NF-kappaB2 (p100) limits TNF-alpha-induced osteoclastogenesis.

Abstract

Bone undergoes a continuous cycle of renewal, and osteoclasts--the cells responsible for bone resorption--play a pivotal role in bone homeostasis. This resorption is largely mediated by inflammatory cytokines such as TNF-alpha. In this issue of the JCI, Yao et al. demonstrate that the NF-kappaB precursor protein NF-kappaB2 (p100) acts as a negative regulator of osteoclastogenesis (see the related article beginning on page 3024). TNF-alpha induced a sustained accumulation of p100 in osteoclast precursors, and TNF-alpha-induced osteoclast formation was markedly increased in Nfkb2-/- mice. They also found that TNF receptor-associated factor 3 (TRAF3) is involved in the posttranslational regulation of p100 expression. These results suggest that blockade of the processing of p100 is a novel strategy to treat TNF-alpha-related bone diseases such as RA.

Full text

Page 1

????????????
The Journal of Clinical Investigation http://www.jci.org
???κ?????????????????????α?????????
??????????????????
Sakae Tanaka1 and Hiroyasu Nakano2
1Department of Orthopaedic Surgery, Faculty of Medicine, University of Tokyo, Bunkyo-ku, Tokyo, Japan. 2Department of Immunology,
Juntendo University School of Medicine, Bunkyo-ku, Tokyo, Japan.
Bone undergoes a continuous cycle of renewal, and osteoclasts — the cells
responsible for bone resorption — play a pivotal role in bone homeosta-
sis. This resorption is largely mediated by inflammatory cytokines such as
TNF-α. In this issue of the JCI, Yao et al. demonstrate that the NF-κB precur-
sor protein NF-κB2 (p100) acts as a negative regulator of osteoclastogenesis
(see the related article, doi:10.1172/JCI38716). TNF-α induced a sustained
accumulation of p100 in osteoclast precursors, and TNF-α–induced osteo-
clast formation was markedly increased in Nfkb2–/– mice. They also found
that TNF receptor–associated factor 3 (TRAF3) is involved in the posttrans-
lational regulation of p100 expression. These results suggest that blockade
of the processing of p100 is a novel strategy to treat TNF-α–related bone
diseases such as RA.
Osteoclasts are primarily implicated in
physiologic and pathologic bone resorp-
tion. They are derived from hematopoi-
etic stem cells, and their differentiation is
critically regulated by supporting cells and
cytokines. The important cytokines regu-
lating osteoclast differentiation are M-CSF
and RANKL (1). In particular, the central
role of RANKL, which belongs to the TNF-α
superfamily, has been demonstrated by the
fact that the targeted disruption of Rankl
or its receptor Rank induced severe osteope-
trosis (increased bone density) in mice due
to the complete lack of osteoclasts, while
knockout of osteoprotegerin, a natural
antagonist of RANKL, conversely caused
marked bone loss (2).
The remarkable clinical success of anti–
TNF-α therapies such as anti–TNF-α
antibody and soluble TNF receptor has
established a critical role of TNF-α in
inflammatory diseases such as RA and
Crohn disease (3). Anti–TNF-α strategies
not only ameliorate the inflammatory
conditions in these disorders but also sup-
press bone erosion in RA, indicating an
essential role of TNF-α in the pathologic
bone destruction. However, in spite of such
strong clinical evidence, the relationship
between TNF-α and RANKL signaling in
osteoclast development is not necessarily
clear. The osteoclastogenic effect of TNF-α
independent of RANKL has been particu-
larly controversial. Although several stud-
ies have demonstrated that TNF-α directly
promotes osteoclast formation in vitro in
the absence of RANKL (4–6), the ability of
TNF-α–induced osteoclast formation is
limited, and the administration of TNF-α
does not induce osteoclast formation in
Rank-deficient mice in vivo (7). This may
be at least partly because RANK, but not
TNF receptor 1 or TNF receptor 2, recruits
the adaptor molecule TNF receptor–associ-
ated factor 6 (TRAF6), which is essential for
osteoclast development. However, the pos-
sibility that molecule(s) induced by TNF-α
negatively regulate osteoclast differentia-
tion has not been excluded.
???κ??????????????????????
?????????????????????????
?????????????????????????????
NF-κB is a collective term referring to
dimeric transcription factors that belong
to the Rel family, and NF-κB is involved in
various aspects of physiologic and patho-
logic events. NF-κB is composed of five
members: RelA (p65), RelB, c-Rel, NF-κB1
(p50 and its precursor p105), and NF-κB2
(p52 and its precursor p100). It is currently
well known that two major signaling path-
ways, the canonical or classical pathway
and the noncanonical or alternative path-
way, lead to NF-κB activation (8) (Figure 1).
In the canonical pathway, the RelA/p50
complex is sequestrated with inhibitors of
κB (IκBs) in the cytoplasm. After a chal-
lenge with inflammatory stimuli such as
TNF-α, IL-1, and LPS, phosphorylation
and subsequent degradation of IκBs are
rapidly induced, and the released RelA/p50
complex translocates to the nucleus. On
the other hand, the noncanonical pathway
is induced by CD40 ligand, RANKL, and
lymphotoxin β (LTβ). Upon stimulation
with these ligands, NF-κB2 (p100) retains
RelB in the cytoplasm as the RelB/p100
complex is processed to p52, then the RelB/
p52 complex translocates to the nucleus.
Notably, CD40 ligand, RANKL, and LTβ
activate both the canonical and noncanon-
ical pathways, whereas TNF-α, IL-1, and
LPS solely activate the canonical pathway.
The essential role of NF-κB in osteoclast
development was first observed in knock-
out mice. Although targeted disruption of
either Nfkb1 or Nfkb2 alone did not affect
skeletal development, double knockout of
these genes induced osteopetrosis in mice
due to a defect in osteoclast differentiation
(9, 10). These results suggest that there are
redundant roles of the canonical and non-
canonical NF-κB pathways in osteoclast
differentiation. NF-κB–inducing kinase
(NIK) is known to be essential for the phos-
phorylation and subsequent processing of
p100 to p52, and Novack et al. showed that
deletion of the Nik gene resulted in the
accumulation of p100 in osteoclast pre-
cursors, which caused impaired osteoclast
differentiation in vitro by retaining the
RelB/p100 complex in the cytoplasm (11).
They also found that the deletion of Nfkb2
restored the impaired osteoclastogenesis in
Nik–/– precursors (12).
In this issue of the JCI, Yao et al. (13)
provided in vitro and in vivo evidence that
p100 is induced by TNF-α in osteoclast
precursors and acts as a potent negative
regulator of TNF-α–induced osteoclast for-
mation (Figure 1). They found that TNF-α
induced a sustained accumulation of p100
in osteoclast precursors, while RANKL
more efficiently processed p100 to p52
through NIK activation. To analyze the
Conflict of interest: The authors have declared that no
conflict of interest exists.
Citation for this article: J. Clin. Invest. doi:10.1172/
JCI40629.

