Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Nov 30, 2004; 101(48): 16711–16712.
Published online Nov 22, 2004. doi:  10.1073/pnas.0407335101
PMCID: PMC534741

Postmenopausal osteoporosis, T cells, and immune dysfunction

Because of its dramatic appearance, the osteoclast has been long recognized as a bonedegrading cell, but its ontogeny remained enigmatic until the late 1970s when Donald Walker demonstrated that mice with enhanced bone mass caused by dysfunctional osteoclasts are cured by parabiosis or infusion of spleen cells, laying the foundation for the generation of osteoclasts, in vitro, from marrow preparations. The next critical step came in 1990, when Tatsuo Suda's group (1) demonstrated that macrophages, placed in proper culture conditions, differentiate into bone fide osteoclasts, establishing membership of the bone resorptive cell in the monocyte/macrophage family. Unexpectedly, however, commitment of macrophages to the osteoclast phenotype required contact of the precursor cell with marrow mesenchymal cells. In 1998, the key, stromal cell-residing, osteoclastogenic molecule was identified as RANK ligand (RANKL), a member of the tumor necrosis factor (TNF) superfamily (2).

Postmenopausal osteoporosis, which follows ovarian failure, reflects an absolute acceleration in the rate of bone resorption due to increased osteoclast recruitment. The realization that the osteoclast is a relative of the macrophage, a cell that secretes and is regulated by inflammatory cytokines, raised the possibility that the same molecules also modulate the bone resorptive cell and, in fact, TNF-α and IL-1 proved agonistic in this regard. As a consequence, Pacifici hypothesized in 1987 (3) that estrogen deprivation enhances the production of proresorptive inflammatory cytokines and established that such is the case by demonstrating that IL-1 and TNF-α are secreted by peripheral blood mononuclear cells derived from osteoporotic women.

Neutralization of TNF-α and IL-1 protects against ovariectomy-induced bone loss.

These experiments, however, raised a challenging conundrum. Whereas generation of osteoclasts from bone marrow macrophages required the presence of stromal cells (1), this was not true when the resorptive polykaryon was derived from circulating monocytes, suggesting an alternative source of the “osteoclastogenic factor” (4). Indeed, these peripheral blood-derived monocytic cultures subsequently were found to contain substantial numbers of T lymphocytes. This observation, taken with the finding that the distribution of CD4+ and CD8+ T lymphocyte subsets are altered in osteoporotic women, raised the possibility that the T cell plays a role in the pathogenesis of postmenopausal bone loss and indicated that a relationship exists between skeletal regulation by estrogen and the immune system. The fact that T cell deficient nu/nu (nude) mice are protected from ovariectomy-induced bone loss strongly buttresses this posture (5). In a recent issue of PNAS, Pacifici's group (6) provided compelling evidence that the bone-sparing effects of estrogen are mediated by macrophage-produced transforming growth factor β (TGF-β) specifically targeting to T cells.

The means by which estrogen deprivation enhances osteoclast recruitment appears to involve T cell production of TNF-α and IL-1. Each of these cytokines is incapable of promoting osteoclast recruitment alone. Both, however, induce the RANKL expression by marrow stromal cells that is enhanced in postmenopausal women (7) and synergizes with RANKL in the osteoclastogenic process (5). Hence, neutralization of TNF-α and IL-1 protects against ovariectomy-induced bone loss, as does deletion of the p55 TNF receptor that transmits the cytokine's osteoclastogenic signal. In keeping with a central role for T cell-produced TNF-α in the pathogenesis of postmenopausal osteoporosis, reconstitution with wild-type lymphocytes, but not those incapable of expressing the cytokine, restores the bone-wasting properties of ovariectomy in nude mice (8). This enhancement of cytokine production by estrogen withdrawal does not occur on a per-cell basis but is a manifestation of stimulated T cell proliferation in vivo (8). Consistent with this finding, estrogen does not alter cytokine production by isolated T cells in vitro.

These observations suggest that estropryvic bone loss reflects increased numbers of activated T cells producing inflammatory cytokines, leading, in turn, to RANKL expression and enhanced osteoclast formation and function. The missing component of the paradigm was the mechanism by which estrogen deprivation activates T lymphocytes, thus inducing proliferation. Resolution of the issue came with the observation that ovariectomy enhances marrow macrophage expression of class II transactivator (CIITA), a transcriptional coactivator that up-regulates class II MHC and, consequently, antigen presentation to T cells (9). In fact, ovariectomy-induced bone loss is prevented by blunting antigen-presenting-cell-induced T cell activation.

