Our understanding of the mechanism by which stimulators of resorption activate and maintain osteoclastic responses in bone was greatly advanced by the discovery of a TNF-related ligand and two TNF receptor–related proteins. These proteins have acquired multiple names because four different groups identified them independently. The ligand, referred to here as RANK-L/OPG-L, was originally cloned and sequenced as a receptor activator of NF-κB ligand (RANK-L) (16
) and as a TNF-related activation induced cytokine (TRANCE) (17
) by investigators who were examining T-lymphocyte products. Subsequently, two groups identified the same protein as osteoprotegerin ligand (OPG-L) (18
) and osteoclast differentiation factor (ODF) (19
) based on its ability to stimulate osteoclast formation.
The receptor for RANK-L/OPG-L on responsive cells is named RANK (16
). In addition, a variety of cells produce a soluble decoy receptor that has been termed osteoprotegerin (OPG) (20
). OPG is released from cells and binds RANK-L/OPG-L, thus preventing activation of RANK. The majority of stimulators of resorption enhance the production of RANK-L/OPG-L, and some also inhibit OPG production in stromal and osteoblastic cells. Hence, regulation of RANK-L/OPG-L and OPG appears to be a critical mechanism by which stimulators of resorption influence the activity of osteoclasts in bone (see the recent JCI
Commentary by F.P. Ross for a detailed review of this subject [ref. 21
Along with their potent effects on osteoclast development, RANK-L/OPG-L, OPG, and RANK have important roles as regulators of immune cell function. This is most clearly demonstrated in RANK-L/OPG-L–deficient (22
) and RANK-deficient mice (23
). In addition to osteopetrosis arising from a complete absence of osteoclasts, mice of these two mutant genotypes show significant alterations in both immune cells and immune cell organs. Thymic cellularity and size are significantly decreased in RANK-L/OPG-L–deficient mice. Furthermore, differentiation of CD44–
precursor cells into CD44–
cells — an important step in maturation to CD4+
single-positive thymocytes — is blocked in the thymus of RANK-L/OPG-L–deficient mice. A similar block was seen in OPG-overexpressing transgenic mice, suggesting that RANK-L/OPG-L is a critical factor in T-lymphocyte maturation in the thymus at the stage of pre–T-cell receptor expression.
An equally striking observation was that B-lymphocyte maturation in the bone marrow of RANK-L/OPG-L–deficient mice arrests at the progression of B220+CD25– pro–B lymphocyte precursors to B220+CD25+ pre–B lymphocytes. Despite the common developmental origin of osteoclasts and monocytes and the fact that RANK-L/OPG-L can act as a dendritic cell survival factor, RANK-L/OPG-L–deficient mice show normal dendritic cell and monocyte development. However, T lymphocytes from these mice show poor induction of the cytokines IFN-γ, IL-2, IL-4, IL-5, and IL-6 upon stimulation with anti-CD3 and anti-CD28.
Perhaps the most intriguing aspect of RANK-L/OPG-L–deficient mice is their complete lack of lymph-node development, a defect that does not appear to result from a T-lymphocyte homing deficiency. Interestingly, Peyer’s patches are present in RANK-L/OPG-L–deficient mice, although they are severely reduced in size. RANK-deficient mice manifest a similar phenotype, except that their thymi are of normal size and cellularity. This difference raises the interesting possibility that RANK-L/OPG-L may interact with another receptor in the thymus.
In normal physiology, osteoclast formation is regulated by production of membrane-bound RANK-L/OPG-L and soluble OPG in stromal/osteoblastic cells under the influence of stimulators of resorption (Figure ). However, RANK-L/OPG-L production by activated T lymphocytes can also stimulate osteoclast formation and bone resorption, and IL-1, TNF, and PGE2 can affect osteoclasts directly. Kong et al. (25
) recently demonstrated that activated T cells produce both membrane-bound and soluble RANK-L/OPG-L, which can activate osteoclastogenesis. These authors also found that transfer of spontaneously activated ctla4–/–
T lymphocytes into RANK-L/OPG-L–deficient mice caused bone loss, implying that T lymphocyte–derived RANK-L/OPG-L can directly stimulate osteoclast formation in vivo.
Figure 1 Regulation of osteoclast formation. In normal physiology, osteoclast formation is regulated by production of membrane-bound RANK-L/OPG-L and soluble OPG in stromal/osteoblastic cells under the influence of stimulators of resorption. In inflammatory states, (more ...)