In 2015, Waning et al. [10] described changes to the bone microenvironment in both mouse models and in patients with bone metastases from breast and lung cancer. The work of this group identified that bony metastases from cancers resulted in inflammation, with resultant liberation of TGF-β from bone. The result of this release was the generation of ROS in the sarcoplasmic reticulum of skeletal muscle, where it interacts with the ryanodine receptor, oxidizing it and causing a calcium leak resulting in muscle weakness. The effects of TGF-β on the ryanodine receptor could be reversed by use of either a TGF-β1 receptor blocker or the anti-resorptive agent zoledronate. This study confirmed the work of Abrigo et al. [7] in identifying ROS generation as a mode of action leading to muscle weakness in metastatic cancer. Another piece of the puzzle, the mechanics of the release of TGF-β from resorbing bone, was supplied by the work of Dallas et al. [11]. This group reported that TGF-β was stored in the bone extracellular matrix bound to Latent TGF-β Binding Protein-1 and that osteoclasts, with the aid of matrix metalloproteinases (MMP)-2 and -9, can cleave this binding protein from TGF-β, liberating it from the extracellular matrix during resorption. Figure 1 illustrates the known pathway of TGF-β release from bone. What was, at the time, unclear was whether this TGF-β liberation from bone with paracrine catabolic effects on muscle was unique to metastatic cancers. As it turns out, Pin et al. [12] subsequently described muscle wasting in cancer without metastases, inasmuch as IL-6, tumor necrosis factor-alpha (TNF-α), interferon-gamma (IFN-¥) and IL-1-β are chronically elevated in blood, peripheral tissues, and the central nervous system of patients with cancer cachexia. These cytokines can be produced by either the immune cells or tumor cells. Moreover, myostatin, a member of the TGF-β family of proteins, and one that inhibits muscle protein synthesis, has a catabolic effect on bone by inhibiting osteoblast differentiation by decreasing the osteocyte-derived production of exosomal microRNA (mIR) 218 [13]. Myostatin also increases RANK Ligand activity in stimulating osteoclast formation. To compound matters, Chen et al. [14] showed that the inhibition of activin receptors and of myostatin inhibited the Smad 2/3 pathway, producing muscle hypertrophy. The driving factor in the muscle hypertrophy was an increase in muscle protein synthesis and a reduction in ubiquitin ligase activity—findings seen with the inhibition of TGF-β release from bone. Thus, the relationship between the TGF-β molecule liberated from bone during resorption, and muscle myostatin activity, remains unclear.

However, it is known that both TGF-β1 and myostatin bind to the same receptors, which, when bound, will phosphorylate the Smad 2/3 signaling pathway, resulting in stimulation of the ubiquitin ligase muscle protein catabolic pathway and suppression of the Akt/mTor muscle protein anabolic pathway.

Not only is muscle wasting produced by cancer metastases and cancer without metastases, but chemotherapeutic agents, such as carboplatin [15] and cisplatin [16], are capable of producing bone loss and consequent muscle wasting with zoledronate preventing the muscle wasting in both instances. The authors of these papers did not specify that activation of TGF-β was involved in the pathogenesis of the muscle wasting.

We do not have evidence implicating bone-released TGF-β’s involvement in immobilization to date. However, a recent study by Gugala et al. [31] provides suggestive information. Studies of a rat model of critical illness myopathy developed by Llano Diez et al. [32] have shown that by pharmacologically immobilizing rats and mechanically ventilating them for 10 days, a model that reproduces the effects of immobilization and the mechanical ventilation of patients in intensive care for COVID-19 respiratory disease, the loss of appendicular skeletal muscle mass and, in particular, myosin, occurs in tight correlation with the loss of trabecular bone, as determined by serial micro computed tomography of the femurs of rats studied from 0–10 days post-immobilization. While immobilization is the main feature of this model, it is difficult to discount a role of inflammation that may occur as a result of mechanical ventilation. This is because mechanical ventilation itself causes an inflammatory response by upregulating the NLRP3 inflammasome [33], resulting in the increased production of IL-1 by the monocytes and macrophages of the innate immune system. Therefore, the model is not entirely “clean”, in a manner similar to burn injury. However, the tightness of the correlation between muscle and bone loss in this experimental setting is suggestive that bone resorption is involved in the pathogenesis of muscle wasting in this model. The definitive experiment would involve the use of bisphosphonates or other anti-resorptive agents to block the bone loss and then evaluate the muscle wasting. However, studies of atrophic muscle fibers in this condition have been reported to express TGF-β ligands [34]. Similarly, studies of sarcopenia have reported an increase in circulating TGF-β1, phosphorylated Smad 3, and myostatin, again suggesting the involvement of TGF-β1 in these conditions [35]. Yet, it is unclear whether muscle myostatin is upregulated by TGF-β released from bone, although, as we have seen [21], this may not occur in aging, sarcopenic muscle. However, this relationship has not been studied in other conditions, so it is not clear that this is categorically the case.