2.2. Mechanics of the Collagen Hydrogel
Figure 2 shows the compression moduli of collagen hydrogels prepared from collagens at 20 and 30 mg/mL concentrations dissolved in 0.05 and 0.1% acetic acid. The resulting pH was set in two manners: (1) acidic with no compensation and (2) compensated to optimal pH according to the titration curves and predicted models. Each sample was measured approximately 15 and 60 min after printing (with N = 5) for each sample. There is a significant difference in compression modulus between 20 and 30 mg/mL with different types of preparation. Therefore, collagen concentration plays the main role in overall stiffness (unpaired
t-test; for more details, see
Supplementary Material S1).
According to linear regression parameter estimates, all three parameters significantly affect the resulting stiffness of collagen bioink see
Table 1 and
Table 2. The increase in strength over time is evident. After 60 min, the measured modulus did not increase further; therefore, the gels reached their final stiffness (model estimate ‘Incubation time’). A comparison of gels prepared at different concentrations of acetic acid showed an increased compression modulus of gels prepared in 0.05% AA for both collagen concentrations (model estimate ‘AA concentration’—negative coefficient).
The collagen titration method was essential at the beginning of bioink preparation. Without the compensation of pH with the addition of NaOH, the gel’s stiffness is significantly reduced. However, the addition of NaOH in the optimal ratio also significantly increased gel stiffness (model estimate ‘pH optimization’).
The compressive modulus of collagen hydrogels is known to increase with increasing collagen concentration because collagen molecules form a densely packed network at higher concentrations, resulting in higher mechanical strength and stiffness of the hydrogel [
35]. However, there is a limit to how much the compressive modulus can increase with an increasing concentration, as excessively high concentrations can form a more brittle and less elastic hydrogel [
35]. Determining a specific value for the compression modulus of collagen hydrogels is challenging. Each research group uses its own methodology for preparing and measuring hydrogels, so it is not easy to generalize and compare the results.
Acetic acid is used to solubilize collagen and adjust the pH for gelation. The acetic acid concentration can affect the hydrogel’s structure and mechanical properties [
36]. An increasing acetic acid concentration can form a denser network of collagen fibrils, resulting in a higher Young’s modulus. However, other sources state that a higher concentration of acetic acid can lead to a lower pH and a less crosslinked collagen structure, resulting in a lower Young’s modulus [
37]. Careful control of the concentration of acetic acid is essential for achieving the desired mechanical properties for tissue engineering applications.
Acetic acid and sodium acetate (formed during the neutralization of acetic acid by sodium hydroxide) are typical examples of buffers. A buffer refers to a solution that can withstand changes in pH when an acidic or basic component is added. It has the ability to neutralize small amounts of added acid or base, thereby keeping the pH of the solution relatively constant and stable. If the concentration of acetic acid is lower, then the buffering capacity of the entire system is also lower. This makes it easier to achieve a neutral pH, which, in turn, leads to the formation of a gel.
2.3. Cell Viability and Morphology
Figure 3 and
Figure 4 show a comparison of static culture and active perfusion ‘dynamic’ culture samples when cultured in a growth culture medium in the hydrogel with an initial collagen concentration of 20 or 30 mg/mL. In both types of gels, a decrease in cell number can be observed in static culture from the beginning to day 7 of culture. Cell loss was insignificant on the third day of culture, but cells were round, indicating poor cell viability. On the fifth day of culture, there was already a noticeable decrease in cell number, and by the seventh day, almost no viable cells were found in the culture. The culture of the samples with active perfusion stimulated the cells to grow and divide, and the cells elongated and intermingled. Their number increased until the fifth day of culture, but by the seventh day, their number was reduced because their differentiation potential was suppressed by growing in a proliferative culture medium.
In bioprinted collagen hydrogels, the typical cell morphology can vary depending on the specific cell type used for culture. Generally, cells in collagen hydrogels exhibit a rounded morphology as they adapt to their new environment and begin to proliferate. As cells grow and remodel the surrounding collagen matrix, they can take on a more elongated or spindle-like shape. Cells embedded deep in the hydrogel, especially in static cultures, have a poor nutrient supply, so their shape is often round. However, the cells on the surface are already elongated like fibroblasts [
38]. In general, bioprinted collagen hydrogels that support high cell viability and functionality tend to exhibit well-organized, elongated cell structures that resemble native tissue [
39], which was achieved in our samples with active perfusion.
Moreover, cells remodeled the printed substrate according to its dimensional changes, as illustrated in macroscopic views and charts in
Figure 3 and
Figure 4. According to linear regression estimates, all observed parameters have a significant role in substrate remodeling; see
Table 3 and
Table 4. The type of cultivation promoted over a prolonged cultivation period has the most significant impact on geometry change.
