suture fiber content, the higher the pneumatic pressure required for 3D printing (Figure 3A). Increasing the pneumatic pressure causes more biomaterial ink to come out through the nozzle, which can increase the strand width. Therefore, a suture fiber content of 0.5% was the highest content that we were able to practically add in the gelatin biomaterial ink.

One of the greatest advantages of 3D printing technology is that we can fabricate sophisticated 3D scaffolds as designed. These advantages enable the fabrication of patient-specific scaffolds and desired disease models. With regards to printing accuracy of the biomaterial ink, synthetic polymer biomaterial inks (PEG, PVA, PLA, PLGA, and PCL) are superior to natural biomaterial inks [41,42]. However, synthetic polymer biomaterial inks are less biocompatible than natural polymer biomaterial inks, and the acidic by-products of their degradation can be toxic to surrounding cells [43]. In order to solve these problems, studies are needed to improve the printing accuracy of natural polymer biomaterial inks. Figure 4A,B shows a comparison of the sizes of the CAD model and the 3D printed scaffold. Observation of the printing accuracies for various contents of added suture fibers indicated a dramatic difference in the height of the 3D printed scaffold. Measurements of the heights of the 3D printed scaffolds showed that the use of a suture fiber content of 0.5% yielded a 3D printed scaffold with a printing accuracy of 97.1%, considerably higher than the 84.4% printing accuracy when without suture fibers. These results can be explained in terms of the modulus of the biomaterial inks (Figure 2C). When the modulus of the biomaterial ink was high enough for printing, the biomaterial ink maintained the round-type strand shape extending in a straight line when coming out through the nozzle. When the modulus of the biomaterial ink was not sufficiently high, the printed biomaterial ink could not maintain this shape and instead collapsed, with this collapse reducing the height of the 3D scaffold and consequently impairing the printing accuracy. These results show that the best printing accuracy and best printability were achieved by the printed biomaterial ink including a 0.5% suture fiber content, which also showed a dramatically increased modulus.

Another major advantage of 3D printing is the ability to use this technique to adjust the 3D scaffold pore sizes as desired. The presence of pores is very important for the human organ scale 3D scaffolds. A high interconnectivity of pores in the 3D scaffold can facilitate the supply of nutrients and oxygen to the cells in the 3D scaffold and increase the surface area of the 3D scaffold [44]. To achieve such interconnectivity, the pores at the surface area and those in the cross-sectional area must be well maintained. Figure 4C–N shows the surface and cross-section morphologies of the 3D printed scaffolds. The surfaces of the 3D printed scaffold strands were very rough and filled with micropores. Hydrogels with micropores and rough surfaces have already been reported to be better than hydrogels with smooth surfaces for tissue engineering [45]. The change in the 3D scaffold pore shape is also an important factor in 3D printing, so we evaluated the printability (Pr) value of each biomaterial ink. If the value of Pr is greater than 1, the shape of the pores is uneven and the uniformity of the strand is not good. Conversely, if value of Pr is lower than 1, the size of the pores is reduced compared to the CAD design and the size of the strands is increased. In Figure 4C–F, the Pr value was closest to 1 at 0.3% and 0.5%, which is consistent with the strand width (Figure 3C). The current 3D printed scaffolds were observed to have macropores, specifically with dimensions of 500–700 um. These macropores observed on the surfaces and in the cross-sections of the 3D printed scaffold could thus make it possible to supply nutrients and oxygen to the inside of organ scale 3D scaffolds.

The mechanical strength of a 3D scaffold directly affects cell fate and tissue regeneration [46]. As reported by several groups, the modulus of native soft tissues and organs ranges from 0.1 kPa to 1000 kPa [25,47]. These reports showed that scaffolds for tissue engineering must have mechanical strength similar to those of the target tissues or organs in order to secure dimensional stability when exposed to surrounding human tissue movement and external stimuli. As shown in Figure 5, the Young’s modulus of the pure gelatin 3D printed scaffold was measured to be 24.0 ± 9.0 kPa. This low mechanical strength would limit the applicability of pure gelatin 3D printed scaffolds to only weak soft tissues such as brain, breast, and lung tissues. However, improving the mechanical strength of a gelatin hydrogel would allow it to be applied to muscle, skin, and tendon regeneration, especially since gelatin hydrogels already display an intrinsically high ability to be elongated. As shown in Figure 5A, the tensile stress of the 3D printed scaffold was markedly improved by increasing its suture fiber content. While increasing this suture fiber content was observed to be associated with a decrease in the elongation ratio, this effect was very slight. This result was probably due to the occurrence, during the 3D printing process, of a shear-induced alignment of the suture fibers as seen in the confocal microscopy images (Figure 7C–F). The significant differences between the tensile stress of the 3D printed scaffolds were also reflected in the considerable differences between their Young’s modulus. As shown in Figure 5B, the Young’s modulus of the 3D printed scaffold including a suture fiber content of 0.5% was, at 147.5 ± 14.0 kPa, about 6-fold higher than the 24.0 ± 9.0 kPa Young’s modulus of the 3D printed scaffold without suture fibers. As expected, the added suture fibers played an important role in improving the mechanical strength of the gelatin 3D scaffold.