Page 2

????????????
The Journal of Clinical Investigation http://www.jci.org
effect of p100 in osteoclast development,
they investigated whether the deletion of
Nfkb2 promotes TNF-α–induced osteo-
clast formation. TNF-α–induced osteoclast
formation was increased in bone marrow
cells from Nfkb2–/– mice to a level com-
parable to that induced by RANKL treat-
ment. Consistent with the in vitro obser-
vation, bone resorption was significantly
increased in Nfkb2–/– mice compared with
WT mice when TNF-α was injected over
the calvaria. Although there is no obvious
basal bone phenotype in Nfkb2–/– mice (7),
the results strongly suggest that TNF-α–
dependent upregulation of p100 acts as a
negative regulator for osteoclast develop-
ment under pathologic conditions. TNF-α
treatment also upregulated osteoclastogen-
esis in Rank–/–Nfkb2–/– and Rankl–/–Nfkb2–/–
mice both in vitro and in vivo, confirming
that the enhancement of TNF-α–induced
osteoclast formation in Nfkb2–/– mice does
not depend on secondary production of
RANKL by TNF-α.
To further substantiate a negative regu-
latory role of p100 in TNF-α–induced
osteoclast formation under RA-like inflam-
matory conditions, Yao et al. (13) crossed
Tnfa-Tg mice with Nfkb2–/– mice. Tnfa-Tg/
Nfkb2–/– mice spontaneously developed
more severe inflammation and joint ero-
sion along with an increase in osteoclast
number compared with Tnfa-Tg/Nfkb2+/–
mice. Collectively, these studies showed
that p100 acts as a negative regulator for
TNF-α–induced osteoclast formation
under pathologic conditions.
Finally, Yao et al. (13) presented an inter-
esting observation that TRAF3 is involved
in the posttranslational regulation of
p100 expression. They found that TNF-α
increased the protein level of TRAF3 by
suppressing TRAF3 degradation, which was
reversed by RANKL. Previous studies have
shown that TRAF3 binds to and promotes
degradation of NIK by forming a complex
with TRAF2 and cellular inhibitor of apop-
tosis protein 1/2 (c-IAP1/2), and therefore,
stabilization of TRAF3 by TNF-α may
decrease the level of NIK, resulting in the
upregulation of p100 (14–16). Conversely,
knockdown of Traf3 using siRNA promoted
TNF-α–induced osteoclast formation
through downregulation of p100 (13).
Dejardin et al. (17) and Novack et al. (11)
reported that an agonistic antibody against
LTβ receptor or TNF-α upregulated the
expression of p100 in an NF-κB–dependent
fashion. Taken together, these studies have
shown that the expression level of p100 is
regulated transcriptionally by NF-κB and/
or posttranslationally by TRAF3, although
the detailed molecular mechanisms are not
fully elucidated (Figure 1).
????????????????????????????????????
???????????
There still remain many questions to be
answered. First, how does TNF-α cause
pathologic bone resorption in vivo even
though it induces a negative regulator for
osteoclastogenesis, i.e., p100? Is it because
????????
??????????????????????????????????????????α??????????????????????????????????????????????????????????????????????????????????????????
???????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????
??????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????
???κ??????????????????????????????????????????????????????????????????????????????????????????????????α?????????????????????????????????????????
?????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????
?????????????????????????????????????????????????????????????????????????????????????????????????????κ?????????????????????????????????????
???????????????????????????????????????????α???????????????????????????????????????????????????????????????????????????????????????????????
????????????????????????????????????????????????????????????????????????????????????????????????α????????????????????????????????????????????
????????????????????????????????????????????????????????????????????????