In 20 mg/mL bioink, there was a change in area from approx. 200 mm2 to approx. 60 mm2 when active perfusion for 7 days was used, whereas in static conditions, it was lowered to approx. 125 mm2. Especially for static cultivation, there was a major change after 1 day of cultivation; however, this change stagnated. In active perfusion, the change was observed on all days of cultivation. Thickness was reduced from approx. 1.2 mm to 0.85 in active perfusion to 1.06 in static.
In 30 mg/mL bioink, there was a change in area from approx. 200 mm2 to approx. 125 mm2 when active perfusion for 7 days was used, whereas in static conditions, it was lowered to approx. 175 mm2. This change developed on all days of cultivation. Thickness was reduced from approx. 1.2 mm to 1.05 mm in active perfusion to 1.07 mm in static. This thickness change is not so high compared to 20 mg/mL collagen but is still statistically significant according to the linear regression model.
An increased rate of remodeling in active perfusion correlates with higher viability and metabolic activity.
Figure 5 shows the results of static and active perfusion ‘dynamic’ culturing in hydrogels at 20 mg/mL concentrations for two hydrogel thicknesses when cultured in a culture medium for differentiation into smooth muscle cells. The results were relatively comparable for static culture and culture with active perfusion regarding proliferation and cell number. In both cultures, there is an increase in cell number at all intervals; the cells are elongated and, by the fifth day, have already formed a compact structure. However, there is shrinkage and thickening of the collagen hydrogel at the same time, and on day 7, due to collagen remodeling, the shrinkage was so significant that the results could not be analyzed without distortion of the data.
However, there is a significant difference between the two cultures in producing calponin, which regulates the growth and differentiation of smooth muscle cells. On the first day of culture, the differences are not significant. However, on the third day of culture, the calponin production is several times higher in the culture with active perfusion, as well as on the fifth day of culture. This indicates a much higher potential for cells to differentiate into smooth muscle when cells are supported by active perfusion, which agrees with the existing literature [
29,
31].
Figure 6 shows the results of static culture and culture with ‘dynamic’ active perfusion in a 30 mg/mL hydrogel for two hydrogel thicknesses when cultured in a culture medium for differentiation into smooth muscle cells.
For these samples, the resulting collagen concentration is 15 mg/mL, which is already a relatively high concentration for cells in static culture, resulting in gradual cell death, especially in the thicker sample in which there is insufficient diffusion of gases and metabolites. Despite this, the cells form an elongated shape, indicating partial differentiation of the cells into smooth muscle cells, as shown by the SMC marker calponin. On the other hand, under active perfusion, the cells prosper, their number increases, and in 5 days, they form a fully grown structure with high differentiation toward SMC. As in the previous case, the collagen hydrogel was so remodeled and shrunken that it was impossible to evaluate the results on day 7 without biasing the data.
During cultivation, the shrinkage and remodeling of hydrogel samples from both types of cultures occurred independently of the collagen concentration. Hydrogel shrinkage is a natural process that occurs in cell cultures with physical stresses (pressure, electric field, etc.) [
40]. Other mechanisms of its formation include dehydration or cell contraction [
40]. The shrinkage process is essentially the repeated spreading of cells, their pull on the hydrogel in the radial direction, and the reduction of hydrogel elongation over time [
41]. The shrinkage can lead to changes in the mechanical properties of the hydrogel, affecting cell behavior and tissue development. Gel shrinkage induces changes in mass transfer efficiency, cell distribution, and density of adhesive ligands of the surrounding matrix [
42].
Shrinkage can be beneficial, as it can lead to increased mechanical stability and improved cell alignment within the hydrogel; on the other hand, excessive shrinkage can negatively impact the viability and functionality of cells within the hydrogel, as it can lead to increased stress and strain on cells [
40]. In our samples, the reduction of the pattern area from approx. 200 mm
2 (initial size 14 × 14 mm) dropping to 25 mm
2 was achieved (
Figure 3,
Figure 4,
Figure 5 and
Figure 6). Morphology and viability assays showed that even under this significant change, cells are still viable (in samples with active media perfusion) and form the structure of future tissue. The reduction in the area of the sample is further supported by the fact that the cells metabolize the surrounding collagen into their own extracellular matrix, thus remodeling the entire sample.
To demonstrate this fact, we again evaluated cell area and sample thickness for 20 and 30 mg/mL collagen, both cultivations and initial print heights in the differentiation medium. Also, linear regression model estimates were made (see
Table 5 and
Table 6). The differentiation medium promoted remodeling and collagen shrinkage in both cultivation types. Not even a significant difference was achieved by type of cultivation, which was also confirmed by the regression model.
