Myoblast 3D bioprinting to burst in vitro skeletal muscle differentiation

Abstract Skeletal muscle regeneration is one of the major areas of interest in sport medicine as well as trauma centers. Three‐dimensional (3D) bioprinting (BioP) is nowadays widely adopted to manufacture 3D constructs for regenerative medicine but a comparison between the available biomaterial‐based inks (bioinks) is missing. The present study aims to assess the impact of different hydrogels on the viability, proliferation, and differentiation of murine myoblasts (C2C12) encapsulated in 3D bioprinted constructs aided to muscle regeneration. We tested three different commercially available hydrogels bioinks based on: (1) gelatin methacrylate and alginate crosslinked by UV light; (2) gelatin methacrylate, xanthan gum, and alginate‐fibrinogen; (3) nanofibrillated cellulose (NFC)/alginate‐fibrinogen crosslinked with calcium chloride and thrombin. Constructs embedding the cells were manufactured by extrusion‐based BioP and C2C12 viability, proliferation, and differentiation were assessed after 24 h, 7, 14, 21, and 28 days in culture. Although viability, proliferation, and differentiation were observed in all the constructs, among the investigated bioinks, the best results were obtained by using NFC/alginate‐fibrinogen‐based hydrogel from 7 to 14 days in culture, when the embedded myoblasts started fusing, forming at day 21 and day 28 multinucleated myotubes within the 3D bioprinted structures. The results revealed an extensive myotube alignment all over the linear structure of the hydrogel, demonstrating cell maturation, and enhanced myogenesis. The bioprinting strategies that we describe here denote a strong and endorsed approach for the creation of in vitro artificial muscle to improve skeletal muscle tissue engineering for future therapeutic applications.


