Generating Biomaterials from Agricultural Waste: Decellularized Plant-based Scaffolds for Tissue Engineering

Tissue engineering has become an integral part of the bioproduction economy, with advances in both regenerative medicine, and potential production of consumer goods via cellular agriculture. However, the field is still mired in commercialization challenges, as high costs for culture materials, media components, and biomaterial scaffolds limit the scalability of tissue engineered products. In order to be market-competitive and produce tissue engineered products at-scale, production costs must be lowered.

Our laboratory has demonstrated use of decellularized plants as low-cost scaffolds for mammalian cell growth. We have established that bovine satellite cells can grow and differentiate readily on decellularized spinach-derived scaffolds, and used broccoli florets as a microcarrier for suspension culture within a bioreactor. Plant-derived scaffolds are edible, and primarily cellulose, a well-established biomaterial. However, these studies largely employ agricultural plants, fruits, and vegetables as starting materials for scaffolds. These foodstuffs have an implicit market value on their own, and therefore may be costly to employ at-scale as biomaterial scaffolds.

Agricultural harvest waste like corn husk or rice hulls present an alternative, low cost, and highly-abundant scaffold source, with an estimated 3.8 billion metric tons of crop residue generated per year. Repurposing these wastes represents a significant opportunity to add value back to conventional agricultural crops within the framework of a bioproduction economy. In this work, we harvest wastes - corn husks and jackfruit rind - as scaffolds for muscle tissue bioproduction.

Materials and Methods: Fibrous jackfruit “rind” (white pulp) was separated from the edible pulp surrounding seeds. Corn husk was decellularized via submersion in 1% sodium dodecyl sulfate (SDS) in DI water for 5 days, followed by 2 days in 1% Polysorbate-20 and 3% bleach, followed by 1 day in DI water. Jackfruit was similarly processed under 10% SDS/3% Polysorbate-20/10% bleach for 48 hours. Solutions were exchanged daily. Cyquant and histological analysis (Safranin O/Fast Green) was used to validate DNA removal and effective decellularization. To evaluate scaffold stiffness, ~6 cm samples of each scaffold (both normal and decellularized) was loaded into an Instron uniaxial tensile testing device and pulled at 10mm/min, and young’s modulus quantified.

For cell culture, ~1cm² scaffolds were prepared and sterilized in ethanol for 30 mins, followed by a 1-hour acclimation in serum-containing culture medium. Primary bovine satellite cells (PBSCs) and a quail myoblast cell line (QM7s) were trypsinized and 2mL of cell solution was seeded into 15mL conical tubes at 10⁶ cells/mL, containing 100mg of plant scaffolds. For cell and scaffold visualization and protein content analysis, samples were maintained in growth (10% serum) media for 72 hours, and were then maintained for 1 week in low-serum (2%) differentiation media. Samples were removed and either processed for protein analysis using a standard BCA assay, or fixed in 4% PFA, stained with DAPI/phallodin, and imaged using an inverted microscope.

Results, Conclusions, and Discussions: Decellularization was effective for both scaffold types, showing visible loss of color (Figure 1a). Corn husk and jackfruit native samples contained 9.41+/-4.46 and 4.00+/1.64 ug DNA/g tissue respectively, and decellularized corn husk jackfruit fibers 0.17+/-0.06 and 0.07+/-0.00ug DNA/g tissue respectively. Scaffold stiffness was reduced by decellularization: Corn husk is an anisotropic material, and stiffness was reduced post-decellularization when pulled parallel to the aligned vasculature (56.67 +/- 16.71 MPa vs 12.95 +/- 2.43 MPa before/after decellularization) but not in the perpendicular direction. Jackfruit stiffness was not significantly reduced (Figure 1b).

Culture media-conditioned, non-cell seeded jackfruit samples had an average of 0.61+/-0.35 ug/ul lysate protein per gram of tissue mass while PBSC-seeded and QM7-seeded samples had 11.45 +/-2.24 and 12.90+/-1.99 ug/ul lysate/gram respectively (Figure 1c, 1e). There was no significant difference between the protein detected in jackfruit samples seeded with QM7 cells and PBSCs. PBSC-seeded corn husk protein content was not significantly different to corn husk conditioned in culture media. However, corn husk seeded with QM7s had a measurable increase in protein content (16.28 +/- 3.55 ug/ul lysate/gram) to both samples seeded with PBSCs, and media-conditioned samples (9.57 +/- 1.56 and 6.35 +/- 1.43 ug/ul lysate/gram, respectively).

Low stiffness (closer to native muscle tissue at ~8kPa) and higher protein content suggests jackfruit fiber may represent more viable scaffold for tissue engineering, but both plants represent attractive candidates for low-cost scale-up of bioproduction. Ongoing work is targeted at refining scaffold preparation to better mimic mechanical properties of native muscle tissue, and exploring other harvest wastes as scaffold candidates.


Acknowledgements: The authors thank Bret Judson and the BC Imaging Core for support with this project.