The high water absorption capacity along with the excellent biocompatibility of natural polymers is one of the important reasons that natural polymers are suitable for use as biomaterial inks. Water-containing hydrophilic polymer materials are called hydrogels. Such hydrogels have cell-friendly properties such as facilitating rapid diffusion of nutrients, oxygen, and water-soluble compounds, providing protection from external stimuli, and improving cell adhesion and migration [19]. Therefore, water absorption is an essential property of the 3D scaffolds. In Figure 6A, the swelling ratio curves were acquired for the 3D printed scaffolds, in order to determine their ability to absorb water. For each 3D printed scaffold, the swelling ratio increased during the first 2 h to a value of approximately 2000% and plateaued stably at this level for the next 2 days of the swelling test. However, the swelling ratio of the 3D printed scaffold containing 0.5% suture fiber was, at 1969% at 24 h, slightly lower than those of the other scaffolds. This somewhat decreased ratio was due to its increased suture fiber content having resulted in a higher-density hydrogel structure, as can be inferred by its increased mechanical strength (Figure 5). Although there was a slight decrease in the swelling ratio due to the increase in the suture fiber content, it still exhibited a high swelling rate close to 2000%. The suture fiber reinforced 3D scaffold can deliver nutrients to culturing cells by rapidly absorbing liquid such as medium while maintaining their original structure.

3D scaffold implanted in vivo must be stable for enough time to provide an optimal environment for cell adhesion and proliferation [48]. Moreover, an overly rapid degradation of a scaffold can trigger an overly strong immune response, causing side effects at the transplant site [49]. Therefore, gelatin hydrogels have been subjected to various crosslinking methods to find ways to stop them from degrading rapidly during in vivo applications [28,29,30]. Physical crosslinking methods such as dried heat treatment (DHT) are more suitable for pure gelatin 3D scaffolds than chemical crosslinking methods [23]. However, suture fiber reinforced 3D scaffolds cannot be effectively crosslinked using the DHT method. DHT needs to be conducted at a high temperature (110 °C) and in a vacuum condition in which suture fibers denature and are unable to maintain their strength [50]. Accordingly, in the current work, suture fiber reinforced 3D scaffolds were crosslinked using a chemical reagent (EDC/NHS), and the rates of enzymatic degradation of these crosslinked scaffolds were confirmed by collagenase (2 U/mL). The rate of degradation of each 3D printed scaffold was determined by measuring the remaining mass of the scaffold as a function of time and plotting, as shown in Figure 6B, the remaining masses each as a percent of the original mass. At 5 days of the collagenase reaction, the remaining mass as a percentage of the original mass gradually increased with increasing suture fiber content. This remaining mass percentage for the 3D printed scaffold containing 0.5% suture fiber was approximately 48%, considerably higher than the approximately 29% remaining mass for the 3D printed scaffold without suture fibers. These results imply that supplementing suture fibers into a 3D printed scaffold could markedly improve the stability of the scaffold under physiological conditions.