| INTRODUCTION
Skeletal muscle has the great capacity to self-repair and regenerate in response to common acute injuries, such as exercise-induced damage (Giarratana et al., 2020;Ronzoni et al., 2021). This is principally due to a resident stem cell population that is mainly involved in skeletal muscle homeostasis and regeneration. It has been demonstrated that these muscular progenitor cells are able to fuse, forming myotubes even if treated with recombinant proteins (Agosti et al., 2020;Perini et al., 2015;Ronzoni et al., 2011Ronzoni et al., , 2017. However, when muscle loss becomes irreversible (e.g., in case of severe trauma, invasive surgeries, degenerative diseases, or because of aging), lesions are so critical that they impair muscle functionality (Young, 1964). In this scenario, muscle regenerative medicine can provide solutions (Langridge et al., 2021;Ronzoni et al., 2020).
Several studies focused on the production of an ideal structure to induce muscle tissue regeneration, including biochemical components to ensure efficient myogenic differentiation and maturation, resulting in thick and elongated myotube formation (Kang et al., 2016). However, the current challenge is to ensure the uniform growth of muscle cells inside the biomaterial and to induce a contractile syncytium similar to the native skeletal muscle structure (Chen, 1993) despite, over the years, different biomaterials and scaffold designs have been experimentally and/or clinically evaluated for the repair of skeletal muscle tissue.
In particular, porous three-dimensional (3D) scaffolds have been manufactured using natural or synthetic polymers (Melchels et al., 2012), hydrogels (Baar et al., 2005;Fedorovich et al., 2008;L'Heureux et al., 2006;Stevens et al., 2009;Visser et al., 2013), decellularized extracellular matrix (dECM), and their composites (Ott et al., 2008). Several advantages emerge from the use of such natural hydrogels, such as mimicking skeletal muscle environment, providing bioactive signaling for muscle differentiation, and reabsorbing the biomaterial to allow the in vivo interaction of myofibers (Lev & Seliktar, 2018). Advantageous is also the use of dECMs that preserve the native tissue architecture, facilitate the adhesion of muscular cells and promote the regeneration of the tissue area in which the damage is (Lev & Seliktar, 2018;Wolfe & Sell, 2011). Nevertheless, there are some limitations associated with the use of such natural materials; for instance, the inadequate supply of nutrients to the cells in the central portion of the bioconstruct or, regarding dECMs, long incubation times to observe the effective functional recovery of the damaged tissue is required (Smoak & Mikos, 2020). As for the synthetic polymeric matrices, they do not guarantee good cell adhesion, they are poorly absorbable and there is a greater risk of activation of immune response of the patients. Therefore they are not considered biocompatible (Lev & Seliktar, 2018). Costantini et al. (2017) encapsulated C2C12 murine myoblast into gelatin methacryloyl hydrogel (CELLINK ® GelMA -CELLINK AB, Gothenburg, Sweden) using 3D mold to evaluate 3D cell culture in terms of in vitro myogenesis; moreover, they demonstrated that both hydrogel stiffness and geometrical confinement play a crucial role in the differentiation of myogenic precursors in a threedimensional environment. Otherwise, Seyedmahmoud et al. (2019) encapsulated C2C12 not only in CELLINK ® GelMA, but also in CELLINK ® GelMA mixed with different percentages of alginate (6% and 8%). They demonstrated that alginate percentage can provide a more favorable mechanical microenvironment for murine myoblasts (C2C12) cell proliferation and an optimal niche to induce muscle tissue formation.
Bauer and colleagues (Costantini et al., 2018) demonstrated that spreading and proliferation of C2C12 cells encapsulated into alginate-based hydrogel were impacted by both stiffness and stress relaxation behavior of the substrates created by 3D molding. In addition, Matthias et al. (Costantini et al., 2018)  Furthermore, thanks to the advancement of additive manufacturing, three-dimensional bioprinting is nowadays a widely adopted technique for both manufacturing 3D scaffolds and constructs in various tissue engineering approaches (Nikolova & Chavali, 2019). In fact, BioP not only allows the production of scaffolds whose geometry can be controlled thanks to the use of specific software, but it can also be exploited for the manufacturing of different scaffolds based on different biomaterials in which different cell types can be encapsulated (Derby, 2012;Leong et al., 2003;Murphy & Atala, 2014). The outcome of BioP, which is a complex process defined by several steps, is conditioned by the printing technology and biomaterial adopted, which defines when combined with cells the so-called bioink (Groll et al., 2019;Matai et al., 2020;Ng et al., 2019).
Bioprinting techniques can be classified according to the printing methods, in particular, it is possible to distinguish three main BioP techniques: inkjet, extrusion, and vat-polymerization (AmerDababneh & Bioprinting Technology, 2014). These techniques vary in precision and accuracy in the deposition of the material, stability, and cell survival.
The inkjet-based BioP was the first technique to be implemented.
The bioink solution is manipulated by generating droplets which are deposited on a substrate using a small nozzle. The jet delivered can be of three types: continuous, on command (drop-on-demand) and electrodynamic (Gudapati et al., 2016). This technique offers many advantages thanks to its simplicity, versatility, and control in the bioink deposition of the allowing to control the bioink volume to be deposited. The disadvantage is that inkjet technique does not allow to process high viscosity bioink.
Extrusion-based BioP is a combination of a pneumatic or mechanical fluid dispensing system and an automatic robotic system for the extrusion and the 3D printing (Jiang et al., 2019).
The bioink is dispensed by a deposit system on a substrate on which, thanks to a light, chemical solutions or thermal transistors, the crosslinking of the bioink takes place, thus obtaining the deposition of cells encapsulated in cylindrical filaments, allowing the creation of 3D structures. The mechanical extrusion of the bioink solution involves the use of a piston or a screw, while the pneumatic extrusion involves the use of compressed air. Although the extrusion BioP is the most used technique in this field, there are some limitations for the realization of the desired structure such as the shear effort and the limited selection of the material due to the need to encapsulate the cells inside the bioink and its rapid gelling.
Vat polymerization-based bioprinting uses different photoinitiators and UV light during the bioprinting process for crosslinking the hydrogel (Ng et al., 2020). Although this technique allows for the creation of high-resolution 3D constructs, the UV light used for crosslinking can damage the cells with a consequent reduction in the ability of cells to proliferate and differentiate.
Given such premises, also in the case of BioP for muscle regeneration, the selection of appropriate biomaterials and the resulting bioink is vital to obtain desired biological outcomes. Among the various solutions proposed by the literature and thanks to their features, hydrogels combined with MP cells (C2C12), are commonly used as bioink for skeletal muscle regeneration (Langridge et al., 2021;Malda et al., 2013). In fact, hydrogels are known to be material with high biocompatibility and biodegradability. In addition, their mechanical properties could be modulated by the amount of chemical, temperature, or photo-crosslinking, to modify the elastic modulus to be as much similar as skeletal muscle tissue (Fischer et al., 2020). Hydrogel-based bioinks interact with cells in vitro and in vivo, so their viscosity may be optimized to maintain cell integrity and viability during the printing process. For this purpose, it is possible to use natural (chitosan, alginate, collagen, fibrin, etc.) and synthetic

