Transcriptional Time Course After Rotator Cuff Tear

The effect of tenotomy, neurotomy, and dual injury on mouse rotator cuff muscles: Consequences for the mouse as a preclinical model

A common animal model of muscle pathology following rotator cuff tear (RCT) is a tenotomy of the supraspinatus and infraspinatus, often combined with neurotomy of the suprascapular nerve, which induces a more robust atrophy response than tenotomy alone. However, the utility of this model depends on its similarity to human muscle pathology post-RCT, both in terms of the disease phenotype and mechanisms of muscle atrophy and fatty infiltration. Given the clinical prevalence of nerve injury is low and the muscular response to denervation is distinct from mechanical unloading in other models, an understanding of the biological influence of the nerve injury is critical for interpreting data from this RCT model. We evaluated the individual and combined effect of tenotomy and neurotomy across multiple biological scales, in a robust time-series in the mouse supraspinatus. Muscle composition, histological, and gene expression data related to muscle atrophy, degeneration-regeneration, fatty infiltration, and fibrosis were evaluated. Broadly, we found tenotomy alone caused small, transient changes in these pathological features, which resolved over the course of the study, while neurotomy alone caused a significant fatty atrophy phenotype. The dual injury group had a similar fatty atrophy phenotype to the neurotomy group, though the addition of tenotomy did marginally enhance the fat and connective tissue. Overall, these results suggest the most clinically relevant injury model, tenotomy alone, does not produce a clinically relevant phenotype. The dual injury model partially recapitulates the human condition, but it does so through a nerve injury, which is not well justified clinically.

Defining Endogenous Mitochondrial Transfer in Muscle After Rotator Cuff Injury

BACKGROUND

Rotator cuff muscle degeneration leads to poor clinical outcomes for patients with rotator cuff tears. Fibroadipogenic progenitors (FAPs) are resident muscle stem cells with the ability to differentiate into fibroblasts as well as white and beige adipose tissue. Induction of the beige adipose phenotype in FAPs has been shown to improve muscle quality after rotator cuff tears, but the mechanisms of how FAPs exert their beneficial effects have not been fully elucidated.

PURPOSE

To study the horizontal transfer of mitochondria from FAPs to myogenic cells and examine the effects of β-agonism on this novel process.

STUDY DESIGN

Controlled laboratory study.

METHODS

In mice that had undergone a massive rotator cuff tear, single-cell RNA sequencing was performed on isolated FAPs for genes associated with mitochondrial biogenesis and transfer. Murine FAPs were isolated by fluorescence-activated cell sorting and treated with a β-agonist versus control. FAPs were stained with mitochondrial dyes and cocultured with recipient C2C12 myoblasts, and the rate of transfer was measured after 24 hours by flow cytometry. PdgfraCreERT/MitoTag mice were generated to study the effects of a rotator cuff injury on mitochondrial transfer. PdgfraCreERT/tdTomato mice were likewise generated to perform lineage tracing of PDGFRA+ cells in this injury model. Both populations of transgenic mice underwent tendon transection and denervation surgery, and MitoTag-labeled mitochondria from Pdgfra+ FAPs were visualized by fluorescent microscopy, spinning disk confocal microscopy, and 2-photon microscopy; overall mitochondrial quantity was compared between mice treated with β-agonists and dimethyl sulfoxide.

RESULTS

Single-cell RNA sequencing in mice that underwent rotator cuff tear demonstrated an association between transcriptional markers of adipogenic differentiation and genes associated with mitochondrial biogenesis. In vitro cocultures of murine FAPs with C2C12 cells revealed that treatment of cells with a β-agonist increased mitochondrial transfer compared to control conditions (17.8% ± 9.9% to 99.6% ± 0.13% P < .0001). Rotator cuff injury in PdgfraCreERT/MitoTag mice resulted in a robust increase in MitoTag signal in adjacent myofibers compared with uninjured mice. No accumulation of tdTomato signal from PDGFRA+ cells was seen in injured fibers at 6 weeks after injury, suggesting that FAPs do not fuse with injured muscle fibers but rather contribute their mitochondria.

CONCLUSION

The authors have described a novel process of endogenous mitochondrial transfer that can occur within the injured rotator cuff between FAPs and myogenic cells. This process may be leveraged therapeutically with β-agonist treatment and represents an exciting target for improving translational therapies available for rotator cuff muscle degeneration.

CLINICAL RELEVANCE

Promoting endogenous mitochondrial transfer may represent a novel translational strategy to address muscle degeneration after rotator cuff tears.

Gene Expression in Synovium of Rotator Cuff Tear Patients Determined by RNA Sequencing