The biocompatibility of the 3D printed scaffolds was assessed by culturing HDFs on a 3D printed scaffold. The cell proliferation rate was evaluated based on an DNA assay. The morphologies and spreading of the cells were confirmed using confocal microscopy images. As shown in Figure 7A, all experimental groups supported steady cell growth for the duration of the cell culture. There was no significant difference in cell numbers between the experimental groups on day 1 and day 3. However, at 7 days and 14 days of cell culture, the cell growth rate tended to be higher for the cell cultured 3D scaffolds with a higher content of suture fibers. The number of HDFs increased rapidly from day 7 to day 14 of cell culture. At day 14, the DNA assay showed more cell proliferation having occurred on the cell cultured 3D scaffolds containing a higher suture fiber content. The most cells (1,331,707 ± 48,788 cells/scaffold) were observed for the cell cultured 3D scaffold containing 0.5% suture fiber. As shown in Figure 7B–I, the morphologies of the HDFs grown on the cell cultured 3D scaffold were visualized by performing confocal microscopy on samples in which the HDFs nuclei (DAPI, blue) and actin filaments (phalloidin, red) were stained. The confocal microscopy images taken at 14 days of cell culture showed that the HDFs were evenly distributed over the surfaces of the 3D printed scaffolds. Also, the number of DAPI-stained HDFs present on the 3D printed scaffold containing 0.5% suture fiber was higher than those on the other experimental groups. These results, which are consistent with the results of the DNA assays shown in Figure 7A, may have been due to the obvious interconnected and hierarchically structured pores and good dimensional stability of the 3D printed scaffold containing 0.5% suture fiber (as shown in Figure 4). Such structural and dimensional stability features of 3D printed scaffolds play an important role in cellular behaviors such as cell proliferation, migration, infiltration, and morphogenesis because they affect the cellular microenvironment [51]—but these features may also be influenced by cell-mediated contraction.

Contractile forces generated by cells are one of the important factors that impair the dimensional stability of a scaffold. B. A. Harley et al. reported a contractile force of about 11~41 nN generated by dermal fibroblast in a collagen-glycosaminoglycan (GAG) scaffold [52]. M. Eastwood et al. reported that a 0.1 nN/cell contractile force occurred in a collagen hydrogel [53]. The cell-mediated contractile force affects the cellular microenvironment by changing the size and microstructure of the 3D scaffold. These changes in the size of the 3D scaffold cause the implanted 3D scaffold to separate from the surrounding natural tissue. In addition, changes in 3D scaffold microstructure features such as pore size and porosity caused by cell-mediated contraction affect cell migration, infiltration, and nutrient exchange. Therefore, it is important to maintain the dimensional stability of the 3D scaffold by minimizing cell-mediated contraction. As shown in the digital images in Figure 8A, at day 14 of cell culture, the sizes of the cell cultured 3D scaffolds containing a low content of suture fibers were smaller than those of the scaffolds with a high suture fiber content. As shown in Figure 8B, the areas and thicknesses of the cell cultured 3D scaffolds were measured at several time points, and the contraction of each cell cultured 3D scaffold was indicated by showing the decrease in the cell cultured 3D scaffold area (%) over 14 days. At 14 days of cell culture, the area of the cell cultured 3D scaffold containing 0.5% suture fiber was 97.8% of the initial area, quite a bit higher than the 83.8% of the initial area maintained by the cell cultured 3D scaffold without suture fibers. As shown in Figure 8C–N, this difference was also confirmed upon inspection of the corresponding SEM images. It was confirmed that numerous pores present on the strand surfaces of all the cell cultured 3D scaffolds were not visible by cells and ECM at 14 days of cell culture. As shown in Figure 8C–J, the strand morphology was changed by the proliferated cells. In the cell cultured 3D scaffolds containing 0% and 0.1% suture fiber, the strand size was reduced, the strand uniformity decreased, and, as shown in Figure 8K,L, the pores collapsed with regards to their cross-section areas; on the other hand, the cell cultured 3D scaffolds containing 0.3% and 0.5% suture fiber had uniform strands and maintained the pore structure in the cross-section images of Figure 8E,F,I,J,M,N. To reduce cell-mediated contraction, several studies have adjusted the gel strength by increasing the degree of crosslinking or the concentration of matrix substances or introduced GAG. [54,55]. We have now also provided compelling evidence that introducing suture fibers could contribute to the dimensional stability of 3D scaffolds and increase the mechanical strength of these scaffolds so that they can resist cell-mediated contraction.

