Muscle tissue engineering in fibrous gelatin: implications for meat analogs

Bioprocessing applications that derive meat products from animal cell cultures require food-safe culture substrates that support volumetric expansion and maturation of adherent muscle cells. Here we demonstrate scalable production of microfibrous gelatin that supports cultured adherent muscle cells derived from cow and rabbit. As gelatin is a natural component of meat, resulting from collagen denaturation during processing and cooking, our extruded gelatin microfibers recapitulated structural and biochemical features of natural muscle tissues. Using immersion rotary jet spinning, a dry-jet wet-spinning process, we produced gelatin fibers at high rates (~ 100 g/h, dry weight) and, depending on process conditions, we tuned fiber diameters between ~ 1.3 ± 0.1 μm (mean ± SEM) and 8.7 ± 1.4 μm (mean ± SEM), which are comparable to natural collagen fibers. To inhibit fiber degradation during cell culture, we crosslinked them either chemically or by co-spinning gelatin with a microbial crosslinking enzyme. To produce meat analogs, we cultured bovine aortic smooth muscle cells and rabbit skeletal muscle myoblasts in gelatin fiber scaffolds, then used immunohistochemical staining to verify that both cell types attached to gelatin fibers and proliferated in scaffold volumes. Short-length gelatin fibers promoted cell aggregation, whereas long fibers promoted aligned muscle tissue formation. Histology, scanning electron microscopy, and mechanical testing demonstrated that cultured muscle lacked the mature contractile architecture observed in natural muscle but recapitulated some of the structural and mechanical features measured in meat products.


Figure S1
Rheological Mapping of Gelatin Solution a) Characteristic flow diagram, flow curve, and yield stress measurements of gelatin-water 20% (w/w) solutions at 50°C. b) Oscillatory mapping of gelatin-water 20% (w/w) solutions in frequency (constant 10% strain) and strain domains (constant 1 Hz) at 50°C. c) Time sweep of gelatin solutions mixed with MTG crosslinker at 50°C, 1 Hz frequency and under 10% strain. Time sweep begins 30 seconds after initial mixing to due to the time required for mixing and removing loading effects on the sample. Each curve of the same measurement repents independent replicates.

Figure S2
Fourier transform infrared spectroscopy of gelatin powder compared with gelatin fibres. Gelatin powder was porcine type 300A. In all cases, gelatin fibres were spun by immersion rotary jet spinning using a 20 % w/w gelatin precursor solution with gelatin dissolved in DI H2O at 50 °C. Gelatin fibres crosslinked using microbial transglutaminase (green), EDC-NHS (blue), or noncrosslinked (red) all show amide peaks preserved during the spin and crosslinking processes.

Figure S3
Scanning electron microscopy images of decellularized rat hind limb skeletal muscle (A) and microfibrous gelatin (B). Gelatin was spun by immersion rotary jet spinning, using a 20 % w/w gelatin precursor solution (dissolved in DI H2O at 50 °C) spun into a precipitation bath with 70:30 ethanol:water composition Figure S4 Muscle cell adhesion to gelatin fibers. a Brightfield microscopy of sparsely distributed gelatin fibers (left panel), rabbit skeletal muscle myoblast cell (RbSkMC) adhesion to gelatin fibers (middle panels), and bovine aortic smooth muscle cell adhesion to gelatin fibers (right panels); scale bars from left to right are 100 μm, 30 μm, 10 μm, 30 μm, and 10 μm. b Immunohistochemical staining of cytoskeletal (F-actin) and adhesion (vinculin) proteins in RbSkMC attached to gelatin fibers. Cell morphology was consistent with attachment to and alignment with underlying gelatin fibers; scale bars from left to right are 10 μm and 20 μm. c Immunohistochemical staining of cytoskeletal (F-actin) and adhesion (vinculin) proteins in BAOSMC attached to a curved gelatin fiber: The red dotted line in the middle and top-right panels indicates fiber shape. Cell morphology was consistent with attachment to and alignment with the underlying gelatin fiber; scale bar is 10 μm.

