Fresh whole porcine hearts were acquired from Nebraska Scientific (Omaha, NE) and cut down to sections not exceeding 2 mm³. The tissues were then placed in 2.5 L of a 1% SDS solution for 48 h, stirred continuously. The solution was replaced at 24 h, and it was ensured no agglomerations had formed. The cardiac tissue was then treated with 1% Triton X-100 for 30 min followed by 3 days of PBS washes, where the buffer was replaced every day, and the tissues are stirred continuously. To sterilize the decellularized tissue, it was treated with 1% peracetic acid for 4 h and washed five times with sterile PBS and sterile distilled water. All decellularization steps were done at room temperature. The strained tissue was then frozen at -80°C overnight and lyophilized. All decellularization techniques followed were taken from Pati et al. with minimal modification (Pati et al., 2014).

Human Aortic Smooth Muscle Cells were grown in vascular growth media (ATCC PCS-100-030) plus smooth muscle kit (ATCC PCS-100-042) under normal environmental culture conditions at 37°C with 5% CO₂ in a fully aseptic environment. Before printing, cell membranes were stained using PKH26 red fluorescent live cell membrane stain (Sigma MINI26-1KT).

The lyophilized dECM was digested and re-suspended in order to form a printable gel material. Lyophilized dECM was mechanically pulverized using burr grinding methods, then 3 g of dECM were digested with 30 mg of pepsin from porcine gastric mucosa (Sigma P7125-100G, ≥3,200 units/mg protein) in 100 ml of 0.5 M acetic acid for 48 h at room temperature with continuous stirring (Pati et al., 2014). Centrifugation at 500 x g for 10 min prior to pH adjustment was done to allow separation of any particles that were not solubilized; this allowed for a homogenous texture when printing. Ice cold 10M NaOH was used to adjust the pH of the dECM solution to the physiological range; while the dECM was kept below 10°C to avoid premature crosslinking of the solution. A Bio X bioprinter (Cellink, Boston, MA) equipped with a cooling printhead and an 18-gauge blunt needle was used for all printed samples. Printhead temperature was maintained at 14°C to avoid premature crosslinking of the bioink, print bed temperature was kept at 30°C to help begin the crosslinking process prior to the 1-hour incubation period to achieve full gelation. A print stabilizing solution of 1% Xanthium Gum was used to ensure the stability of the print prior to gelation, a modification of the technique described in Noor et al. (2019). For the cell-laden bio printed structure a cell suspension of 1 × 10 (Limperopoulos et al., 2002) cells/mL was mixed with the solubilized, pH-adjusted dECM immediately prior to bioprinting. Printhead and print bed temperatures were maintained at 14 and 30°C, respectively.

Autodesk Fusion 360 was used for Computer-Aided Design (CAD). The tensile testing shape was drawn to ASTM D412- Die A for thermoplastic elastomers, with minor modifications due to the biological nature of the samples. Dimensions for the aorta construct were taken from previous literature (Walther et al., 1985; Jarvisalo et al., 2001; Skilton et al., 2005; Taketazu et al., 2017).

The tissue samples were processed using a Thermo Scientific Spin Tissue Processor Microm STP-120. Tissues were fixed using 4% formalin at 4°C overnight, followed by dehydration. Samples were dehydrated in increasing ethanol concentrations (starting with 70, 95, and 100%), followed by immersion in xylene two times and paraffin infiltration. Paraffin-embedded tissues were sectioned at 4–6 μm using a Shandon Finesse ^® E/ME microtome. All samples were deparaffinized in three xylene washes then rehydrated in decreasing ethanol concentrations (100, 95, 70, and 50%), with each wash lasting 3 min. Hematoxylin and eosin (H&E) staining was done post-deparaffinization. Following rehydration, the samples were exposed to hematoxylin solution for 2 min, then rinsed with running water for 1-minute, distilled water for 30 s, and 95% alcohol for an additional 30 s. Immediately after, the samples were exposed to eosin solution for 1 min, then washed with increasing amounts of alcohol (95% x2, 100% x2) for 2 min each, then two xylene washes each for 2 min. Trichrome staining was done using Masson’s Trichrome Staining Kit; all kit procedures were precisely followed. All samples were mounted using Cytoseal™60 (Thermo Scientific REF#8310-4).

For scanning electron microscopy imaging, samples were fixed using 4% formalin and dehydrated in increasing ethanol concentrations prior to lyophilization. All samples were sputter-coated for 20 s and viewed using a Hitachi S-4800-II SEM. A Nikon AZ100 inverted light microscope was used for histological imaging.