| MATERIALS AND METHODS
Murine myoblasts were mixed with three commercial hydrogels (CELLINK ® GelMA A, CELLINK ® GelXA FIBRIN, CELLINK ® FIBRIN) and extruded by pneumatic extrusion-based bioprinter (INKREDIBLE + ® ). In the resulting constructs, C2C12 proliferation and differentiation were analyzed at different time points (24 h, 7, 14, 21, and 28 days) using morphological tests (Live/Dead staining and immunofluorescence [IF]). Molecular biology tests were also performed to quantify the gene expression of specific myogenic markers involved in muscle fiber maturation.

| Hydrogels and crosslinkers
The experiments were performed using commercially available Gelatin-based hydrogel and alginate (CELLINK ® GelMA A), Xantan gum and Fibrinogen hydrogel (CELLINK ® GelXA FIBRIN) and nanofibrillated cellulose (NFC)/alginate-fibrinogen-based hydrogel (CEL- Gelatin-based and alginate hydrogel (CELLINK ® GelMA A). The chemical composition of this hydrogel is a blend of CELLINK® GelMA and alginate, offering a higher printability compared to pure CEL-LINK ® GelMA hydrogels. This is due to the provided softening of the alginate and to essential properties of native ECM that allow cells to proliferate and spread. CELLINK ® GelMA A 3D constructs were crosslinked by photopolymerization, or through the addition of the ionic crosslinking solution (50 mM CaCl2).

Xanthan gum and Fibrinogen hydrogel (CELLINK ® GelXA FIBRIN).
This hydrogel incorporates GelMA base, xanthan gum and alginate to enhance printability and stability of the 3D constructs, while fibrin improves muscle cell proliferation and differentiation. A combination of photoinitiator-assisted and ionic crosslinking was applied. In Table 1 are summarized the hydrogels and relative crosslinkers used for each round of 3D printing experiments. In addition, rheological tests were carried out directly by CELLINK (CELLINK AB) for each hydrogel ( Figure S1).

| Bioprinting process
Before starting the printing process, the bioprinter was placed under a sterile hood and UV light was turned on for 1 h to sterilize all the materials and surfaces. Hydrogel was mixed with C2C12 cells (10:1 ratio). The Cartridge was filled with bioink, then nozzle connected (inner diameter 0.25 mm) and finally placed into the printhead. The axes were homed, the z-axis was calibrated, and the pressure and printing speed was set according to standard guidelines (10-15 kPa and 1000 mm/min respectively for all bioinks tested). The 3D constructs were bioprinted on a Petri dish, then the crosslinking process was performed as follows. For chemical crosslinking, CaCl 2 droplets were applied to cover the whole 3D structure and immediately after, the samples were incubated for 5 min at room temperature (RT). The crosslinking solution was subsequently removed from the constructs and DMEM culture complete medium was added. Dishes were then incubated at 37°C and 5% CO 2 . Only the chemical crosslinking process was repeated weekly before medium refreshment to keep the three-dimensional structure unchanged and avoiding degradation. For UV crosslinking, 3D constructs were exposed once to UV light at 365 nm for approximately 3/5 s.

| 3D structure
To mimic morpho-physiology muscle fiber structure, 3D geometry lines formed by one layer were bioprinted. Line length was set at 20 mm, while the line thickness is given by the combination of pressure and printing speed. In this case, it is equal to 0.35 mm ( Figure 1). Given the simplicity of the structure considered, we directly implemented the G-code of the 3D virtual model.

| Cell culture of 3D constructs
3D bioprinted constructs were cultured up to 28 days in DMEM complete medium at 37°C and 5% CO 2 . The culture medium was refreshed every 3 days. 3D constructs were crosslinked every 3 days for 5 min. Following 4 days of BioP, the differentiation process of C2C12-laden bioink was induced by using a differentiation medium (DM) composed by DMEM supplemented with 2% fetal bovine serum.