Page 3

????????????
The Journal of Clinical Investigation http://www.jci.org
TNF-α–induced upregulation of p100 is
somehow blocked under inflammatory
conditions, such as in that in RA synovial
joints, or do pathologic conditions induce
signaling molecule(s) that suppress the
inhibitory effect of p100? Second, while
p100 acts as an IκB-like molecule that
binds to RelB, p100 also binds to the RelA/
p50 dimer (11, 18). Given that Rela–/– or
Relb–/– single-knockout mice do not show
an altered bone phenotype, p100 may sup-
press TNF-α–induced osteoclast formation
by retaining both RelA/p50 complex and
RelB in the cytoplasm. Thus, it will be inter-
esting to investigate the bone phenotype of
Rela–/–Relb–/– mice. Third, the mechanism of
how TNF-α suppresses the degradation of
TRAF3 has not been clarified. It will also be
intriguing to test whether TNF-α–induced
osteoclast formation is enhanced in Traf3–/–
mice. Given that Traf3–/– mice die soon after
birth (14), transfer of Traf3–/– bone marrow
cells to WT mice would be a feasible way
to investigate the role of TRAF3 in TNF-α–
induced osteoclast formation in vivo.
Finally, what is the role of bone-forming
osteoblasts in TNF-α–induced pathologic
bone resorption, since TNF-α is known to
induce RANKL expression in osteoblasts?
These intriguing questions remain for
future study.
In conclusion, Yao et al. (13) have con-
vincingly demonstrated that NF-κB2p100
plays a negative role in suppressing TNF-α–
induced osteoclast formation under patho-
logic conditions using various animal
models. Thus, blockade of the processing
of p100 might be a novel strategy to treat
various bone diseases such as RA, in which
TNF-α–induced osteoclast formation plays
a crucial role in the progression of diseases.
Address correspondence to: Sakae Tanaka,
Department of Orthopaedic Surgery, Fac-
ulty of Medicine, The University of Tokyo,
7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033,
Japan. Phone: 81-3-3815-5411, ext. 33376;
Fax: 81-3-3818-4082; E-mail: tanakas-ort@
h.u-tokyo.ac.jp.
1. Tanaka, S., Nakamura, K., Takahasi, N., and Suda,
T. 2005. Role of RANKL in physiological and path-
ological bone resorption and therapeutics target-
ing the RANKL-RANK signaling system. Immunol.
Rev. 208:30–49.
2. Theill, L.E., Boyle, W.J., and Penninger, J.M. 2002.
RANK-L and RANK: T cells, bone loss, and mam-
malian evolution. Annu. Rev. Immunol. 20:795–823.
3. Scott, D.L., and Kingsley, G.H. 2006. Tumor necro-
sis factor inhibitors for rheumatoid arthritis.
N. Engl. J. Med. 355:704–712.
4. Azuma, Y., Kaji, K., Katogi, R., Takeshita, S., and
Kudo, A. 2000. Tumor necrosis factor-alpha induc-
es differentiation of and bone resorption by osteo-
clasts. J. Biol. Chem. 275:4858–4864.
5. Kim, N., et al. 2005. Osteoclast differentiation inde-
pendent of the TRANCE-RANK-TRAF6 axis. J. Exp.
Med. 202:589–595.
6. Kobayashi, K., et al. 2000. Tumor necrosis factor
alpha stimulates osteoclast differentiation by a
mechanism independent of the ODF/RANKL-
RANK interaction. J. Exp. Med. 191:275–286.
7. Li, J., et al. 2000. RANK is the intrinsic hematopoietic
cell surface receptor that controls osteoclastogenesis
and regulation of bone mass and calcium metabo-
lism. Proc. Natl. Acad. Sci. U. S. A. 97:1566–1571.
8. Bonizzi, G., and Karin, M. 2004. The two NF-kap-
paB activation pathways and their role in innate and
