Tissue-Engineered Human Skeletal Muscle Model of Rheumatoid Arthritis for Disease Modeling and Drug Testing

Tissue-engineered skeletal muscle can be designed and optimized to serve as a platform for disease modeling and drug testing. In vitro models such as these can be used to explore basic research questions that may be difficult to study in vivo. To achieve this, it is important that the engineered skeletal muscle mimic its in vivo equivalent both phenotypically and functionally. Our engineered human skeletal muscle constructs (myobundles) generate quantitative contractile forces in response to electrical stimulation. The 3D myobundles provide a more realistic in vitro environment than that experienced by cells cultured in 2D monolayers.

The overall goal of this work was to develop an in vitro myobundle model of rheumatoid arthritis (RA), a chronic inflammatory disorder, to (1) characterize muscle function of RA patients, (2) further our understanding of the underlying disease mechanisms, and (3) test potential therapeutics that can reduce muscle damage and loss in RA. We first characterized myobundles made with cells from the vastus lateralis muscle of RA patients and aged healthy donors, as well as from the hamstring muscle of young healthy donors as a benchmark. Next, we investigated RA myobundle sensitivity to pro-inflammatory cytokine exposure, compared to aged healthy controls. Finally, we evaluated the effect of pharmacologic agents on functional recovery of RA myobundles.

Surprisingly, in 3D culture, contractile force production by RA myobundles was greater compared to aged controls. In support of this finding, assessment of RA myofiber maturation showed increased area of sarcomeric α-actinin expression over time compared to aged controls. Furthermore, a linear regression test indicated a positive correlation between sarcomeric α-actinin protein levels and tetanus force production in RA and controls. Our findings suggest that medications prescribed to RA patients may maintain—or even enhance—muscle function, and this effect is retained and observed in in vitro culture.

We demonstrated that RA myobundles were more sensitive to IFN-γ treatment leading to reduced contractile force and reduced contractile protein levels compared to aged healthy controls. RNA sequencing (RNA-seq) and gene set enrichment analysis (GSEA) was performed to identify pathways associated with altered gene expression. Gene sets that were enriched in IFN-γ-treated RA myobundles, but not IFN-γ-treated controls, were genes upregulated in response to hypoxia and genes upregulated during unfolded protein response. From the hypoxia gene set, Pim1 and MT-1 were identified as potential therapeutic targets for treating RA-associated muscle dysfunction. Furthermore, we showed that treatment with tofacitinib fully restores contractile force and contractile protein levels in IFN-γ-treated RA myobundles.

To our knowledge, this represents the first study to use tissue-engineered human muscle to characterize muscle function of RA patients. Our in vitro RA myobundle model enables us to (1) model key variables affecting the progression of RA and (2) serve as a platform for pharmaceutical testing allowing for ineffective drugs to be quickly identified. Since chronic inflammation and muscle loss play a role in other diseases such as osteoarthritis, sarcopenia, and cachexia in heart failure and cancer, this work serves as a proof-of-principle for modeling and treating inflammation and fibrosis of muscle.