The Young’s modulus of the 3D printed scaffold containing 0.5% suture fiber, while 6 times the modulus of the 3D printed scaffold without suture fibers (Figure 5), was still not as high as the Young’s modulus of human skin, and hence this scaffold cannot be used to replace human skin. Nevertheless, a higher modulus can be obtained by carrying out cell proliferation and an ECM remodeling process. D. Kang et al. mixed HepG2/C3A/endothelial cells with collagen/alginate bioink and printed them in order to mimic the hepatic lobule structure [56]. They showed that the compressive strength of the printed hepatic lobe was about 2 times higher on day 5 than on day 1. The mechanical strength of the suture fiber reinforced 3D scaffolds used in this study also increased with cell culture time. As shown in Figure 9A,B, the compressive stress of each cell cultured 3D scaffold was higher on day 14 of the cell culture than on day 1. Moreover, as shown in Figure 9C, the Young’s modulus of each cell cultured 3D scaffold on day 14 was more than twice the Young’s modulus on day 1. The Young’s modulus of the cell cultured 3D scaffold containing 0.5% suture fiber was 2.496 ± 0.019 MPa at 14 days of cell culture. The improvements in the mechanical strengths of the 3D printed scaffolds with cell culture time were probably due to the proliferation of cells and ECM secreted from cells (Figure 8C–N). This increase in mechanical strength and resistance to cell-mediated contraction indicated that a suture fiber reinforced 3D scaffold with high dimensional stability and mechanical strength can stably maintain its structure for a sufficient amount of time in a physiological environment.

Gelatin biomaterial inks containing suture fibers were prepared by adding suture fibers into a gelatin aqueous solution. Briefly, the suture bundles (Vicryl, Ethicon, Somerville, NJ, USA) were cut with surgical blades into filaments each having a length of 3 mm. By using a mortar, the bundles were formed into suture fibers. Gelatin (Bloom 225, type A, MP Biomedicals, Solon, OH, USA) was mixed with distilled water (DIW) at 70 °C for 3 h to form a solution with a final gelatin concentration of 3.5% (w/v) [23]. The suture fibers and 3.5% gelatin aqueous solution were then mixed at room temperature. Various biomaterial ink formulations containing suture fibers at content of 0% (w/v), 0.1% (w/v), 0.3% (w/v), and 0.5% (w/v), respectively, were prepared. Inclusion of suture fibers in the biomaterial inks was confirmed using scanning electron microscopy (SEM) (CX-200TM; COXEM, Daejeon, Korea). Briefly, the gelatin biomaterial inks containing suture fibers were transferred to a −80 °C deep freezer (MDF-U74V, SANYO, Osaka, Japan) and frozen overnight. The frozen samples were dried by freeze-drying for 3 days. The dried samples were coated with gold using an ion sputter coater (SPT-20; COXEM, Daejeon, Korea) and characterized using SEM at 20 kV.

The rheological properties of the suture fibers added gelatin biomaterial inks were confirmed using an AR 2000ex rheometer (TA instruments, New Castle, DE, USA) with a cone–plate geometry (diameter of 40 mm, angle of 1°). All samples were heated to 37 °C before being used in the experiment. The viscosities of the samples were measured as a function of shear rate, with the shear rate increased from 0.1 to 10/s at 37 °C. Viscosity measurements were also taken at various temperatures, starting at 40 °C and then decreasing at a cooling rate of 1.0 °C/min to 10 °C. Here, the oscillations were applied at a frequency of 1 Hz, and the shear rate was 0.1/s. The storage modulus (G′) and loss modulus (G″) of each biomaterial ink was also determined, by placing the biomaterial ink on the rheometer plate and lowering the temperature from 37 °C to 20 °C in 30 min. Then, frequency sweep experiments were conducted at a fixed strain of 1% from 0.1 to 10 Hz at 20 °C. The complex shear modulus (G*) was calculated using Equation (1) [57]:(1)ComplexshearmodulusG*=G′2+G″2

A low-temperature 3D printing system was built by modifying a commercial 3D printer [23]. The modified 3D printer was driven using a customized operating program (MotionMaster) (KIRAMS, Seoul, Korea). All 3D scaffolds, except those used for taking measurements of mechanical properties, were designed (Solidworks) to have dimensions of 19.2 mm × 19.2 mm × 3 mm (5 layers). During 3D printing, the temperature of the internal environment was maintained at 20–22 °C, and the cryogenic plate temperature was kept at −10 °C. After transferring 8 mL of a biomaterial ink to a 10 mL syringe, the rubber plunger of the syringe was replaced with a PLF-E plunger (MUSASHI, Mitaka, Japan). Three-dimensional printing was carried out while changing the 3D printing parameters to evaluate biomaterial ink printability, with pneumatic pressures of 20–80 kPa and printing speeds of 5–10 mm/s. The needle size was fixed at 18 gauge (inner diameter: 0.84 mm). A porous 3D scaffold was obtained as each layer adhered to the previous layer to form a 0°/90° strand structure. The printed strand width was confirmed by using vernier calipers (Mitutoyo, Kawasaki, Japan). Immediately after carrying out the 3D printing, the 3D scaffolds were transferred to a −80 °C deep freezer and frozen overnight. Then, the 3D scaffolds were dried by freeze-drying for 24 h. The lyophilized 3D scaffolds were stored at −20 °C, to prevent them from absorbing water, until they were used for later experiments.