| Live/dead staining
To evaluate cell viability, we used the Live/Dead staining (Invitrogen); 500 μL of a solution consisting of 1.5 ml of Phosphate Buffered Saline (PBS), 3 μL of EthD-1 and 1.5 μL of calcein, was added to 3D constructs. Samples were incubated for 45 min in the dark, then the solution was removed, and cell nuclei were counterstained with 500 μL 4 0 ,6-diamidino-2-phenylindole (DAPI) for 10 min according to the protocol. Fluorescent image acquisition was carried out by semiconfocal microscope (ViCo confocal, Nikon).
Viability and differentiation tests were performed as well as morphological and gene expression analysis at six different time points (1, 4, 7, 14, 21, and 28 days in culture).

| Total RNA extraction and quantitative realtime PCR
Expression levels of myogenic genes were analyzed on 3D bioprinted constructs by Quantitative real-time PCR (RT-qPCR).
Total RNA derived from each sample was extracted and isolated at different time points using 300 μL of lysis buffer (TRIzol Reagent).
Total RNA extraction was performed by using Direct-zol RNA Miniprep's reagents following the manufacturer protocol (Zymo Research). Total RNA was then quantified by NanoDropTM (Thermo-Fisher Scientific). cDNAs obtained from 350 ng of RNA were reverse transcribed using iScript™ cDNA Synthesis Kit (Biorad) and quantitative PCR analysis was performed using oligonucleotide primers

| Immunofluorescence assay
Immunofluorescence assay on in vitro 3D constructs was performed (1:40). Samples were counterstained with DAPI to detect nuclei, washed three times with a washing buffer, and ultimately mounted.
Finally, sections were observed with a semi-confocal microscope (ViCo confocal, Nikon), supported by the ImageJ PRO 6.2 software. cells were not merged forming myotubes. This is probably due to a non-homogeneous diffusion of the crosslinking solution or to lower oxygen and nutrient levels within the 3D constructs (Figure 2i).

| Live/dead staining
Finally at 21 and 28 days, C2C12 cells merged forming myotubes even in the most central part of the 3D structure, and the alignment was promoted by the linear shape of the printed construct (Figures 2l,m).
Regarding C2C12 cells laden in CELLINK ® GelXA hydrogel, 94% viability was observed at all the time points analyzed (Figure 2n-o).
Nevertheless, at 7 and 14 days in culture, cells kept a round shape and slowly start to elongate only at day 21 especially at the borders of the constructs (Figure 2o-q).
Live/Dead staining in proliferative conditions was also performed on CELLINK ® FIBRIN 3D constructs, crosslinked with CaCl 2 and Thrombin. We observed no advantages on cell viability, adhesion, spreading, and differentiation (data not shown).

| Gene expression analysis of cell-laden structures by quantitative real-time PCR
Gene expression analyses were performed to evaluate and validate the observed differentiation rate of C2C12 cells laden into CEL-LINK ® FIBRIN hydrogel at 7, 14, 21, and 28 days and into CELLINK ® GelXA FIBRIN hydrogel at 7, 14, and 21 days in culture in proliferative and differentiative conditions ( Figure 5 and Figure 6).

The expression levels of myogenic genes such as MyoD and MCK
in the 3D structures were detected by RT-qPCR normalized by the PGK gene.
Regarding the CELLINK ® FIBRIN hydrogel, after 7 days of culture in proliferative conditions, MyoD and MCK were expressed 1.2fold higher than in 3D cultures (Figure 5a).
Similarly, after 7 days in DM, the expression of both genes was 1.8-fold higher in 2D than in 3D (Figure 5a, p > 0.05). Thus, at the Gene expression was also evaluated for the other hydrogel (CELLINK ® GelMA A), but no statistical differences were highlighted among the samples (data not shown).
In conclusion, CELLINK ® FIBRIN hydrogel, as indicated also by Live/Dead staining, improves myogenic gene signature, and proves to be the best bioink to promote myoblast alignment along the printed filament.

