Proposed Standard Nomenclature for New Tumor Necrosis Factor Family Members Involved in the Regulation of Bone Resorption †
This article is published simultaneously in “Bone” © American Society for Bone and Mineral Research and Elsevier Science, Inc.
Recently, three new family members of the tumor necrosis factor (TNF) ligand and receptor signaling system that play a critical role in the regulation of bone resorption have been identified and cloned. These also have been shown to play an important role in regulating the immune system. A proliferation of synonyms for these molecules has led to miscommunication and redundancy. To resolve this, the President of the American Society for Bone and Mineral Research (ASBMR) appointed a special committee to recommend a standard nomenclature. After considerable deliberation and after vetting by workers in the field, the Committee recommends the names of receptor activator of NF-κB (RANK) for the membrane receptor, RANK ligand (RANKL) for the ligand, and osteoprotegerin (OPG) for the decoy receptor.
BONE RESORPTION IS a complex, multistep process. This involves (1) commitment, differentiation, and fusion of osteoclast precursors in bone marrow to form multinucleated mature osteoclasts; (2) regulation of the functional activity of the mature osteoclast; and (3) when construction of the resorption cavity has been completed, termination of activity and osteoclast removal by apoptosis. As reviewed elsewhere, (1–3) this process involves the activation of multiple genes and the action of multiple hormones and cytokines.
However, by the early 1990s, it had become clear that these known factors were insufficient to support osteoclast formation and function and that one or more unknown factors produced by osteoblasts were necessary for completion of osteoclast formation and activation. It was shown that most of the hormones and cytokines involved in the regulation of bone resorption, for example, parathyroid hormone, calcitriol, interleukin-1β (IL-1β), tumor necrosis factor (TNF), IL-6, and prostaglandin E2, act via receptors on osteoblasts, rather than on osteoclasts. The major exception to this general rule is calcitonin, which acts by a receptor on osteoclasts. McSheehy and Chambers(4) showed that when the agents stimulating bone resorption were applied directly to isolated osteoclasts, there was no increase in their ability to form pits on bone slices, but when conditioned medium from osteoblasts treated with these agents was applied, pit-forming activity was increased. Udagawa et al.(5) then showed that when ST-2 cells, a mouse marrow stromal cell line capable of differentiating into osteoblasts, were cocultured with mouse bone marrow containing osteoclast precursors, differentiation to mature osteoclasts occurred. However, when the two cell types were cocultured separated by a semipermeable membrane, the osteoclasts did not differentiate, implying the necessity of cell-to-cell contact. This hypothetical membrane-bound molecule that was induced on osteoblasts/stromal cells by bone-resorbing factors was named the osteoclast differentiation factor (ODF).
IDENTIFICATION AND CLONING OF TNF FAMILY SIGNALING MOLECULES
Readers are referred to recent reviews(6,6) for details and more complete references on the three classes of molecules involved in this signaling system.
The decoy receptor
Before the osteoblast factor responsible for the completion of osteoclast formation was isolated, two groups (8–10) independently isolated and cloned a resorption-inhibiting molecule that acted as a decoy receptor for the unknown ligand, which they respectively named osteoprotegerin (OPG) and osteoclastogenesis inhibitory factor (OCIF). This was found to be a novel member of the TNF receptor (TNF-R) superfamily. Other investigators, who also had isolated the molecule in immune cells, named it the TNF receptor-related molecule 1 (TR-1(11) and the follicular dendritic cell receptor 1 (FDCR-1).(12) Unlike other TNF-R superfamily members, this molecule lacks transmembrane and cytoplasmic domains and, thus, can be secreted as a soluble protein. Overexpression of this decoy receptor in transgenic mice leads to nonlethal osteopetrosis caused by decreased osteoclast formation and function, and targeted gene ablation leads to mice with severe osteoporosis and to arterial calcification. When the soluble decoy receptor binds to the ligand, it prevents the ligand from binding to its biological receptor, thereby inactivating it.