Figure S7
Proliferation of rabbit skeletal muscle myoblast cells (RbSkMC) from RbSkMC:gelatin aggregates. The aggregates were assembled in tissue culture polystyrene plates containing dispersed short length (~20 μm) gelatin fibers. After 14 days in culture, the resulting aggregates were transferred by micropipette to fresh flasks to confirm cell viability and proliferation following transfer. These brightfield microscopy images confirmed RbSkMC transferred as aggregates proliferated at rates comparable to non-aggregated counterparts (~daily doubling).

Figure S8
Bovine aortic smooth muscle cell (BAOSMC) tissue formation on a 3D biomimetic fibrous gelatin scaffold. Immunohistochemical staining of filamentous actin (F-actin, Red) and cell nuclei (DAPI, white) confirmed cell confluence and cell alignment, guided by the underlying fibrous gelatin, which is observable as light-grey resulting from gelatin autofluorescence in the DAPI channel. Culture day is 21.

Figure S9
Rabbit skeletal muscle myoblast cell (RbSkMC) tissue formation on a 3D biomimetic fibrous gelatin scaffold. Immunohistochemical staining of filamentous actin (F-actin, Red) and cell nuclei (DAPI, white) confirmed cell confluence and cell alignment, guided by the underlying fibrous gelatin, which is observable as light-grey resulting from gelatin autofluorescence in the DAPI channel. Culture day is 21.

Figure S10
Rabbit skeletal muscle myoblast cell (RbSkMC) tissue formation on a 3D biomimetic fibrous gelatin scaffold. Immunohistochemical staining of filamentous actin (F-actin, Red) and cell nuclei (DAPI, white) confirmed cell confluence and cell alignment, guided by the underlying fibrous gelatin, which is observable as light-grey resulting from gelatin autofluorescence in the DAPI channel. Successive reduction of F-actin channel intensity (from left to right) reveals the underlying gelatin fibres and their influence on cell and cell nuclei anisotropy and alignment. Culture day is 21.

Figure S11
Sample preparation and representative microstructural comparisons of food products. a) Plated samples prepared by 1 cm biopsy punch of several meat products (including bacon, turkey breast, prosciutto, beef tenderloin, rabbit muscle, and ground beef). b) Hematoxylin and eosin (H&E) stains (top two rows) and scanning electron microscopy (SEM, bottom rows) of rabbit skeletal muscle (freshly isolated gracilis muscle from hind limb), bacon, and a processed 'fish ball' product. Scale bars are 200 μm (top row), 50 μm (middle row), and 20 μm (bottom row).

Figure S12
Rheological Mapping of Gelatin Scaffolds and Tissue Engineered Substrates Oscillatory mapping of scaffolds manufactured by spinning gelatin into a) 70% ethanol and b) 80% ethanol in addition to oscillatory mapping of c) rabbit cells seeded onto the scaffolds and d) bovine cells seeded onto the scaffolds. Frequency (at constant 1% strain) and Amplitude sweeps (at constant 1 Hz) occurred at 37°C. Frequency sweeps before heating occurred at 37°C and 70°C after heating (at constant 1% strain). Each curve of the same measurement repeats independent replicates except for the frequency sweep before and after heating where each curve is for either 37°C or 70°C.

Figure S13
Sample preparation and representative force curves obtained by texture profile analysis (TPA). a) Tissue cultured sample with 1 cm diameter hole removed by biopsy punch. b) Cylindrical sample placed between plates in a rheometry system (i) was subjected to compression and shear tests (ii) within an evaporation trap that prevented solvent evaporation (iii). c) Force curves obtained by running two sequential compressions on a tissue cultured sample for TPA. Maximum force during compressions and areas under the curves are used to estimate TPA parameters.

Figure S14
Texture profile analysis (TPA) dual compression curves for fibrous gelatin scaffolds (a, b) and selected food products (c-f). a,b) Fibrous gelatin scaffolds obtained by dissolving 20% w/w gelatin in water and spinning fibers into precipitation baths composed of either 70:30 Ethanol:water (a, G20E70) or 80:20 Ethanol:water (b, G20E80). c-f) TPA curves obtained by compression of bacon (c), prosciutto (d), processed fish balls (e), and smoked turkey breast (f).