| DISCUSSION
In this study, we demonstrated the impact of different types of hydrogels on the viability, proliferation, and differentiation of murine myoblasts encapsulated in 3D constructs and manufactured by pneumatic extrusion based BioP.
Skeletal muscle tissue engineering characterizes a revolutionary branch of regenerative medicine which aims to recreate in vitro muscles to be studied ex vivo and ultimately for the substitution of diseased or damaged muscle tissue. Up to the present time, many different strategies have been proposed even if not fully suitable for a potential therapeutic application (Fuoco et al., 2016;Levenberg et al., 2005;Shadrin et al., 2016;Sicari et al., 2014). One of the main issues encountered was the identification of the best hydrogel to achieve sarcomerogenesis and the parallel-oriented myofiber organization resembling the correct skeletal muscle structure. Therefore, to improve skeletal muscle tissue engineering, innovative techniques are required to produce engineered constructs with precise 3D structures. To date, pioneering technologies are revolutionizing many different manufacturing fields, including tissue engineering (Costantini et al., 2018). Especially 3D BioP techniques showed a prodigious potential for the rapid and cost-effective fabrication of cellularized structures, to build human-sized myo-constructs (Agosti et al., 2020;Mozetic et al., 2017;Ott et al., 2008). Although in this study we focused on extrusion-based bioprinting there are other works hat use different technique such as inkjet and vat polymerization. For example, inkjet-based bioprinting was used to fabricate biocompatible substrates used for fabricating an electrostimulation device to guide cell alignment and enhance myotubes differentiation (Fortunato et al., 2018).
In this study, we tested multiple commercially available hydrogels characterized by specific composition and rheological capabilities to understand which is the best biomaterial that promotes the formation of a functionalized myo-construct. We did not perform a thorough rheological characterization of the different bioinks used to understand the shear stress experienced by the cells during the printing process (Lucas et al., 2020  Statistically significant values are indicated as *0.05 < P < 0.01 and **P < 0.01. Analysis of variance test was performed to evaluate data significance This could be related to the specific formulation and structure of each biomaterial. The internal structure of the hydrogels is crucial to metabolite transport inside the 3D constructs. Nutrient, oxygen and protein spreading, as well as cell migration and differentiation are supported by diffusion within any matrix with embedded cells .These findings denote a remarkable improvement, as it has been shown that fibrinogen-related biomaterials stimulate cell adhesion, spreading, and differentiation of multiple cell sources including myogenic progenitor cells, especially due to their biodegradable and non-immunogenic features (Almany & Seliktar, 2005;Centola et al., 2013;Fuoco et al., 2012Fuoco et al., , 2014Fuoco et al., , 2015. These characteristics, joint with the hydrogels composed of Fibrinogen/Gelatin allowed the creation of myo-constructs containing myogenic progenitors (C2C12) in precisely defined constructs promoting myotube formation and alignment (Figures 2 and 3). Even if recent studies investigated 3D printing techniques for skeletal muscle tissue engineering (Karande et al., 2004;Mironov et al., 2008), the results achieved were still poor, highlighting an unsatisfactory structural organization both in vitro and in vivo. Conversely, in this paper we showed a significant morphological organization of the myotubes, resembling mature sarcomerogenesis ( Figure 4). Finally, while the use of any of these hydrogels requires further optimization to maximize their functional and myogenic properties, the obtained results provide a knowledge advance in the field and a promising tool for skeletal muscle tissue engineering.

| CONCLUSION
We performed a comparative study of hydrogel behavior testing their myogenic properties over a long-time course (28 days) to analyze how the biomaterial matrix could improve muscle precursor cell (C2C12) viability and differentiation. The linear 3D printed structures were tested in vitro to assess their ability to stimulate myogenesis. Our results clearly showed that CELLINK ® FIBRIN and slightly less CELLINK ® GelXA FIBRIN hydrogels demonstrated the best potential to support the in vitro long-term differentiation of skeletal muscle cells in 3D constructs. After 21-28 days in culture, myogenic cells were able to fuse together forming structurally aligned myotubes, with high expression levels of specific skeletal muscle markers such as Myogenic Differentiation 1 and MCK genes.
Due to all these findings, the results reported herein denote a significant enhancement to improve skeletal muscle tissue engineering.