Thus, a scenario has unfolded that suggests that postmenopausal osteoporosis is a disease of enhanced immune reactivity (Fig. 1). But, if so, what is the antigen(s) driving the process? Although the answer is not yet in hand, a reasonable hypothesis involves naturally arising CD4+ regulatory T cells. These lymphocytes promote immunologic self-tolerance and negatively control immune responses to a broad spectrum of normally harmless environmental antigens such as microbes and allergens to which we are constantly exposed (10). Thus, by releasing the suppressive effect of regulatory T cells, estrogen deprivation may prompt proliferation of effector T cells while at the same time enhancing antigen-presenting cell function. An alternative but not mutually exclusive means by which arrested regulatory T cells may stimulate effector T cells in postmenopausal women would be by the up-regulation of potentially hazardous self-reactive lymphocytes that circulate in normal individuals.

Fig. 1.
Immune modulation of postmenopausal osteoporosis. The decline in circulating estrogen (E2) attending menopause stimulates IFN-γ expression by T cells, enhancing, in turn, CIITA in macrophages. The resulting induction of antigen presentation activates ...

IFN-γ is a potent activator of antigen-presenting cells. T cell expression of IFN-γ is enhanced after estrogen withdrawal, and in this circumstance the immune regulatory molecule mediates CIITA expression and the resulting T cell proliferation (9). According to Gao et al. (6), estrogen-induced TGF-β, specifically targeting to T cells, suppresses IFN-γ synthesis, seemingly attenuating antigen presentation and T cell activation and proliferation. Mice whose T cells are unable to react to TGF-β are not only unresponsive to the bone sparing effects of estrogen after ovariectomy but also experience a lower developmental bone mass with age, presumably reflecting IFN-γ-driven osteoclastogenesis. Finally, in an elegant example of experimental gene therapy, the authors prevent ovariectomy-induced bone loss by TGF-β overexpression. Thus, TGF-β appears to be central to normal skeletal development as well as maintenance of bone mass after menopause.

The relationship of the immune system to the pathogenesis of postmenopausal osteoporosis, pioneered largely by the Pacifici laboratory, has progressed into a compelling story, but not one without controversy, particularly with regard to the roles of IFN-γ and TGF-β. For example, the authors' observation that stimulated IFN-γ expression is key to the enhanced osteoclastogenesis of estrogen deprivation stands in the face of claims that the immune regulatory molecule is, in fact, antiosteoclastogenic. Hence, mice lacking the IFN-γ receptor have increased numbers of osteoclasts in response to inflammatory stimuli such as lipopolysaccharide (11).

The role of TGF-β in the osteoclastogenic process is perhaps even more confounding. Consistent with the conclusions of Gao et al. (6), the growth factor, which is abundant in bone matrix, is an established repressor of immune cell function in vivo and in vitro, exerting an antiproliferative effect on activated T cells (12). It arrests naive T cell maturation into TH1 progeny in circumstances such as suboptimal antigen presentation and blunts IFN-γ-stimulated CIITA expression by macrophages (13), observations in keeping with the cytokine's inhibitory effect on estropryvic osteoclast recruitment. Furthermore, others have shown that exogenous TGF-β decreases osteoclast numbers in vivo (14). Clinical support for a role of TGF-β in estrogen-mediated skeletal sparing comes from the fact that bones of women treated with the steroid are richer in the growth factor than their untreated counterparts (15). On the other hand, direct exposure of macrophages to TGF-β, in vitro, accelerates their differentiation into osteoclasts by a mechanism involving reversal of the antiosteoclastogenic effects of IFN-γ and IFN-β (16). Furthermore, TGF-β overexpression in bone results in an osteoporosis-like phenotype associated with increased osteoclastic activity (17).

The apparent paradoxical effects of IFN-γ and TGF-β on the osteoclast likely reflect the context in which the cytokines find themselves and the cells which they are targeting. In this regard, T lymphocytes also play a central role in a family of disorders characterized by enhanced osteoclast recruitment and function caused by production of inflammatory cytokines in amounts much greater than those attending menopause. These intensely inflammatory conditions include active rheumatoid arthritis (18) and periodontal disease (19) in which T cells synthesize an abundance of RANKL. These cells also express large amounts of IFN-γ, whose dominant impact on osteoclasts in these disorders appears to involve direct suppression of precursor differentiation by blunting the RANKL/RANK signaling pathway (11). A similar argument may hold with regard to TGF-β, which is abundantly produced by macrophages in the context of inflammation. Perhaps in this circumstance the growth factor's primary role is not to inhibit osteoclast recruitment by blunting antigen presentation and T cell activation as occurs with estrogen withdrawal but to promote osteoclast formation by direct induction of the resorptive cell's precursors. Regardless of the ultimate mechanistic distinctions between the osteoclastogenesis characterizing postmenopausal osteoporosis as compared with inflammatory osteolysis, the relationship between the immune system and the skeleton is here to stay, and the T cell is likely to be a major player.