The receptor ligand
By using the decoy receptor as a probe, the ligand was isolated and cloned independently by two groups(13,13) and was named ODF and OPG ligand (OPGL), respectively. This ligand, a novel member of the TNF superfamily, stimulates osteoclast differentiation, enhances the activity of mature osteoclasts, and prolongs osteoclast life by inhibiting apoptosis. In the presence of permissive concentrations of macrophage colony-stimulating factor (M-CSF; also called colony-stimulating factor 1), the ligand is both necessary and sufficient for osteoclast formation and, thus, for bone resorption. Targeted deletion of the ligand in mice leads to osteopetrosis, shortened bones, impaired tooth eruption, and immunologic abnormalities. The ligand was found to be identical with the molecule previously reported in the immunology literature as TNF-related, activation-induced cytokine (TRANCE(15) or receptor activator of NF-κB ligand (RANKL)(16) in which it was reported to stimulate T cell growth and dendritic cell function.
The membrane receptor
The receptor for the ligand is identical with a receptor previously identified on immune cells and named RANK.(16,17) It also has been named osteoclast differentiation and activation receptor (ODAR).(18) When the ligand binds to this receptor, there is activation of several signaling pathways that lead to biological activity. Mice made deficient in RANK by targeted deletion of the gene have severe osteopetrosis and fail to develop peripheral lymph nodes.
THE NOMENCLATURE PROBLEM
Provided permissive concentrations of M-CSF are present, these three new TNF superfamily molecules—the receptor ligand, the membrane receptor, and the decoy receptor—constitute the final common pathway for regulation of osteoclast formation and function and, as such, mediate the effects of large numbers of upstream hormones and cytokines. Thus, they provide new insights into the biology of bone resorption and into the pathophysiological mechanisms of the many metabolic bone diseases that are characterized by relative or absolute increases in levels of bone resorption. Over the 3 years since the first of these molecules was cloned, research progress has been truly breathtaking. However, the nomenclature used by different investigators is redundant, confusing, and unwieldy. Indeed, there are 14 names (Table 1) for the three components of this system. This has adversely affected communication among investigators and hampered an understanding of the system by clinicians and investigators who are not working in this area. For those who do, it has caused increased difficulty in writing research communications because of the need to provide multiple synonyms.
Table Table 1.. Listing of Terms Used for the Three TNF Family Molecules Involved in Regulation of Osteoclast Formation and Function
To address this problem, in August, 1999, Dr. Jane Aubin, the President of the American Society for Bone and Mineral Research (ASBMR), appointed a special committee to develop a recommended standard nomenclature. The Committee was constituted to include workers in the field that had expressed an interest in developing a standard nomenclature, representatives of the groups that were codiscoverers of the ligand and decoy receptor, and representatives of four major journals that publish many of the papers on bone research—the Journal of Bone and Mineral Research, Bone, Endocrinology, and Endocrine Reviews.
THE COMMITTEE'S DELIBERATIONS AND RECOMMENDATIONS
The Committee met initially on October 3, 1999 during the ASBMR annual meeting in St. Louis and, subsequently, by an international conference call on October 21, 1999. After an in-depth discussion on the scope of the problem, there was unanimous agreement that the standard nomenclature to be developed should consider (1) priority in publication, (2) relevance to physiological processes, (3) current usage in the bone field, and (4) simplicity and acceptability to the majority of workers in the area of bone biology.
The Committee first addressed the names given to the three molecules by the 1998 TNF Conference, that is, TNFSF-11, TNFSFR-11A, and TNSFR-11B for the ligand, receptor, and decoy receptor, respectively. Although there were certain advantages to this nomenclature, it was unanimously felt that its disadvantages outweighed these. In particular, such a complex system likely would be confusing to most ASBMR members and other workers in the bone field. Also, it apparently has not been widely accepted by the nonbone scientific community. The Committee next discussed the feasibility of combining existing names. Although this would defuse some of the competition between groups, it would violate the requirement for simplicity. Finally, it was felt that it was too late to develop a new nomenclature system that might provide a more rational way of describing the structure and function of the new molecules. Moreover, such a new system would be difficult to devise at present because we do not yet understand fully the molecules' extraskeletal functions.
