Factors Affecting Glucose Uptake in Tissue-engineered Human Skeletal Muscle

Tissue-engineered skeletal muscle myobundles offer a promising approach for developing a human in vitro model of healthy and diseased muscle for drug development and testing. Their three-dimensional structure offers a better model of the organization of native skeletal muscle than monolayer culture does, and their amenability to an array of functional measures provides a multifaceted account of the tissue’s health. One such functional measure is the metabolic state of the myobundle, which in an in vitro model of healthy skeletal muscle should reflect the metabolism of native muscle. However, skeletal muscle cultured in vitro is exposed to artificially high levels of nutrients meant to promote cell growth, and it exhibits altered glucose uptake, with high rates of glycolysis and a dampened insulin response. Inflammation is closely linked with skeletal muscle metabolic dysfunction, and the field could benefit from a human tissue-engineered myobundle model of inflammation to elucidate mechanisms of disease progression and to examine drug safety and efficacy in inflamed muscle.

We first characterized the glucose uptake and insulin response of tissue-engineered myobundles in the basal state and in response to treatment with metformin and an HDAC inhibitor. We then imparted greater physiological relevance to the system via altered culture conditions and examined the impact on myobundle metabolic and contractile function. Finally, we validated the ability of pro-inflammatory cytokine exposed-myobundles to recapitulate key aspects of inflammation-mediated skeletal muscle dysfunction.

We found that myobundles exhibit insulin sensitivity similar to that of in vivo skeletal muscle, but insulin responsiveness is substantially lower, in accordance with other in vitro studies. Metformin treatment stimulated a robust increase in basal glucose uptake, and treatment with the HDAC inhibitor 4-PBA enhanced myobundle contractile function and insulin responsiveness. We showed that altering myobundle culture conditions to provide more physiologic nutrient availability was sufficient to metabolically reprogram the myobundles to a less glycolytic state. To model an inflammation-mediated disease state, we exposed myobundles to pro-inflammatory cytokines and found that the inflamed myobundles exhibited contractile dysfunction, a robust secretion of pro-inflammatory cytokines, and increased basal glucose uptake while still retaining a functional response to metformin treatment. Overall, this work validates the suitability of a tissue-engineered skeletal muscle model for detecting metabolic perturbations mediated by drug treatment, nutrient availability, and inflammatory conditions.