See companion article on page 16618 in issue 47 of volume 101.


1. Udagawa, N., Takahashi, N., Akatsu, T., Tanaka, H., Sasaki, T., Nishihara, T. & Suda, T. (1990) Proc. Natl. Acad. Sci. USA 87, 7260–7264. [PMC free article] [PubMed]
2. Lacey, D. L., Timms, E., Tan, H. L., Kelley, M. J., Dunstan, C. R., Burgess, T., Elliott, R., Colombero, A., Elliott, G., Scully, S., et al. (1998) Cell 93, 165–176. [PubMed]
3. Pacifici, R., Rifas, L., Teitelbaum, S., Slatopolsky, E., Miller, R., Bergfeld, M., Lee, W., Avioli, L. V. & Peck, W. A. (1987) Proc. Natl. Acad. Sci. USA 84, 4616–4620. [PMC free article] [PubMed]
4. Matayoshi, A., Brown, C., DePersio, J. F., Haug, J., Abu-Amer, Y., Liapis, H., Kuestner, R. & Pacifici, R. (1996) Proc. Natl. Acad. Sci. USA 93, 10785–10790. [PMC free article] [PubMed]
5. Cenci, S., Weitzmann, M. N., Roggia, C., Namba, N., Novack, D. & Pacifici, R. (2000) J. Clin. Invest. 106, 1229–1237. [PMC free article] [PubMed]
6. Gao, Y., Qian, W.-P., Dark, K., Toraldo, G., Lin, A. S. P., Guldberg, R. E., Flavell, R. A., Weitzmann, M. N. & Pacifici, R. (2004) Proc. Natl. Acad. Sci. USA 101, 16618–16623. [PMC free article] [PubMed]
7. Eghbali-Fatourechi, G., Khosla, S., Sanyal, A., Boyle, W. J., Lacey, D. L. & Riggs, B. L. (2003) J. Clin. Invest. 111, 1221–1230. [PMC free article] [PubMed]
8. Roggia, C., Gao, Y., Cenci, S., Weitzmann, M. N., Toraldo, G., Isaia, G. & Pacifici, R. (2001) Proc. Natl. Acad. Sci. USA 98, 13960–13965. [PMC free article] [PubMed]
9. Cenci, S., Toraldo, G., Weitzmann, M. N., Roggia, C., Gao, Y., Qian, W. P., Sierra, O. & Pacifici, R. (2003) Proc. Natl. Acad. Sci. USA 100, 10405–10410. [PMC free article] [PubMed]
10. Sakaguchi, S. (2004) Annu. Rev. Immunol. 22, 531–562. [PubMed]
11. Takayanagi, H., Ogasawara, K., Hida, S., Chiba, T., Murata, S., Sato, K., Takaoka, A., Yokochi, T., Oda, H., Tanaka, K., et al. (2000) Nature 408, 600–605. [PubMed]
12. Gorelik, L. & Flavell, R. A. (2002) Nat. Rev. Immunol. 2, 46–53. [PubMed]
13. Lee, Y., Han, Y., Lu, H., Nguyen, V., Qin, H., Howe, P., Hocevar, B., Boss, J., Ransohoff, R. & Benveniste, E. (1997) J. Immunol. 158, 2065–2075. [PubMed]
14. Beaudreuil, J., Mbalaviele, G., Cohen-Solal, M., Morieux, C., De Vernejoul, M. C. & Orcel, P. (1995) J. Bone Miner. Res. 10, 971–977. [PubMed]
15. Bord, S., Beavan, S., Ireland, D., Horner, A. & Compston, J. E. (2001) Bone 29, 216–222. [PubMed]
16. Fox, S. W., Haque, S. J., Lovibond, A. C. & Chambers, T. J. (2003) J. Immunol. 170, 3679–3687. [PubMed]
17. Erlebacher, A. & Derynck, R. (1996) J. Cell Biol. 132, 195–210. [PMC free article] [PubMed]
18. Kong, Y. Y., Feige, U., Sarosi, I., Bolon, B., Tafuri, A., Morony, S., Capparelli, C., Li, J., Elliott, R., McCabe, S., et al. (1999) Nature 402, 304–309. [PubMed]
19. Teng, Y.-Y. A., Nguyen, H., Gao, X., Kong, Y.-Y., Gorczynski, R. M., Singh, B., Ellen, R. P. & Penninger, J. M. (2000) J. Clin. Invest. 106, R59–R67. [PMC free article] [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem Substance links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...