Uniaxial tensile testing was done at room temperature in order to determine the mechanical properties of native tissue and decellularized printed tissue. One group of samples was taken from a fresh whole porcine heart; rectangles were cut vertically along the sagittal plane. Each sample’s thickness was determined by measurements using a digital caliper. The second sample group consisted of printed decellularized ECM made to ASTM standard D638—Type I and also measured using a digital caliper. All specimens were patted dry, fixed to grips, and placed in the uniaxial electromechanical tensile testing machine (MTESTQuattro), where they were subjected to a strain rate of 5 mm min⁻¹ at room temperature until rupture. Forces were measured by a 200 lb. load cell and elongation by the internal sensor as well as by displacement of the reference grid.

The stress-strain relationships for the samples tested were obtained using the Mooney-Rivlin Model for incompressible hyperelastic materials (Peattie et al., 1996; Rubin and Krempl, 2010; Ruiz de Galarreta et al., 2016). Where: T e n s i l e   s t r e s s   σ = F A o (1) S t r e t c h   r a t i o   λ = Δ L L 0 − 1 (2)

Because the mechanical testing conducted was a uniaxial tensile test: σ 1   1 − σ 3   3 = 2 C 1 ( λ 2 − 1 λ ) − 2 C 2 ( 1 λ 2 − λ ) (3)

And: σ 2   2 − σ 3   3 = 0 (4)

In this case, the model can be further simplified under the assumption of simple tension making: σ 2   2 = σ 3   3 = 0 (5)

Therefore: σ 1   1 = ( 2 C 1 + 2 C 2 λ ) ( λ 2 − 1 λ ) (6)

This data was then plotted to find the predicted stress-stretch ratio diagrams. Young’s modulus was derived from the slope at the linear region of the graph. Poisson’s ratio was determined by taking the transversal elongation and dividing by axial compression.

Chemical decellularization of the cardiac tissue resulted in visual, physical changes to the structure and state of the tissue sections (Figures 1A–F). When subjected to temperatures of 37°C post enzymatic digestion the decellularized ECM showed three different stages of crosslinking. At a pH of 3 it was a low viscosity fluid, transitioned to being highly viscous at a pH of 5, and finally crosslinked successfully at a pH of 7 (Figures 1G–I). Hematoxylin and eosin staining of the cardiac tissue pre-decellularization confirmed normal morphology of the cellular structures, no nuclear irregularity or clumping of the chromatin to indicate cancerous areas. Staining with Masson’s Trichrome showed collagenous areas throughout the section; a normal extracellular matrix was visible and necessary for future experiments’ success. Once chemical decellularization, enzymatic digestion, and gelation of the decellularized ECM was completed, the stains were repeated to determine the presence of cellular matter and collagen. Staining showed a lack of cellular structures and nuclear matter; collagen fibrils are uniformly seen throughout the decellularized sample (see Figure 2). Randomly oriented collagen fibrils can be seen in the decellularized ECM’s dense network, whereas the printed decellularized ECM’s topography was similar to that of the porcine cardiac tissue (see Figure 3).

Video stills of the tensile testing allowed for precise comparison of the change in length (ΔL) and change in width (ΔH) (Figure 4) in addition to the information gathered for the stress-strain ratio curves. Due to the samples’ large deformations when undergoing forces, a Mooney-Rivlin model was used to find the samples’ effective stresses (Figure 5). The response of the cardiac tissue (Figures 5A, C) indicates an overall higher tensile strength and greater plasticity than that of the printed decellularized ECM (Figures 5B, D). Young’s Modulus and Poisson’s Ratio for the individual samples are shown in Table 1. The printed decellularized ECM (Ē = 920 kPa) exhibited 2.7 times greater elasticity than the porcine cardiac tissue (Ē = 330 kPa). Similarly, the porcine cardiac tissue displayed 1.8 times the deformation ( ν ¯ = 3.93 ) prior to failure than the printed decellularized ECM ( ν ¯ = 2.13 ) . Neither value showed a significant (p < 0.05) difference between the two sample types.

Due to the highly viscous nature of the bioink used a larger gauge needle was selected for bioprinting. The dog bone and cylindrical structures were printed without deformation issues or bleeding of the bioink into the support bath (Figure 6). The printed aortic structures maintained their original shape and dimensions post crosslinking, as did the tensile testing samples. Confocal imaging of the bioprinted HASMC indicates normal cell attachment and morphology 24 h post-print, with no necrotic areas as seen in Figure 6D.