*3.6. Morphology of Cells on the Hydrogels*

Bright-field images of hASC, L929, and SaOS-2 (Figure 5) cultured on silk hydrogels for 1 and 5 days showed the cell morphology after the initial attachment and in the exponential growth phase, respectively. Fluorescent images of hASCs (Figure 6) demonstrated live and dead cells, stained by calcein AM and PI dyes, respectively. The number of cells visualized from the images were in accordance with the cell proliferation results. It can be recognized that cells attached to hydrogels containing SF within 1 day after seeding. In contrast, cells on hydrogels with an increasing content of eADF4(C16) were less stretched and presented a lower attachment. On pure 2% eADF4(C16), cells could not attach well and preferably formed cell aggregates.

**Figure 5.** Bright-field images of hASC, L929 and SaOS-2 cells cultured on 2% SF:eADF4(C16) hydrogels on day 1 (**top** panel) and day 5 (**bottom** panel) (scale bar = 50 μm).

**Figure 6.** Fluorescence images of calcein AM (Live) and PI (Dead) stained hASC cells cultured on 2% SF:eADF4(C16) hydrogels on day 1 and day 5 (scale bar = 500 μm).

#### **4. Discussion**

Blended SF:eADF4(C16) hydrogels were fabricated in order to combine the advantages of eADF4(C16) in facilitating rapid gelation and enhanced interactions with cells provided by SF. As shown in Figure 1, it can be clearly seen that the addition of eADF4(C16) can induce faster gelation of SF, i.e., the gelation of 5:5 and 3:7 hydrogels occurred within 48 h. The gelation of pure 2% and 3% SF solutions were not observed after one week, which was in agreement with a previous study [20]. As expected, an increase of protein concentration from 2% to 3% led to a shorter gelation time, due to a higher opportunity in chain-chain interactions [5,10]. Additionally, the presence of cations, especially the mixtures of monovalent and divalent cations as in DMEM, significantly shortened the gelation time of SF:eADF4(C16) blends. This finding confirmed the previous study in which the addition of DMEM and divalent cations, such as Ca2+, triggered a faster eADF4(C16) hydrogel formation [11]. Also, accelerated gelation of SF was achieved in presence of the divalent cation Ca2+, but not the monovalent K+ [21]. However, our results showed that mixtures of monovalent cations, such as in PBS, and NSS, were able to affect the gelation of SF:eADF4(C16). We propose that the presence of cations mostly influences assembly of eADF4(C16) rather than that of SF, due to the polyanionic characteristics of eADF4(C16). Interactions with counterions result in a decrease of chain repulsion and facilitates molecular interactions.

Typically, the sol-gel transition of silk proteins, either by spontaneous gelling or physical intervention, relates to the self-assembly process [22]. Conformational changes from random coil in sol state to the highly ordered beta sheet structure in gel state were in accordance with the gelation of SF and eADF4(C16). The hydrogen bonding and hydrophobic interaction between hydrophobic glycine-alanine repeats or poly-alanine sequences result in a formation of beta sheet stacks and a subsequent gel formation [5,10]. Our findings showed a reduction of random coil structure as well as an increased beta sheet content after the gelation of all samples (Figure 3B). Notably, the eADF4(C16) showed a lower degree of beta sheet structures than that of SF. As the heavy chain of SF, which mainly directs the formation of beta sheets, possesses a molecular weight of 391 kDa [4], the presence of smaller eADF4(C16) chains with a molecular weight of 47.7 kDa [23] could disturb the molecular organization, resulting in the detected reduction of beta sheet formation upon an increase of eADF4(C16) content.

Cell interactions of SF:eADF4(C16) hydrogels were evaluated using the primary cell, hASC, and two cell lines, L929 mouse fibroblasts and SaOS-2 human osteoblast-like cells. For all cells, those cultured on pure eADF4(C16) hydrogels were rounded, loosely attached to the hydrogel surface, and preferably formed cell aggregates (Figure 5). The surface hydrophilicity, determined from water contact angle measurement of blended films (Figure S1), and the protein adsorptivity of the hydrogels (Figure 3C) could not strongly influence the cellular adhesion. In addition, as the isoelectric point of SF and eADF4(C16) was approximately 3.9 and 3.5, respectively [24,25], the negatively charged materials at the physiological pH (pH 7.4) could not well support cellular attachment through electrostatic interaction with negative-charged cell surface.

