Development of 3D neuromuscular bioactuators

Abstract

Neuronal control of skeletal muscle bioactuators represents a critical milestone toward the realization of future biohybrid machines that may generate complex motor patterns and autonomously navigate through their environment. Animals achieve these feats using neural networks that generate robust firing patterns and coordinate muscle activity through neuromuscular units. Here, we designed a versatile 3D neuron-muscle co-culture platform to serve as a test-bed for neuromuscular bioactuators. We used our platform in conjunction with microelectrode array electrophysiology to study the roles of synergistic interactions in the co-development of neural networks and muscle tissues. Our platform design enables co-culture of a neuronal cluster with up to four target muscle actuators, as well as quantification of muscle contraction forces. Using engineered muscle tissue targets, we first demonstrated the formation of functional neuromuscular bioactuators. We then investigated possible roles of long-range interactions in neuronal outgrowth patterns and observed preferential outgrowth toward muscles compared to the acellular matrix or fibroblasts, indicating muscle-specific chemotactic cues acting on motor neurons. Next, we showed that co-cultured muscle strips exhibited significantly higher spontaneous contractility as well as improved sarcomere assembly compared to muscles cultured alone. Finally, we performed microelectrode array measurements on neuronal cultures, which revealed that muscle-conditioned medium enhances overall neural firing rates and the emergence of synchronous bursting patterns. Overall, our study illustrates the significance of neuron-muscle cross talk for the in vitro development of neuromuscular bioactuators.

FIG. 1.

3D co-culture in the engineered platform. (a) SEM images of (i) the full platform and (ii) one of the target wells showing the pillar profile. (b) Schematic of the tissue seeding process illustrating (i) target cell-ECM seeding, (ii) compacted target tissues, (iii) neurosphere seeding, and (iv) co-culture tissue. (c) Phase contrast images of a muscle strip from (i) top and (ii) side views. (d) Phase contrast image of the full platform after neurosphere seeding. Scale bars: [(a-i) and (d)] 500 μm and [(a-ii), (c-i), and (c-ii)] 200 μm.

FIG. 2.

Formation of functional neuromuscular junctions. (a) (i) Muscle strips categorized by their contraction pattern in response to optical stimulation of neurons. (ii) Force-time traces of representative samples from groups 1 and 2. The blue rectangle indicates the optical stimulation. (iii) Definitions of contraction force and period. (iv) and (v) Comparison of spontaneous and evoked contraction forces in groups 1 and 2. (vi) Comparison of spontaneous and evoked contraction rates in group 2. The values in panels iv, v, and vi are averaged over the ten contractions immediately preceding and following optical stimulation for spontaneous and evoked contractions, respectively. Box plots represent the 25th, 50th, and 75th percentiles with whiskers representing 1.5×IQR, n = 19 muscle strips for group 1, n = 48 muscle strips for group 2, and ^**p < 0.005 (student's t-test). (b) (i) Brightfield image of a muscle strip and (ii) confocal image of the region outlined in (i) illustrating muscle fibers. (iii) Zoomed views of the region outlined in (ii) showing a muscle fiber with cross-striations. (c) (i) Confocal images illustrating a neurite (β-tubulin III, green) extending toward and making connections with post-synaptic receptor clusters on the muscle (AChR, red). (ii) and (iii) Orthogonal views of the regions indicated by the arrowheads in the rightmost panel in (c-i). Scale bars: (b-i) 100 μm and [(b-ii) and (c)] 10 μm.

FIG. 3.

Neuronal outgrowth toward different targets. Brightfield and corresponding confocal images of representative samples from (a) case 1 and (b) case 2. (c) (i) Confocal image of motor neurons with the outlines in dashed lines illustrating the four different sectors that correspond to the regions between the neurosphere and the different targets and the definition of % outgrowth where “I” refers to the total fluorescence light intensity. (ii) Comparison of relative outgrowth toward the different targets in cases 1 and 2. Values are % outgrowth toward each type of target, and box plots represent the 25th, 50th, and 75th percentiles with whiskers representing 1.5×IQR, n = 5 co-culture samples each for cases 1 and 2, ^**p < 0.005, and ^**p < 0.0005 (student's t-test). All scale bars: 500 μm.

FIG. 4.

Bidirectional cross talk in developing co-cultures. (a) Representative brightfield images and overall time course of the ratio of active vs quiescent muscle strips in (i) co-cultures and (ii) muscle-only cultures. (b) Spontaneous contraction forces in muscle-only and co-culture samples at days 3, 5, and 7. Values are spontaneous contraction force averaged over a 30 s recording per muscle strip at each day, and box plots represent the 25th, 50th, and 75th percentiles with whiskers representing 1.5×IQR, n = 27 muscle strips for co-culture, n = 20 muscle strips for muscle-only at each day, ^*p < 0.05, and ^**p < 0.005 (Mann Whitney U test). (c) Comparison of the percentage of cross-striated muscle fibers between muscle-only and co-culture groups. Bars represent mean ± SD, n = 6 muscle strips for each group, and ^**p < 0.005 (Mann Whitney U test). Confocal images at the right show sample muscle fibers with and without cross-striations. (d) MEA raster plots of representative samples from (i) control and (ii) CM groups at day 9. Black dashed lines represent the individual spikes, blue dashed lines represent the bursts, and pink boxes outline the synchronous bursts. (e) Time evolution of the MEA burst rate of neurons in control and muscle CM groups. Box plots represent the 25th, 50th, and 75th percentiles with whiskers representing 1.5×IQR, the values are average burst rates per electrode over 10 min recording from the entire well, n = 12 wells each for control and CM at each day, ^**p < 0.005, and ^***p < 0.0005 (student's t-test). (f) Conceptual illustration of bidirectional cross talk and its functional outcomes. Scale bars: (a) 500 μm and (c) 10 μm.