Rotator cuff tear (RCT) is a common shoulder disorder related to pain and dysfunction. However, the pathological mechanism of RCT remains unclear. Thus, this study aims to investigate the molecular events in RCT synovium and identify possible target genes and pathways as determined by RNA sequencing (RNA-Seq). The synovial tissue was biopsied from 3 patients with RCT (RCT group) and 3 patients with shoulder instability (Control group) during arthroscopic surgery. Then, differentially expressed (DE) mRNAs, long non-coding RNAs (lncRNAs) and micro RNAs (miRNAs) were comprehensively profiled by RNA-Seq. Gene ontology (GO) enrichment, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway, and competing endogenous RNA (ceRNA) network analysis were performed to identify the potential functions of these DE genes. 447 mRNAs, 103 lncRNAs and 15 miRNAs were identified differentially expressed. The DE mRNAs were highlighted in inflammatory pathway including up-regulated T cell costimulation, positive regulation of T cell activation, and T cell receptor signaling. Down-regulated fatty acid degradation pathway and 5'-AMP-activated protein kinase (AMPK) signaling in RCT group are also enriched. Validation assay showed that the expression of pro-inflammatory molecules including IL21R, CCR5, TNFSF11, and MMP11 was significantly increased in RCT group compared with Control group. CeRNA analysis further revealed lncRNA-miRNA-mRNA regulatory networks involving IL21R and TNFSF11 in RCT. Activated synovial inflammation is the remarkable event of RCT. Importantly, increased T cell activation and disordered fatty acid metabolism signaling might play a significant role. ceRNA networks involving IL21R and TNFSF11 identified could potentially control the progression of RCT. In conclusion, our findings could provide new evidence for the molecular mechanisms of RCT and might identify new therapeutic targets.

Transcriptional time course after rotator cuff repair in 6 month old female rabbits

Introduction: Rotator cuff tears are prevalent in the population above the age of 60. The disease progression leads to muscle atrophy, fibrosis, and fatty infiltration, which is not improved upon with surgical repair, highlighting the need to better understand the underlying biology impairing more favorable outcomes. Methods: In this study, we collected supraspinatus muscle tissue from 6 month old female rabbits who had undergone unilateral tenotomy for 8 weeks at 1, 2, 4, or 8 weeks post-repair (n = 4/group). RNA sequencing and enrichment analyses were performed to identify a transcriptional timeline of rotator cuff muscle adaptations and related morphological sequelae. Results: There were differentially expressed (DE) genes at 1 (819 up/210 down), 2 (776/120), and 4 (63/27) weeks post-repair, with none at 8 week post-repair. Of the time points with DE genes, there were 1092 unique DE genes and 442 shared genes, highlighting that there are changing processes in the muscle at each time point. Broadly, 1-week post-repair differentially expressed genes were significantly enriched in pathways of metabolism and energetic activity, binding, and regulation. Many were also significantly enriched at 2 weeks, with the addition of NIF/NF-kappaB signaling, transcription in response to hypoxia, and mRNA stability alongside many additional pathways. There was also a shift in transcriptional activity at 4 weeks post-repair with significantly enriched pathways for lipids, hormones, apoptosis, and cytokine activity, despite an overall decrease in the number of differentially expressed genes. At 8 weeks post-repair there were no DE genes when compared to control. These transcriptional profiles were correlated with the histological findings of increased fat, degeneration, and fibrosis. Specifically, correlated gene sets were enriched for fatty acid metabolism, TGF-B-related, and other pathways. Discussion: This study identifies the timeline of transcriptional changes in muscle after RC repair, which by itself, does not induce a growth/regenerative response as desired. Instead, it is predominately related to metabolism/energetics changes at 1 week post-repair, unclear or asynchronous transcriptional diversity at 2 weeks post-repair, increased adipogenesis at 4 weeks post-repair, and a low transcriptional steady state or a dysregulated stress response at 8 weeks post-repair.

Spatial transcriptomics tools allow for regional exploration of heterogeneous muscle pathology in the pre-clinical rabbit model of rotator cuff tear

Background

Conditions affecting skeletal muscle, such as chronic rotator cuff tears, low back pain, dystrophies, and many others, often share changes in muscle phenotype: intramuscular adipose and fibrotic tissue increase while contractile tissue is lost. The underlying changes in cell populations and cell ratios observed with these phenotypic changes complicate the interpretation of tissue-level transcriptional data. Novel single-cell transcriptomics has limited capacity to address this problem because muscle fibers are too long to be engulfed in single-cell droplets and single nuclei transcriptomics are complicated by muscle fibers’ multinucleation. Therefore, the goal of this project was to evaluate the potential and challenges of a spatial transcriptomics technology to add dimensionality to transcriptional data in an attempt to better understand regional cellular activity in heterogeneous skeletal muscle tissue.

Methods

The 3′ Visium spatial transcriptomics technology was applied to muscle tissue of a rabbit model of rotator cuff tear. Healthy control and tissue collected at 2 and 16 weeks after tenotomy was utilized and freshly snap frozen tissue was compared with tissue stored for over 6 years to evaluate whether this technology is retrospectively useful in previously acquired tissues. Transcriptional information was overlayed with standard hematoxylin and eosin (H&E) stains of the exact same histological sections.

Results

Sequencing saturation and number of genes detected was not affected by sample storage duration. Unbiased clustering matched the underlying tissue type-based on H&E assessment. Connective-tissue-rich areas presented with lower unique molecular identifier counts are compared with muscle fibers even though tissue permeabilization was standardized across the section. A qualitative analysis of resulting datasets revealed heterogeneous fiber degeneration–regeneration after tenotomy based on (neonatal) myosin heavy chain 8 detection and associated differentially expressed gene analysis.

Conclusions

This protocol can be used in skeletal muscle to explore spatial transcriptional patterns and confidently relate them to the underlying histology, even for tissues that have been stored for up to 6 years. Using this protocol, there is potential for novel transcriptional pathway discovery in longitudinal studies since the transcriptional information is unbiased by muscle composition and cell type changes.