Osteoarthritic Muscle Shows a Blunted Extracellular Matrix Response After Exercise
Knee osteoarthritis is usually framed as a cartilage disease, but the muscle around the joint carries much of the functional burden. In this preprint, investigators examined vastus lateralis tissue in adults with and without moderate-to-severe knee OA and found a pattern that reaches beyond weakness alone: more fibrillar muscle collagen, fewer satellite cells, more PDGFRα+ mesenchymal cells, and a muted protein response in the enzymes that help remodel extracellular matrix after resistance exercise.
The central clinical signal is straightforward. Adults with OA had weaker knee extensors, poorer muscle quality, and a higher proportion of fibrillar collagen than controls, with collagen content averaging 12.2 ± 4.8% versus 8.4 ± 2.9% in the comparison group. Fibrillar collagen also tracked inversely with isokinetic concentric torque, a relationship that fits the mechanical logic of a muscle becoming less efficient as noncontractile material accumulates.
Collagen Accumulation Was Tied to Lower Torque and Smaller Fibers
The tissue findings give that weakness a plausible structural basis. Volunteers with knee OA had smaller myofibres and more irregular fiber shape, both of which point to a muscle compartment under chronic remodeling stress rather than simple disuse. Thigh lean mass, capillary content, and fiber-type distribution did not differ, which narrows the interpretation: the deficit was not explained by a gross loss of limb mass or obvious shifts in fiber type.
The collagen story matters because extracellular matrix is not inert scaffolding. In skeletal muscle, excessive fibrillar collagen can stiffen the tissue, alter force transmission, and interfere with the regenerative niche. When the matrix becomes overly dense, contractile tissue has to work against a thicker mechanical barrier, and that can show up at the bedside as diminished torque long before a patient notices a dramatic change in limb size.
PDGFRα+ Cells and Satellite Cells Point to a Disturbed Repair Niche
Cellular composition in the OA muscle looked different as well. The OA group had more PDGFRα+ mesenchymal cells and fewer satellite cells than controls. Those two compartments are relevant because both participate in matrix biology, and both can shape how muscle responds to injury and loading. Satellite cells support regeneration and repair, while PDGFRα+ mesenchymal cells are increasingly recognized as contributors to fibrotic remodeling and collagen deposition.
In practical terms, that combination suggests a repair environment that may be tilted toward scar-like accumulation rather than balanced renewal. A muscle with fewer satellite cells and more matrix-associated mesenchymal cells may still respond to exercise, but the response could favor stiffness and persistence of collagen rather than efficient turnover. The finding offers a cellular explanation for why knee OA can behave like a disorder of both joint mechanics and muscle biology.
Resistance Exercise Activated MMP Genes, But Protein Remodeling Lagged in OA
The exercise challenge revealed an important split between transcription and protein-level remodeling. In both groups, resistance exercise upregulated MMP2, MMP9, and MMP14 gene expression, and it increased MMP2 and MMP9 enzyme activity. On paper, that looks like a competent remodeling response to mechanical load.
The protein signal was less reassuring. MMP2 and MMP9 protein area fraction rose after exercise in controls, but this response was blunted in the knee OA group, with an age-by-time interaction of p = 0.041. Satellite cell-MMP9 colocalization also changed differently between groups after exercise. The biology here is important: exercise can switch on the machinery for matrix turnover, yet the OA muscle appears less able to convert that signal into a robust extracellular matrix protein response.
This is the sharpest divergence in the dataset. The genes responded, the enzymes responded, but the tissue-level protein remodeling response did not fully follow in OA.
Why the Disconnect Between MMP Signals and Matrix Protein Response Matters
Matrix metalloproteinases are central to collagen turnover, but their biology is tightly regulated at transcriptional, translational, and post-translational levels. A rise in mRNA does not guarantee proportional protein accumulation, and enzyme activity does not always translate into organized matrix degradation or replacement. In a muscle already carrying excess fibrillar collagen, a blunted protein response to loading could permit the matrix to remain abnormally dense even after a stimulus that should promote renewal.
That mechanism fits the clinical phenotype. Weakness in knee OA may not be caused only by pain-related underuse or generalized deconditioning. A dysfunctional extracellular matrix, combined with altered cellular composition, could create a muscle that is mechanically stiff, biologically sluggish, and less adaptable to rehabilitation loading.
What This Preprint Adds to Knee OA Care
The study expands the conversation around osteoarthritis rehabilitation. Resistance exercise remains biologically active in OA muscle, and the transcriptional and enzyme activity data show that the tissue still senses loading. Yet the impaired protein response suggests that standard strengthening may not fully normalize muscle remodeling in some patients with OA, especially those with prominent collagen accumulation.
For clinicians, the immediate relevance is conceptual rather than prescriptive. The knee extensor deficit in OA may reflect a deeper extracellular matrix problem that deserves attention alongside cartilage symptoms, pain control, and mobility training. For researchers, the next step is to determine whether interventions can restore the protein-level matrix response and whether that restoration improves strength, function, or both.
Because this is a preprint, the findings still need peer review and replication. The cohort was also limited to adults with and without moderate-to-severe knee OA, so the results cannot yet be assumed to apply to all osteoarthritis phenotypes or to every exercise prescription. Even so, the tissue pattern is coherent: more collagen, fewer satellite cells, more PDGFRα+ mesenchymal cells, and a weaker extracellular matrix protein response after loading all point in the same direction.