Does Defective Bone Lead to Defective Muscle?

The assumed functions of bone have changed dramatically over the past decade from the obvious such as locomotion and protection of organs, to less obvious, such as functioning as an endocrine organ. Because the locomotion function of bone cannot be accomplished without muscle, the mechanical loading of bone bymuscle was originally assumed to be the only interaction between muscle and bone and that idea has dominated the field. Later it was proposed that muscle can directly regulate and maintain bone mass. The concept was that muscle regulates bone mass through contraction to mechanically load bone and that this mechanical loading increases bone mass. More recently, it has been proposed that muscle can act as an endocrine organ and that potential muscle myokines can affect other organs including bone. Several muscle-derived factors have been proposed to have an effect on bone or bone cells (for review see Girgis and colleagues and Hamrick) which is supported by in vitro and in vivo results. The growing list of factors made by muscle include myostatin, irisin, IGF, IL-5, IL-6, IL-7, IL-8, IL-15, chemokine (C-X-C motif) ligand 1 (Cxcl-1), neurotrophin 3 (NT3), leukemia inhibitory factor (LIF), brain-derived neurotrophic factor, and others. It has also been proposed that the mechanical loading of bone by muscle combined with the factors secreted by contracting muscle can together synergize to regulate bone mass. Conversely, it has recently been shown that bone secretes factors that can target other organs such as pancreas and kidney through relatively bone-specific factors such as osteocalcin and FGF23 (for reviews see Pi and Quarles and Ferron and colleagues). More recent studies using in vitro approaches suggest that bone may regulate muscle mass and function (Mo CL, Romero-Suarez S, Bonewald LF, Johnson ML, Brotto M, unpublished work). The study by Shen and colleagues in this issue of the JBMR is one of the first that has used in vivo studies to support the concept that bone can target muscle. Their article, the impaired muscle function was partially rescued by administration of a bone-specific factor, undercarboxylated osteocalcin (glu-OC). The investigators performed targeted deletion of connexin 43 (Cx43), using the 2.3-kb Col1a1 promoter and reproduced their previous observations of reduced cortical bone thickness and density in these mice. After characterization of the skeletal muscle in these mice, they describe a phenotype of lower fast muscle weight, grip strength, and force. As mRNA for osteocalcin was reduced in the humerus, and carboxylated (also known as the “inactive” form) and undercarboxylated isoforms of osteocalcin (also referred to as the “active” form) were reduced in the serum, they tested the effects of glu-OC on myogenesis in vitro and in vivo. They found that glu-OC increased myotube formation by C2C12 cells and in vivo injections of glu-OC into these transgenic mice rescued muscle cross-sectional area and grip strength. With regard to the specificity of the Cre model used, the 2.3-kb Col1a1, the authors showed activation of osteoblasts and osteocytes but not in muscle. Also, because osteocalcin, a bone-specific factor, corrected the phenotype, this suggests that the effects are not due to deletion of Cx43 in muscle cells. These findings provide new in vivo support for the concept that bone can send signals to regulate muscle. Interestingly, in this study, the bone phenotype occurs before the muscle phenotype. In the mice with conditional deletion of Cx43, the bone defect of enlarged marrow and reduced cortical bone thickness, density, and bone area fraction were evident and significant at 14 days after birth, raising the question of whether this bone defect was present at or before birth. The significant differences in bone at 14 days were present at 28 and 56 days. The reduced body weight and reduced muscle volume and muscle weight were not significantly different from control until days 28 and 56. This suggests that the bone phenotype occurs some time before the muscle phenotype. This observation raises a number of questions. Not onlywill it be important to determine if defective bone models have defective muscle but also whether the bone phenotype precedes the muscle phenotype as in the model in the article by Shen and colleagues. If it is consistently observed that a defective bone phenotype precedes a defective muscle phenotype, this has implications for several clinical conditions and diseases. For example, could the defective bone phenotype in osteogenesis imperfecta be responsible for the muscle phenotype? Could the loss of bone mass and function in osteoporosis precede and be partially responsible for age-related sarcopenia? These would be fertile areas for investigation. As often occurs with many studies, this study raises a number of intriguing questions. How does a gap junction/hemichannel deletion regulate bone mass? It is thought that the major role of Cx43 is to facilitate cellular communication through transport or