Bone tissue is continuously renewed throughout adult life by a process called 'remodeling', which involves a dynamic interplay among bone cells including osteoclasts, osteoblasts and osteocytes. For example, a tight coupling between bone resorption and formation is essential for the homeostasis of the skeletal system. Studies on the coupling mechanism in physiological and pathological settings have revealed that osteoclasts or osteoclastic bone resorption promote bone formation through the production of diverse coupling factors. The classical coupling factors are the molecules that promote bone formation after resorption, but there may be distinct mechanisms at work in various phases of bone remodeling. A recent study revealed that the Semaphorin 4D expressed by osteoclasts inhibits bone formation, which represents a mechanism by which coupling is dissociated. Furthermore, it has been demonstrated that osteoblastic expression of Semaphorin 3A exerts an osteoprotective effect by both suppressing bone resorption and increasing bone formation. Thus, recent advances have made it increasingly clear that bone remodeling is regulated by not only classical coupling factors, but also molecules that mediate cell–cell communication among bone cells. We propose that such factors be called bone cell communication factors, which control the delicate balance of the interaction of bone cells so as to maintain bone homeostasis.
The bony skeleton enables various crucial processes, such as locomotive activity, the storage of calcium and the harboring of hematopoietic stem cells. Bone is a dynamic organ that is continuously being broken down by osteoclasts and subsequently rebuilt with new bone by osteoblasts throughout the course of one's adult life. These activities occur in response to various hormones, cytokines, chemokines and biomechanical external stimuli.
In the initiation phase, osteoclast precursor cells, which originate from cells of hematopoietic lineage, are recruited to the specific bone surface areas and differentiate into mature osteoclasts. It has long been suggested that when osteocytes sense mechanical loading or microdamage, they stimulate the recruitment of osteoclast precursor cells to the bone surface. Recent studies demonstrated that osteocytes act as major source of RANKL to promote osteoclastogenesis in adult bone remodeling,
In the subsequent transition phase, the osteoprogenitor cells of the mesenchymal lineage migrate to the resorbed sites and differentiate into bone-forming osteoblasts. The migration, differentiation and bone formation activity of osteoblasts are regulated by a large number of paracrine, autocrine and endocrine factors. These include bone morphogenetic proteins, Wnts, parathyroid hormone, prostaglandins, growth factors, such as insulin-like growth factors (IGFs), transforming growth factor-β (TGF-β), and acidic and basic fibroblast growth factors, and angiogenic factors such as vascular endothelial growth factor and endothelin-1.
In the formation phase, osteoblasts deposit new bone matrix called osteoid and subsequently mineralize osteoid. When the resorbed lacunae are replenished with new bone, some of the osteoblasts disappear because of apoptosis. A part of osteoblasts become quiescent as bone-lining cells on the surface of the newly formed bone, or further differentiate into osteocytes embedded in the bone matrix (
How is bone formation stimulated after bone resorption? In 1981, Baylink and Colleagues
A number of osteotropic factors, such as IGFs, TGF-β, bone morphogenetic proteins, fibroblast growth factor and platelet-derived growth factor have been thought to function as coupling factors, because they are stored in large amounts in bone matrix and liberated by bone resorption.
TGF-β1 is deposited in bone matrix as an inactive, latent complex along with latency-associated protein. During bone resorption, osteoclasts acidify the resorption pits and secrete proteolytic enzymes, such as cathepsin K and matrix metalloproteinases, so as to degrade and dissolve the bone matrix. As a result, TGF-β1 is released and activated by the dissociation of latency-associated protein. Mice lacking TGF-β1 exhibited a significant decrease in the osteoblast number on the bone surface, but an abnormal accumulation of osteoblasts in the marrow space.
IGF-I and IGF-II are the most abundant growth factors stored in bone, and it is well established that IGFs has a critical role in bone development and remodeling. The action of IGFs in bone is inhibited by IGF binding proteins (IGFBP1, 2, 4, 6), which sequester IGFs from their receptor. The cleavage of such inhibitory IGFBPs is needed to the IGF action.