After considerable discussion, the Committee first tried to reach consensus on naming the membrane receptor. Because it had fewer synonyms, and, because it originally was discovered outside of the bone field, the naming was felt to be less controversial. RANK was chosen over ODAR based on priority and general usage. There was unanimity of opinion that RANK should be used as a term and not as an abbreviation, that is, it should be used without providing the original derivation of “receptor activator of NF-κB.” Having reached this conclusion, the Committee recommended that the ligand be called RANKL because RANK is to be the name of the cognate receptor, even though the receptor activation of NF-κB may not be the major function of the ligand. The name ODF has been used since 1992 and is supported by a great deal of experimental work. However, it was felt to be less meaningful to members of the immunology community who would be unfamiliar with ODF. The same was true for OPGL, and, in addition, it seemed more rational to name the ligand based on the membrane receptor, rather than on the decoy receptor. There was considerable debate about the name of the decoy receptor. Eventually, the name of OPG was chosen because it is used widely and adequately describes its physiological function, that is, to protect bone. Thus, the Committee provisionally recommended the names of RANK for the membrane receptor, RANKL for the ligand, and OPG for the decoy receptor.
These provisional recommendations were then vetted by 35 bone biology investigators1 who were working in this field and by three prominent representatives 2 of the immunology field, the other area in which there is active research on these molecules. There was no major disagreement with the provisionally recommended nomenclature, except that one of the immunology scientists (C.W.) recommended that we more strongly consider calling the ligand TRANCE, the term most commonly used in the immunology literature. However, the Committee noted that this term is rarely used by the bone biology community and that the recommended term, RANKL, also is used often in the immunology literature.
Consequently, the provisional terms were finalized, and our recommendations were sent to Dr. David Goltzman, the current President of the ASBMR. He presented it to the ASBMR Council at their winter meeting on February 8, 2000, where it was approved. It also was approved by the editors of the four bone-related journals.
The use of the final nomenclature now will be required for abstracts submitted to the ASBMR Annual Meeting and for manuscripts published in the four bone-related journals. To facilitate communication and understanding elsewhere, we also recommend its adoption by the broader scientific community.
Scientists participating in vetting from the bone biology are: Brendan F. Boyce, M.D., University of Rochester; Maria Luisa Brandi, M.D., Ph.D., University of Florence; Arthur E. Broadus, M.D., Ph.D., Yale University; Timothy J. Chambers, M.D., St. George's Hospital Medical School; Pierre D. Delmas, M.D., Ph.D., University Claude Bernard of Lyon; David W. Dempster, Ph.D., Helen Hayes Hospital; Rolf Felix, Ph.D., University of Berne; David M. Findlay, Ph.D., Royal Adelaide Hospital; Herbert A. Fleisch, M.D., University of Berne; Toshiro Fujita, Ph.D., University of Tokyo; Carol V. Gay, Ph.D., Pennsylvania State University; Matthew T. Gillespie, Ph.D., St. Vincent's Institute of Medical Research; Shin-Ichi Hayashi, M.D., Ph.D., Tottori University; Michael A. Horton, M.D., Rayne Institute; Shigeaki Kato, Ph.D., University of Tokyo; Sohei Kitazawa, M.D., Ph.D., Kobe University; Masayoshi Kumegawa, D.D.S., Meikai University; Joseph Lorenzo, M.D., VA Medical Center; Toshio Matsumoto, M.D., University of Tokushima; Gregory R. Mundy, M.D., University of Texas at San Antonio; Toshitaka Nakamura, M.D., University of Occupational & Environmental Health; Geoffrey C. Nicholson, M.D., Ph.D., Geelong Hospital; Esturo Ogata, M.D., Ph.D., Cancer Institute Hospital; Claes Ohlsson, M.D., Ph.D., Sahlgrenska University Hospital; Phillip Osdoby, Ph.D., Washington University; Merry Jo Oursler, Ph.D., University of Minnesota at Duluth; J. Wesley Pike, Ph.D., University of Cincinnati; Lawrence G. Raisz; M.D., University of Connecticut Health Center; Gideon A. Rodan, M.D., Ph.D., Merck Research Laboratories; G. David Roodman, M.D., Ph.D., University of Texas; Yoshiki Seino, M.D., Ph.D., Okayama University Medical School; Sakae Tanaka, M.D., Ph.D., University of Tokyo; Steven L. Teitelbaum, M.D., Washington University; H. Kalervo Vaananen, M.D., Ph.D., University of Turku; Alberta M. Zambonin-Zallone, Ph.D., University of Bari.
Scientists involved in vetting from the immunology field: David Baltimore, Ph.D., California Institute of Technology; David Goeddel, Ph.D., Tularik, Inc.; Carl Ware, Ph.D., La Jolla Cancer Center Research Center.