The proliferation of hASC primary cells on all hydrogels was slower, and a lower cell number was obtained than that on the TCP control (Figure 4A). SF and eADF4(C16) hydrogels were biocompatible, as seen from no dead cells in all samples (Figure 6). However, the lack of cell adhesion motifs in eADF4(C16), such as RGD, resulted in a low cellular adhesion to the material [26,27]. Furthermore, cells maintained their spherical forms and preferably formed micro-aggregates due to the loose attachment on the hydrogel surface [28]. Therefore, an extended initial lag phase of primary cells with low proliferation potential could be observed, and any growth phase could not be observed within the experimental period.

For L929 fibroblastic cells, proliferation rate depended on the type of samples, as hydrogels with high SF content supported cellular activities better (Figure 4B). Apart from the reason that eADF4(C16) cannot support proper cellular adhesion [12,29], Yamada et al. reported the presence of fibroblast growth-promoting peptides, VITTDSDGNE and NINDFDED, at the N-terminus of SF heavy chain [30]. Hence, the higher content of SF in the blended materials could be beneficial in promoting the proliferation of L929 fibroblasts.

As shown in Figure 4C, the growth of SaOS-2 on 10:0 hydrogels were statistically similar to the control (TCP) on day 5 and 7, and the values were proportionally reduced with increasing eADF4(C16) content. Furthermore, as noticed from Figure 6, the hASC could spread more on hydrogels with higher SF content. It can be presumed that the material acts as a physical cue directing the cellular activities including cytoskeleton organization and cell morphology [31]. Cells cultured on the stiffer materials exhibit a stretch morphology, generating a greater force on actin cytoskeleton, which favor an osteogenic expression [32,33].

The blended hydrogels, especially the 7:3 and 5:5 SF:eADF4(C16) formulations, exhibited accelerated gelation kinetics together with an enhanced cellular activity. Our results showed that the physical and biological properties of the hydrogels are tunable depending on the blending ratio. The simple blending of two different silk proteins reflects a simple preparation route to obtain the hydrogels with required properties such as biological activity of mechanical features for an intended cell biological application.

#### **5. Conclusions**

Blended SF:eADF4(C16) solutions show accelerated sol-gel transition in combination with the enhanced cell binding of the resulting hydrogels. Faster gelation was noticed with an increment of eADF4(C16) content to at least 50%. At the ratio of 7:3 and 5:5 SF:eADF4(C16), an enhanced cellular adhesion as well as cell proliferation have been

noticed. The developed blended hydrogels supported viability of hASC primary cells. Proliferation of cell lines depended on SF content, as an increasing SF content enhanced cell proliferation. The different behavior of each cell type on the blended SF:eADF4(C16) hydrogels could serve as the fundamental data to design the applications of such hydrogels in the future. In addition, for further works, the mechanical properties of the SF:eADF4(C16) should be evaluated, both the modulus and the thixotropic properties of the hydrogels, to investigate their utilization as a printable bioink. Since the high crystallization of SF could limit the shear thinning and self-recovery of the hydrogels, the addition of eADF4(C16) which is known for its thixotropy [11] could ameliorate the printability of the developed hydrogels.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/10 .3390/polym13234182/s1, Figure S1: Water contact angle of blend SF:eADF4(C16) films.

**Author Contributions:** Conceptualization, C.L., J.R., M.H., T.S. and S.D.; Data curation, C.L.; Formal analysis, C.L.; Funding acquisition, J.R., S.K. and S.D.; Investigation, C.L.; Methodology, C.L., M.H., T.S. and S.D.; Project administration, T.S. and S.D.; Resources, A.V., J.R., Jittima Luckanagul, M.H., T.S. and S.D.; Supervision, A.V., J.R., S.K., J.A.L., M.H., T.S. and S.D.; Validation, J.R., S.K., J.A.L., M.H., T.S. and S.D.; Visualization, C.L.; Writing—original draft, C.L.; Writing—review & editing, A.V., J.R., S.K., J.A.L., M.H., T.S. and S.D. All authors have read and agreed to the published version of the manuscript.

**Funding:** This study was supported by the project "The enhancement of engineering academic collaboration with the Federal Republic of Germany", Ratchadapisek Somphot Fund, Chulalongkorn University. Authors also acknowledged the financial supports from the National Research Council of Thailand (NRCT5-RSA63001-20) and Chulalongkorn University (RES\_64\_117\_21\_014).

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Institutional Review Board of Faculty of Medicine, Chulalongkorn University (IRB No.416/61, Date of Approval: September 10, 2019).

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** Data is contained within the article and Supplementary Materials.

**Conflicts of Interest:** T.S. is co-founder, shareholder and consultant of AMSilk GmbH, Martinsried, Germany.