The membrane-bound ligand ephrinB2, which is an axon guidance molecule, was found to be expressed on mature osteoclasts, whereas the tyrosine kinase receptor EphB4 was found to be expressed on osteoblast lineage cells.
Previous in vitro studies have demonstrated that several factors regulating mineralization or migration of osteoblastic cells are produced by osteoclasts (
The genetic deletion of these factors specifically in osteoclast lineage cells may be a powerful strategy for elucidating the local factors involved in the communication among bone cells. However, caution should be exercised when using conditional knockout mice because bone formation may be influenced not only directly by the absence of a putative coupling factor but also indirectly by a cell-autonomous osteoclast defect, which results in an alteration of production of other coupling factors. Therefore, it may prove challenging to obtain direct evidence for coupling functions in vivo.
Once bone resorption is initiated, various coupling factors are produced and stimulate bone formation. However, in order to completely remove the damaged or aged bone in the initiation phase, osteoblast differentiation and formation within the BMU needs to be suppressed until bone resorption is accomplished. A recent study provided evidence that osteoclasts suppress bone formation through the expression of Sema4D (
Semaphorins are originally described as axonal guidance repellents that induce growth cone collapse during neuronal development and subsequently recognized as attractant and repellent cues for multiple cell types.
The Semaphorins compose a large family of secreted and membrane-bound glycoproteins characterized by a conserved amino-terminal 'Sema' domain. On the basis of structural features and amino-acid sequences similarity, Semaphorins are divided into eight subclasses, of which classes III–VII are vertebral Semaphorins. Sema4D belongs to the class 4 Semaphorin based on its structure of membrane-bound form, but also acts as a soluble factor after proteolytic cleavage. Sema4D, also called CD100, was first identified in the activated T cells and has been extensively explored in the immune system.
In bone, Sema4D was highly and selectively produced by the osteoclasts and its receptor, Plexin-B1 (encoded by the Plxnb1 gene), was expressed on osteoblasts. The soluble form of Sema4D potently inhibited osteoblast differentiation, whereas it did not affect osteoclastogenesis. The differentiation of osteoblasts was markedly enhanced by the addition of the culture supernatant of Sema4d
Sema4d
The binding of Sema4D to Plexin-B1 was shown to result in the activation of the tyrosine kinase ErbB2, which activates Plexin-B1 by phosphorylation. RhoA was subsequently activated by RhoGEF, including PDZ-RhoGEF and LARG, which associate with Plexin-B1. Plxnb1
What are the mechanisms by which Sema4D inhibits bone formation? Sema4D decreased the phosphorylation of insulin receptor substrate-1 (IRS-1) and its downstream molecules, including Atk and ERK, which is the crucial step in IGF signaling that favors osteoblast differentiation. Activation of RhoA or its effector Rho-associated kinase (ROCK) inhibited the IGF-I-stimulated IRS-1 pathway, whereas suppression of RhoA or ROCK activity activated IRS-1 signaling. Collectively, the Sema4D–Plexin-B1–RhoA pathway suppressed osteoblast differentiation by attenuating IGF-I signaling (
Previous in vitro studies have also proposed an inhibitory regulation of coupling between bone resorption and formation (
During bone formation, the generation of additional sites of initiation of the bone remodeling cycle needs to be prevented within the BMU. It is well known that osteoblast lineage cells inhibit osteoclastogenesis by secreting OPG. However, a recent study reported that a substantial anti-osteoclastogenic effect was observed in the conditioned medium of OPG-deficient calvarial cells and they identified another Semaphorin family member Sema3A as the osteoblast-secreted inhibitors of osteoclast differentiation by mass spectrometry.