The 3D scaffolds were crosslinked using a chemical crosslinking agent. An amount of 10 mM of 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide hydrochloride (EDC) (Daejung, Siheung, Korea)/5 mM of N-hydroxysuccinimide (NHS) (Sigma, Saint Louis, MO, USA) was dissolved in a 90% aqueous ethanol solution. Samples of the lyophilized 3D scaffolds were placed in a 6-well plate, and a 10 mM EDC/5 mM NHS solution was added to each well at room temperature. After 24 h, the crosslinked 3D scaffolds (3D printed scaffolds) were washed three times with 90% ethanol. The 3D printed scaffolds were immersed in a phosphate-buffered saline (PBS) (Welgene, Gyeongsan, Korea) solution.

The printing accuracy of the 3D printed scaffolds was compared with the CAD design by applying the method proposed by M.D. Giuseppe et al. [58]. The size Di (horizontal, vertical, and height) of the 3D printed scaffolds were measured using vernier calipers. The printing accuracy (%) was determined according to Equation (2):(2)Printingaccuracy%=1−Di−DD×100 where Di and D represent the 3D printed scaffold size and CAD designed size, respectively.

The surface and cross-section morphologies of the 3D printed scaffolds were visualized using SEM. The 3D printed scaffolds were transferred to a −80 °C deep freezer and frozen overnight. Then, the 3D printed scaffolds were dried by freeze-drying for 24 h. The 3D printed 3D scaffolds were coated with gold using an ion sputter coater and were characterized using SEM at 20 kV. The printability (Pr) value is highly correlated with the circularity (C) of the strand. The circularity of a strand is defined as in Equation (3):(3)C=4πAL2 where L and A represent the mean perimeter and mean area, respectively. If the C value is 1, the strand shape is a complete circle shape. In a square shape, circularity is equal to π/4. To determine the Pr value of each biomaterial ink, we defined the Pr based on a square shape using Equation (4) [59]:(4)Pr=π4×1C=L216A

SEM images of 3D printed scaffolds were analyzed in ImageJ software to determine the perimeter and area of pores (n = 4).

The mechanical properties of the 3D printed scaffolds were determined using a universal testing machine (Instron 5960; Instron^®, Norwood, MA, USA). For the tensile strength test, 3D scaffolds (20 mm × 7 mm × 2 mm) were fabricated and crosslinked. The 3D printed scaffolds were immersed in a PBS solution at 37 °C for 1 h, and the experiment was conducted. The data used to construct the tensile stress–strain curves of the 3D printed scaffolds were collected using a stretching speed of 10 mm/min. The experiment was stopped at the point where the sample broke. The compressive strength test was performed using cell cultured 3D scaffolds (19.2 mm × 19.2 mm × 3 mm). After 1 day and 14 days of culture, the cell cultured 3D scaffolds were washed three times with Dulbecco’s phosphate-buffered saline (DPBS) (Welgene, Gyeongsan, Korea) solutions and fixed with 4% paraformaldehyde (Sigma, Saint Louis, USA) solutions at 4 °C overnight. The fixed cell cultured 3D scaffolds were washed three times with PBS solutions. The experiments were each conducted with a crosshead speed of 3 mm/min at room temperature. The point where the strain rate became 80% was set as the end point of each experiment. For each sample, the Young’s modulus of the tensile and compressive test was calculated from the stress–strain curve.

