The Muscle-Bone Crosstalk: How Myokines, Sclerostin, and Mechanical Load Determine Your Skeletal Fate

Muscle and bone adapt together because they share two inputs: (1) physical forces created by muscle contractions (mechanical loading) and (2) tissue-to-tissue signaling molecules released from muscle and bone. Contraction-associated muscle signals (often grouped as “myokines”) and load-sensitive bone signals help coordinate whether bone-forming cells are activated or restrained.

A central “brake” in this system is sclerostin, produced by osteocytes. When skeletal loading is low, sclerostin tends to rise and dampen Wnt signaling, which reduces bone formation; when bones experience higher strain, sclerostin is typically suppressed, allowing bone formation pathways to proceed.

This framework helps explain why training volume alone does not reliably predict bone mineral density (BMD) across individuals: studies often find stronger links between muscle mass/quality and BMD than between activity volume and BMD, but much of the human evidence is observational and cannot prove causation or identify a single dominant signal.

The muscle–bone link is increasingly described as a bidirectional system with two main channels: mechanical forces (strain on bone created by muscle contractions) and biochemical signaling between tissues [10]. Contracting muscle releases a broad set of molecules often grouped as “myokines,” while bone cells release “osteokines,” and both can influence remodeling and metabolism across tissues [10].

In humans, the most consistent finding is not that any single myokine perfectly predicts bone outcomes, but that muscle mass indices track strongly with bone measures. For example, appendicular muscle mass index has been reported as a leading determinant of bone mineral content and BMD in children, even after accounting for body size and fat mass [4]. Similarly, in a cohort of children, lean mass-related measures were associated with BMD alongside (and sometimes more strongly than) commonly discussed blood markers [8,9].

Mechanistically, candidate signals include irisin and IGF-axis–related proteins, but in most human studies these pathways are inferred from associations rather than demonstrated as causal instructions from muscle to bone [10,15]. Where circulating IGFBP-2 has been measured, it has shown associations with muscle mass and bone geometry/microstructure in adults, consistent with involvement of IGF-axis signaling in the broader muscle–bone network, but not establishing directionality on its own [15].

A widely cited control point in bone’s response to loading is sclerostin. Osteocytes produce sclerostin, which inhibits Wnt/β-catenin signaling and thereby restrains osteoblast-mediated bone formation [10]. In load-deprived contexts, sclerostin expression tends to increase; with higher mechanical strain, sclerostin is often suppressed, which is one mechanism by which loading can shift remodeling toward formation [10].

It is important to separate what is well-established from what is inferred. The load–sclerostin–Wnt relationship is supported by mechanistic work and integrated reviews, but translating that into a simple “stronger muscles → lower sclerostin → higher BMD” chain in free-living humans is less direct because many factors co-vary (age, hormones, inflammation, nutrition, and activity type) [10,11].

On the clinical-phenotype side, muscle quality appears to matter: muscle density (a proxy for intramuscular fat infiltration) has been reported to discriminate hip fracture better than areal BMD in at least one imaging-based study, implying that properties of muscle beyond size relate to fracture-relevant outcomes [14]. This supports the idea that ‘how muscle functions and loads the skeleton’ may be as important as exercise volume alone, even though the molecular intermediates are still being mapped in humans [10,14].

If mechanical loading were the only driver, total training volume would be expected to track tightly with BMD. In practice, results are mixed across populations and study designs, suggesting that “volume” is an incomplete proxy for the type and magnitude of skeletal strain and for underlying biology that governs remodeling.

For example, in hemodialysis patients, higher vigorous and moderate activity volumes were positively associated with skeletal muscle mass but showed no association with BMD or bone metabolism markers [12]. This kind of pattern is consistent with the idea that maintaining muscle and accumulating activity are not always sufficient—especially in clinical populations where bone remodeling may be altered by disease-related factors—to produce measurable BMD differences over typical study windows [12].

Observational data also support that resting muscle mass correlates with bone-relevant traits: skeletal muscle mass has been associated with bone geometry/microstructure and with circulating IGFBP-2 in adults [15]. However, interpreting this as a ‘continuous stream of bone-building signals’ is a mechanistic hypothesis rather than a directly proven human pathway; these studies cannot distinguish whether higher muscle mass drives bone differences, stronger bone supports more muscle, or both reflect shared determinants such as growth, hormones, and nutrition [10,15].

Nutritional status and endocrine context can modulate the muscle–bone system, but the strength of evidence varies by factor and by outcome.

Creatine is one of the better-studied supplements in resistance-training contexts. In a randomized, double-blind, placebo-controlled trial in vulnerable older women, creatine combined with resistance training produced improvements in muscle outcomes and reported bone-related outcomes compared with training alone over 24 weeks, consistent with an indirect pathway where greater strength/muscle adaptation could increase skeletal loading [2]. The study supports a plausible muscle-mediated route to bone effects, but it does not isolate whether changes were driven by altered loading, changes in fall risk/functional performance, or other intermediates.

Broader nutritional adequacy also appears related to both tissues. In post-menopausal women who had undergone thyroidectomy, the Geriatric Nutrition Risk Index and skeletal muscle mass index were associated with BMD, aligning with the view that protein/energy status and muscle mass co-travel with bone outcomes in some settings [6]. Vitamin D status and parathyroid hormone have also been examined as functional biomarkers in childhood and in disease cohorts; these markers relate to bone mass and may interact with muscle strength, but the direction and magnitude of effects depend on age, baseline status, and population studied [9,13]. Overall, these data support ‘modulation’ rather than a single-nutrient switch for the muscle–bone axis [6,9,10,13].

A useful mental model is that bone adapts to the combination of (a) the strains it experiences from movement and muscle contractions and (b) a network of muscle- and bone-derived signals that tune remodeling. Sclerostin’s role as a load-sensitive inhibitor of bone formation fits this model, while human studies most consistently show that muscle mass/quality correlates with BMD and fracture-relevant outcomes—though causality and the exact mediator signals remain incompletely established.