Sema3A, which is secreted protein, is the first identified vertebral Semaphorin and has been the most extensively studied in the nervous system, but now is also recognized to be involved in multiple physiological and pathological processes. Recent reports demonstrated that the Sema3A suppresses the onset of autoimmune diseases such as SLE, rheumatoid arthritis and multiple myeloma, as well as tumor progression.
Sema3A binds to a receptor complex of the ligand-binding subunit Neuropilin-1 and one of the class A Plexins. In a recent study, Sema3A was shown to be predominantly produced by osteoblast lineage cells, whereas the receptor complex was expressed both on osteoblast lineage cells and osteoclast precursor cells.
Viable Sema3a
Anti-resorptive drugs such as bisphosphonates have been the primary therapy for osteoporosis and other remodeling diseases, but a general problem with the existing anti-resorptive treatments is an associated decrease in bone formation resulting from the coupling of resorption with formation. Therefore, it would be desirable to develop a strategy that effectively targets either bone resorption or formation by dissociating resorption from formation. It was clearly demonstrated that the treatment with the Sema4D-specific neutralizing antibody and recombinant Sema3A not only effectively protected against bone loss in mouse model of osteoporosis, but also restored the lost bone by increasing bone formation.
The whole cycles of bone remodeling are finely controlled by multiple communication pathways between osteoblast and osteoclast lineage cells at many stages of differentiation and function in these cells. The coupling mechanism has traditionally been considered to regulate the cell–cell communication at the transition stage. However, now the molecules that mediate intercellular signaling at various stages of bone remodeling may be collectively called bone cell communication factors, which include classical coupling factors. In addition to such factors derived from bone cells, the bone is additionally under the control of factors that are related to the immune, vascular and nervous systems.
The bone remodeling process is divided into the initiation, transition and formation phases. In the initiation phase, mechanical loading and microdamage are sensed by osteocytes, which stimulate the recruitment of osteoclast precursor cells. Osteoclastogenesis is stimulated by the RANKL, macrophage colony-stimulating factor (M-CSF) and ligands for immunoglobulin-like receptors, which are produced by osteoblast lineage cells including osteocytes, and bone resorption starts. Osteoclasts inhibit bone formation during bone resorption through the expression of Sema4D. In the transition phase, classical coupling factors, including IGF-I and TGF-β1, stimulate the migration of osteoprogenitors to the resorbed sites and promote differentiation into osteoblasts. In the bone formation phase, osteoblasts replenish the resorbed area with new bone. Sema3A, which is produced by osteoblast lineage cells, inhibits osteoclastogenesis and simultaneously promotes bone formation in this phase.
During bone resorption, osteoclast-derived Sema4D inhibits osteoblast differentiation in the proximity of osteoclasts and repels osteoblasts by increasing their motility. The binding of Sema4D to its receptor complex, consisting ErbB2 and Plexin-B1, activates RhoA through RhoGEFs, including PDZ-RhoGEF and LARG. The Rho-associated protein kinase Rho-associated kinase (ROCK) inhibits IRS-1 phosphorylation, which is the crucial step in IGF-1 signaling for osteoblast differentiation. The motility of osteoblasts is also controlled by the activation of RhoA-ROCK.
During bone formation, osteoblast-derived Sema3A inhibits osteoclast differentiation and migration, and at the same time stimulates bone formation. The binding of Sema3A to Nrp1 on osteoclasts inhibits ITAM signaling by sequestering Plexin-A1 from TREM-2. Sema3A–Nrp1–Plexin-A1 also inhibits the migration of osteoclast precursor cells by suppressing RhoA activation. The binding of Sema3A to Nrp1 on osteoblasts activates Rac1 through the RacGEF FARP2. Rac1 activation enhances the Wnt-mediated nuclear localization of β-catenin, which is essential for osteoblast differentiation.
This work was supported in part by Grants-in-Aid for the Global Center of Excellence Program from the Ministry of Education, Culture, Sports, Science and Technology of Japan (MEXT), and ERATO, Takayanagi Osteonetwork Project from the Japan Science and Technology Agency.