The swelling ratios of the 3D printed scaffolds were determined from measurements of changes in their weights. First, the 3D printed scaffolds were transferred to a −80 °C deep freezer and frozen overnight. They were then dried by freeze-drying for 24 h. After being lyophilized, the weight of each dried 3D printed scaffolds was measured (W0). Then, the dried 3D printed scaffolds were immersed in a PBS solution at 37 °C. The weight of each 3D printed scaffold was measured at desired time intervals (10, 20, 30, 60, 120, 240, 360, 720, and 1440 min). The swelling ratio (%) was calculated using Equation (5) [60]:(5)Swellingratio%=Wt−W0W0×100

Here, for each 3D printed scaffold sample,andrepresent the weight of the immersed sample and that of the dried sample, respectively. W t W 0

The in vitro degradability of each 3D printed scaffold was confirmed by using collagenase. Each dried 3D printed scaffold was immersed in a PBS solution at 37 °C for 1 h (Wo) and then transferred to a PBS solution containing collagenase (Type I, 2U/mL, Sigma) at 37 °C. The experiment was conducted for 5 days, and the weight of the sample was measured at each time point (Wt). The remaining mass (%) was calculated using Equation (6) [61]:(6)Remainingmass%=WtWo×100

Here, for each 3D printed scaffold sample,andrepresent the weight of the sample after incubation and that of the original sample, respectively. W t W 0

Human dermal fibroblast (HDF) (American Type Culture Collection (ATCC), Manassas, USA) cells were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) (Gibco, Amarillo, TX, USA) supplemented with 10% fetal bovine serum (FBS) (Tissue Culture Biologicals, Long Beach, CA, USA) and 1% penicillin/streptomycin (Welgene, Gyeongsan, Korea). For each experiment, the medium was replaced every other day. After reaching 70–80% confluency, the cells were detached by using trypsin-EDTA and viable cells were counted using a trypan blue assay. To evaluate the cell affinities of the 3D printed scaffolds, the HDFs were cultured on the 3D printed scaffolds. Before being seeded with the cells, these scaffolds were sterilized with 70% ethanol and ultraviolet light for over 12 h. The sterilized 3D printed scaffolds were then washed three times with DPBS, transferred to a 6-well plate (ultra-low attachment surface, Corning, Corning, USA), and then immersed in DMEM for 24 h for medium pre-wetting. After the pre-wetting, the medium was removed from the 6-well plate. The suspensions of HDFs (p = 9, 3 × 10⁴ cells/100 μL/scaffold) were seeded on the sterilized 3D printed scaffolds and incubated at 37 °C with 5% CO₂ for 3 h. After the 3 h incubation, 5 mL of DMEM was added to each well and the medium was changed every second day. Cells were cultivated at 37 °C, 5% CO₂ for 1, 3, 7, or 14 days.

Cell proliferation was evaluated using a DNA assay. The DNA assay (Quant-iT^™ Picogreen^™ dsDNA Assay kit, Invitrogen, Waltham, MA, USA) was used for cell number confirmation. After the desired culture time, the HDF numbers were evaluated according to the manufacturer’s protocol. To quantify the amount of DNA, the fluorescence intensity was measured using a microplate reader (SpectraMax M2e; Molecular Devices, San Jose, CA, USA) at an excitation wavelength of 480 nm and an emission wavelength of 560 nm.

The morphologies of the HDFs on the 3D printed scaffolds were visualized using SEM and confocal microscopy (LSM 710; Carl Zeiss, Oberkochen, Germany). After 14 days of culture, the cell cultured 3D scaffolds were washed three times with DPBS and fixed with 4% paraformaldehyde solution. The fixed cell cultured 3D scaffolds were washed three times with PBS solutions, and then frozen at −80 °C (overnight) and lyophilized for 24 h. The dried samples were coated with gold and imaged using SEM at 20 kV. To obtain confocal images, the fixed specimens were treated with 0.3% Triton-X for 30 min to permeate the cell membrane. Phalloidin (Alexa 594-phalloidin; Molecular probes, Eugene, USA) was used for F-actin staining and DAPI (D9542; Sigma, Saint Louis, USA) for nuclear staining. After the stainings, the HDFs were visualized using a laser scanning confocal microscope system.

Cell-mediated contractions of the 3D printed scaffolds, specifically mediated by the HDFs, were evaluated. After the desired culture time, the horizontal, vertical, and height dimensions of the cell cultured 3D scaffolds were measured using vernier calipers. They were graphed as a percentage of the cell cultured 3D scaffold area versus day 0. The cell-mediated contraction in each case was calculated using Equation (7) [51]:(7)Scaffoldarea%=AtAo×100 where At and A0 represent the 3D printed scaffold areas after the desired time point and at day 0, respectively.

All experiments were independently repeated at least four times, and data are presented with the mean standard error. Statistical analyses were performed using the Student’s t-test. The significance levels were set at * p < 0.05, ** p < 0.01, and *** p < 0.001.