The role of loading in murine models of rotator cuff disease

Abstract

Rotator cuff disease pathogenesis is associated with intrinsic (e.g., age, joint laxity, muscle weakness) and extrinsic (e.g., mechanical load, fatigue) factors that lead to chronic degeneration of the cuff tissues. However, etiological studies are difficult to perform in patients due to the long duration of disease onset and progression. Therefore, the purpose of this study was to determine the effects of altered joint loading on the rotator cuff. Mice were subjected to one of three load-dependent rotator cuff tendinopathy models: underuse loading, achieved by injecting botulinum toxin-A into the supraspinatus muscle; overuse loading, achieved using downhill treadmill running; destabilization loading, achieved by surgical excision of the infraspinatus tendon. All models were compared to cage activity animals. Whole joint function was assessed longitudinally using gait analysis. Tissue-scale structure and function were determined using microCT, tensile testing, and histology. The molecular response of the supraspinatus tendon and enthesis was determined by measuring the expression of 84 wound healing-associated genes. Underuse and destabilization altered forepaw weight-bearing, decreased tendon-to-bone attachment strength, decreased mineral density of the humeral epiphysis, and reduced tendon strength. Transcriptional activity of the underuse group returned to baseline levels by 4 weeks, while destabilization had significant upregulation of inflammation, growth factors, and extracellular matrix remodeling genes. Surprisingly, overuse activity caused changes in walking patterns, increased tendon stiffness, and primarily suppressed expression of wound healing-related genes. In summary, the tendinopathy models demonstrated how divergent muscle loading can result in clinically relevant alterations in rotator cuff structure, function, and gene expression.

Keywords: inflammation; injury; mechanical loading; tendinopathy; tendon.

Figure 1:. Study Design.

Treatment groups, time points, experimental assays, and samples sizes. The same limb from an individual animal was used for bone morphometry and biomechanics. Contralateral limbs were used for either histology or gene expression. Both sexes were included in the study with gait analysis, biomechanics, and bone morphometry containing 10 males, 5 female per group, histology 3 males and 2 females pre group, and gene expression had 3 males and 3 females per group.

Figure 2:. Longitudinal Gait analysis.

A) Mouse gait was captured and digitized by an automated system (CatWalk XT, Noldus). Examples of the functional parameters Forepaw Stance Width and Contact Area are indicated on a resulting gait tracking image. B) All treatments exhibited transient changes in gait, however, returned to baseline values by 4 weeks (n = 14-15/group). *P < 0.05, **P < 0.01 – significantly different from baseline within group. #P < 0.05 – significantly different from Cage Activity at matched time point. Data shown mean ± standard deviation.

Figure 3:

Biomechanics. (A) Cross-sectional area was determined by μCT analysis and used to determine material properties from quasistatic pull to failure testing of the tendon to bone attachment (n = 14–15/group). Dotted line represents the mean value of Cage Activity. *p < 0.05, **p < 0.01 significantly different from Cage Activity. Data shows mean ± standard deviation. (B) Positive Toluidine blue staining indicates the presence of proteoglycans (n = 4–5/group)

Figure 4:. Bone morphometry.

A) μCT analysis of the epiphysis (n = 14-15/group). Dotted line represents mean value of Cage Activity. *P < 0.05, **P < 0.01 significantly different from Cage Activity. Data shown mean ± standard deviation. B) Density-weighted images of the epiphysis in the coronal and transverse planes for an individual sample with corresponding BV/TV.

Figure 5:. Gene Expression.

Heatmap shows the mean relative expression normalized to the Cage Activity control group (n = 6/group; 3 females, 3 males). Functional sub-groups are indicated by brackets. *P < 0.05 compared to Cage Activity.