adaptive immunity. Trends Immunol. 25:280–288.
9. Franzoso, G., et al. 1997. Requirement for NF-kap-
paB in osteoclast and B-cell development. Genes
Dev. 11:3482–3496.
10. Iotsova, V., et al. 1997. Osteopetrosis in mice
lacking NF-kappaB1 and NF-kappaB2. Nat. Med.
3:1285–1289.
11. Novack, D.V., et al. 2003. The IkappaB function of
NF-kappaB2 p100 controls stimulated osteoclasto-
genesis. J. Exp. Med. 198:771–781.
12. Vaira, S., et al. 2008. RelB is the NF-kappaB sub-
unit downstream of NIK responsible for osteo-
clast differentiation. Proc. Natl. Acad. Sci. U. S. A.
105:3897–3902.
13. Yao, Z., Xing, L., and Boyce, B.F. 2009. NF-κB
p100 limits TNF-induced bone resorption in mice
by a TRAF3-dependent mechanism. J. Clin. Invest.
doi:10.1172/JCI38716.
14. He, J.Q., et al. 2006. Rescue of TRAF3-null mice
by p100 NF-kappa B deficiency. J. Exp. Med.
203:2413–2418.
15. Liao, G., Zhang, M., Harhaj, E.W., and Sun, S.C.
2004. Regulation of the NF-kappaB-inducing
kinase by tumor necrosis factor receptor-associ-
ated factor 3-induced degradation. J. Biol. Chem.
279:26243–26250.
16. Skaug, B., Jiang, X., and Chen, Z.J. 2009. The role
of ubiquitin in NF-kappaB regulatory pathways.
Annu. Rev. Biochem. 78:769–796.
17. Dejardin, E., et al. 2002. The lymphotoxin-beta
receptor induces different patterns of gene expres-
sion via two NF-kappaB pathways. Immunity.
17:525–535.
18. Basak, S., et al. 2007. A fourth IkappaB protein
within the NF-kappaB signaling module. Cell.
128:369–381.
???????????????????????????????????????????????
?????????????????????
Gregory A. Dekaban and Sakina Thawer
BioTherapeutics Research Laboratory, Molecular Brain Research Group, Robarts Research Institute and Department of Microbiology and Immunology,
The University of Western Ontario, London, Ontario, Canada.
The role of B cells and autoimmunity as contributing factors to poor neu-
rological outcomes following spinal cord injury (SCI) is poorly understood.
The study by Ankeny et al., in this issue of the JCI, identifies a new immu-
nopathological mechanism arising after SCI in mice (see the related article,
doi:10.1172/JCI39780). The study shows that B cells produce pathogenic
antibodies that impair lesion repair, resulting in worse neurological out-
come. This new understanding of SCI disease pathogenesis, if confirmed in
humans, reveals potential avenues for the development of novel neuropro-
tective immunotherapies.
Conflict of interest: The authors have declared that no
conflict of interest exists.
Citation for this article: J. Clin. Invest. doi:10.1172/
JCI40839.
Primary trauma to the CNS initiates a
series of interrelated responses, including
edema, excitotoxicity, and inflammation,
that lead to secondary injury, resulting in
further expansion of the initial lesion and
additional loss of neurological function.
The treatment of neuroinflammation in
the context of both traumatic brain injury
and spinal cord injury (SCI) still lacks a
standard, universally accepted therapy
that leads to improved neurological out-
comes. There exists a clear, unmet medi-
cal need for an effective antiinflammatory
treatment for the acute and chronic stages
of traumatic brain injury and SCI arising
in general civilian as well in injured mili-
tary personnel populations. Currently, an
important area of SCI research focuses on