Poster 120: Understanding the Biological Basis of Contracture Using a Mouse Model of Single Hindlimb Immobilization-Induced Fibrosis

OBJECTIVES: Arthrofibrosis is characterized by excessive collagen production and adhesions that result in restricted joint motion (stiffness) and pain. The rate of procedural intervention (i.e. manipulation under anesthesia or surgical contracture release) ranges from 0.7% to 35% after anterior cruciate ligament (ACL) reconstruction surgery, and up to 57% after a knee dislocation event. Current treatments are limited to early range of motion after injury, physical therapy, dynamic splinting, manipulation under anesthesia and surgical contracture release, but at present there are no non-surgical therapeutic interventions that can reliably reduce or prevent joint stiffness. To better understand the mechanisms underlying the development of knee stiffness, we have developed a novel mouse model of joint contracture formation based on single hindlimb immobilization using 3D-printed clamshell casts. Previous work has demonstrated that single hindlimb immobilization produces contractures of the knee and ankle that vary in reversibility depending on age and duration of immobilization. The current study seeks to understand the biological mechanisms that lead to hindlimb immobilization-induced contracture formation through analysis of gene expression and histopathology. METHODS: Female and male C57BL/6J mice aged 14-19 weeks underwent single hindlimb immobilization (SHLI) using a custom, 3D-printed clamshell cast for 2 or 3 weeks prior to release (see Figure 1 for study design, n=4 each gender, each time point). Casts were changed and knee motion was assessed weekly using a custom 3D-printed motion measurement system. Mice were sacrificed on the day of release, 4 days after release or 3 weeks after release (3-week immobilization only) and tissues were harvested for histology and immunohistochemistry (IHC). Mouse knee joints were fixed in 4% formaldehyde, decalcified in EDTA and embedded in paraffin for sectioning. Serial 7mm sections were stained with hematoxylin and eosin (H&E), Masson’s Trichrome, Picrosirius Red or Safranin O/Fast Green for histological analyses. Sections were additionally analyzed using immunohistochemistry to identify markers of myofibroblasts (α-smooth muscle actin, αSMA) and macrophages (F4/80). For gene expression analysis, the anterior fatpad and adjacent synovium was harvested from mice immobilized for 3 weeks followed by 4 days release (n=9 male, n=9 female). Tissues were immediately placed in RNAlater and high-quality RNA (RIN>7, 260/280>1.8) was isolated using established protocols. Changes in gene expression were analyzed using the NanoString mouse Fibrosis panel and confirmed with RTqPCR. Histological quantification was done using QuPath, and statistics were completed in Graphpad Prism. RESULTS: Single hindlimb immobilization with the 3D-printed clamshell cast for 3 weeks produces sustained contracture formation in both male and female mice, with an average knee motion loss of 31.9 ± 6.5 degrees for male mice and 37.3 ± 15.6 degrees for female mice, consistent with previously reported results in female mice. There was deposition of excess extracellular matrix, including collagen, in the anterior fat pad and posterior knee capsule in the immobilized joints. These changes were present in both male and female mice and were more pronounced at 3 weeks after immobilization compared to the 2-weeks immobilized group. Mice immobilized for 3-weeks showed more fibrosis after 4 days of free cage motion compared to the day of release. Additionally, we identified increases in αSMA and F4/80- immunopositive cells in the immobilized limbs after 3 weeks of casting and 4 days of release. Analysis of mice immobilized for 3 weeks followed by free cage motion for 3 weeks demonstrated sustained contracture formation with persistent periarticular fibrosis. NanoString gene expression analyses identified more than 230 differentially expressed genes between immobilized and control / contralateral limbs. Specifically, pathway analyses of differentially expressed genes uncovered a robust increase in genes involved in collagen biosynthesis and modification, extracellular matrix synthesis (ECM), ECM degradation, transforming growth factor beta (TGFβ) signaling and epithelial mesenchymal transition (EMT) signaling pathways. Differential expression of selected genes was confirmed using RTqPCR analyses. CONCLUSIONS: Here we have provided an in-depth characterization of the histologic and gene expression changes that occur with mouse single SHLI. Our work demonstrates that mouse SHLI provides a simple and effective preclinical model of immobilization-based knee contracture and peri-articular fibrosis. This model is reproducible, cost effective, permits analyses of the early events that lead to joint fibrosis and stiffness, can be readily employed with transgenic animals, and will allow evaluation of therapeutic interventions. Similar to the human phenotype, we found fibrosis of anterior knee structures including the infrapatellar fat pad, as well as increases in αSMA and F4/80 positive immunostaining, suggesting an increase in myofibroblasts and macrophages, respectively. Importantly, there is consistency between our model and RNA sequencing studies of post-surgical arthrofibrosis in human subjects, which includes upregulation of collagen synthesis and modification, ECM synthesis and degradation and TGF-β signaling. We also observe more fibrosis after free cage motion after release from immobilization, suggesting there may be an opportunity for early intervention. Importantly, our study has identified several potential therapeutic targets involved in the development of joint fibrosis. Future studies should aim to test preventative therapies that specifically target these signaling pathways using our non- surgical model of SHLI-driven fibrosis.