However, there was one interesting finding: in the proliferation medium, the cells were able to remodel structure in terms of area and thickness in the same manner as in the differentiation medium; the thickness change was inverse (negative rank for cultivation in the regression model). In active perfusion, there was a similar scenario as in proliferation media. The initial thickness was reduced from approx. 1.2 or 0.8 mm to approx. 0.9 or 0.6 mm. On the other hand, in static conditions, the thickness increased over cultivation time from an initial approx. 1.2 or 0.8 mm to approx. 1.3 or 0.9 mm. We believe this is caused by shrinkage caused by calponin active and viable cells on the top side of the sample, whereas mainly dead or apoptotic cells are on the bottom side. The active perfusion allows overall diffusion; thus, both sides have viable cells, and substrates are remodeled homogenously. The dynamic samples were homogenously reduced in terms of thickness, and the static samples created more blob-like structures, as seen in cell morphology.
In addition, the MTT metabolic activity was determined. These MTTs are collated with confocal images and LD staining. The metabolic activity in dynamic culture is significantly higher than in static culture, as illustrated in
Figure 7.
Several factors can affect cell viability in collagen hydrogel, including collagen concentration in the hydrogel, stiffness, porosity, or hydrogel geometry [
43]. The high protein concentration in the hydrogel can lead to poor cell viability and limited functionality of the printed constructs. Nutrient supply and/or removal of waste products is critical to maintaining cell viability in a dense collagen gel, which was proven by an experiment where the viability of the static sample was only 20% compared to the perfused sample with 80% live cells [
44]. Incorporating active perfusion during cultivation can significantly improve the functionality and longevity of tissue constructs while cells are exposed to a more physiologically relevant environment [
45]. The same results were obtained in our study. In samples with active perfusion (
Figure 8 above), the culture medium flows around the entire gel surface, increasing the diffusion efficiency and exchange of gases and metabolites. The sample is well populated with dead cells evenly distributed on both halves. The gels adhere to the culture glass on one side for static samples, and the sample is flooded with culture medium in the well (
Figure 8 below). Diffusion is, thus, limited to surfaces without a substrate barrier, insufficient to nourish cells located inside and towards the bottom edge. Therefore, cell viability is reduced at these sites, as shown in
Figure 9. Utilizing hydrogels in bioprinting can shield cells from shear stress and consequent membrane damage, thereby improving their viability in the printed structure. Mechanical stresses experienced during extrusion can damage cell membranes and result in decreased cell viability [
46].
The above results were further confirmed using the SEM method, which visualizes the hydrogel surface in detail.
Figure 10,
Figure 11,
Figure 12 and
Figure 13 show the difference in the inner architecture of the gels after 1, 3, 5, and 7 days of cultivation. On the first day of cultivation, the hydrogels are compact, without significant cavities, except for naturally occurring air bubbles. The compactness of the gel is higher in the case of a higher collagen concentration (30 mg/mL). Cell growth through the hydrogel can be observed on the third and fifth day of culture. Cells contract the gel and remodel the collagenous matrix into its own extracellular matrix. It can be observed that the occurrence of cavities and pores is more pronounced in the active perfusion system. Finally, on the seventh day, the structural changes are the most significant, with the appearance of empty spaces around individual cells, which remodel the collagen matrix.
The changes that occur in the collagen hydrogel’s internal structure are clearly visible at the macroscopic level, i.e., at tens to hundreds of micrometers. The contrary would be at the microscopic level, i.e., at the level of nanometer units, when individual collagen fibers or filaments are already visible (
Figure 14) without their geometrical change over the cultivation period.
Static cultivation with both proliferation and differentiation media has reduced viability. In proliferation media, a large number of apoptotic cells or cell debris were observed in confocal images. Also, according to the SEM images and the geometrical change, the remodeling was minimal. In differentiation media, surface cells were partially differentiated into SMC-like cells. Apoptotic cells in lower layers caused asymmetric tension and remodeling, generating a blob-like formation of the sample. This was also confirmed by the increased thickness of samples after cultivation.
Active perfusion with proliferation media maintains viable cells in overall volume with homogenous remodeling of the substrate as contrasted to SEM images.
Active perfusion with a differentiation medium promotes partial cell differentiation into SMCs within the whole volume and cell viability with strong substrate remodeling and shrinkage. This is demonstrated with significant geometrical change.
When comparing the samples of all culture types and culture media, it can be concluded that the best optimal morphological proliferation is achieved by the bioink sample with active perfusion with the SMC differentiation culture medium. In this sample, cells have high viability throughout the culture period, an optimal elongated shape with protrusions, high calponin production, and a high degree of remodeling of the collagen hydrogel in the ECM. This confirms that the choice of culture with active proliferation and the promotion of cell differentiation by culture medium chemicals have a significant impact on the success of cell cultures in highly concentrated collagen hydrogels. All these milestones are illustrated in
Figure 15.