Inhibition of RANKL-mediated osteoclast differentiation by selective TRAF6 decoy peptides

Biochemical and Biophysical Research Communications 359 (2007) 510–515 www.elsevier.com/locate/ybbrc

Inhibition of RANKL-mediated osteoclast differentiation by selective TRAF6 decoy peptides Ann T. Poblenz a

a,1

, Joerg J. Jacoby

a,2

, Sujay Singh b, Bryant G. Darnay

a,*

Department of Experimental Therapeutics, The University of Texas MD Anderson Cancer Center, Box 143, 1515 Holcombe Blvd., Houston, TX 77030, USA b Imgenex, Inc., San Diego, CA 92121, USA Received 17 May 2007 Available online 30 May 2007

Abstract RANK and RANKL are essential mediators of osteoclastogenesis. RANK interacts with members of the tumor necrosis factor receptor-associated factor (TRAF) family, of which TRAF6 is the critical signaling molecule. We identified a unique TRAF6-binding motif in RANK, which was subsequently co-crystallized with TRAF6 revealing distinct molecular interactions. A cell-permeable TRAF6 decoy peptide (T6DP) was shown to specifically target TRAF6 and inhibit RANKL-mediated signaling. In this study, we identified a core motif for binding to TRAF6 by generating a series of deletion mutants linked via palmitate as a means to internalize the peptide, thus making a smaller scaffold for intracellular delivery. The core motif of RKIPTEDEY inhibited RANKL-mediated osteoclastogenesis and bone resorption. In contrast, TRAF2/5 decoy peptides appeared to have no affect. Thus, disruption of the RANK–TRAF6 interaction may prove useful as a novel target for the development of a small molecule therapeutic agent for the treatment of bone-related diseases. Published by Elsevier Inc. Keywords: Osteoclast; Bone; TRAF6; TRAF2; TRAF5; RANK; RANKL; Peptide

Osteoclast differentiation and activation plays a central role in the development and maintenance of normal bone tissue, which requires osteoblastic matrix deposition and osteoclastic resorption to be closely coordinated. Interference with the process of osteoclastogenesis alters the kinetics of bone remodeling resulting in abnormal bone development [1,2].

Abbreviations: TNF, tumor necrosis factor; GST, glutathione S-transferase; TRAF, TNF receptor-associated factor; RANK, receptor activator of NF-jB; HEK, human embryonic kidney; NF-jB, nuclear factor-jB; RANKL, RANK ligand; TRAP, tartrate-resistant acid phosphatase; SDS, sodium dodecylsulfate. * Corresponding author. Fax: +1 713 745 6133. E-mail address: [email protected] (B.G. Darnay). 1 Present address: Oncology Clinical Science Liaison, Merck & Co., USA. 2 Present address: Department of Thoracic/Head & Neck Medical Oncology, The University of Texas MD Anderson Cancer Center, USA. 0006-291X/$ - see front matter Published by Elsevier Inc. doi:10.1016/j.bbrc.2007.05.151

RANK and RANKL are essential mediators of osteoclastogenesis and have been implicated in various diseases, which include rheumatoid arthritis, osteoporosis, giant cell tumor of bone, Paget’s disease, metastatic breast cancer, multiple myeloma, and familial expansile osteolysis [2]. Osteoprotegerin (OPG) is a soluble, decoy receptor that inhibits RANKL from binding to its cell surface receptor. While knockout mice lacking OPG develop severe osteoporosis, mice lacking either RANKL or RANK exhibit severe osteopetrosis due to a lack of osteoclasts [2]. Thus, RANKL and OPG are the governing factors that regulate normal bone homeostasis. Like most members of the TNF receptor superfamily, the cytoplasmic domain of RANK interacts with adaptor proteins of the TRAF family, which participate in activation of the transcription factor NF-jB and the MAP kinase pathways [3–5]. RANK interacts with TRAF1–3, 5, and 6 [6]. While TRAF1–3 and 5 bind to the C-terminal tail of RANK, TRAF6 binds to a unique motif located within

A.T. Poblenz et al. / Biochemical and Biophysical Research Communications 359 (2007) 510–515

the membrane proximal region of RANK [7], which was subsequently confirmed by crystallographic analysis [8]. Among all the TRAF knockout studies, only TRAF6deficient mice displayed bone abnormalities in that they developed osteopetrosis due to a defect in osteoclastogenesis [9–11]. RANKL signaling is disrupted in cells derived from these mice revealing a major role of TRAF6 in RANK signaling. Recently, multiple TRAF6-binding sites in RANK have been reported, which may allow for qualitative and quantitative regulations in osteoclast formation [12,13]. We have shown that disruption of RANK–TRAF6 interaction with cell-permeable TRAF6 decoy peptides prevents RANKL-mediated signaling and osteoclast differentiation [8]. Collectively, these data support the indispensable role of TRAF6 in RANK signaling and in terminal differentiation of osteoclast progenitors. However, the molecular mechanism by which TRAF6 exerts its biological activity in RANK signaling is not well understood. The role of TRAF2 and TRAF5 in osteoclastogenesis is not as clear as for TRAF6. Mice lacking TRAF5 are healthy and do not show any obvious bone phenotype under normal condition, whereas a lack of TRAF2 in mice leads to embryonic and neonatal lethality, hindering the effort to specify the role of TRAF2 for bone development and remodeling [6]. As TRAF6 is the critical adaptor molecule in the RANK-mediated osteoclast differentiation, we had earlier developed cell-permeable TRAF6 decoy peptides (T6DP), which specifically targets the interaction of TRAF6 with RANK and demonstrated that T6DPs inhibit RANKL-mediated signaling and osteoclast differentiation [8]. In this report, we sought to define a minimal scaffold for binding to TRAF6 and identified a core motif of RKIPTEDEY, which notably is the same sequence that was co-crystallized with TRAF6 [8]. Furthermore, we have employed the use of a palmitic acid linkage at the N-terminus as a means to internalize the peptide, thus making a smaller scaffold for intracellular delivery. In an effort to identify a role for TRAF2 and TRAF5 in RANKL-mediated signaling, we utilized a similar strategy by constructing a palmitic acid-linked decoy peptide that is capable of binding to TRAF2 and TRAF5, designated T2/5DP. Taken together, we present evidence that interfering with the RANK–TRAF6 interaction by targeting TRAF6 with a small, cell-permeable decoy peptide abolishes osteoclast differentiation. These data support the essential role of TRAF6 in osteoclast development and provide a potential novel target for the advancement of a therapeutic agent to treat bone lesions associated with cancer or various diseases such as osteoporosis and arthritis. Materials and methods Cell lines, reagents, and antibodies. Human embryonic kidney 293 (HEK293) cells were obtained from the American Type Culture Collection (ATCC) (Rockville, MD) and cultured as previously described [7,8,14]. Monoclonal anti-FLAG antibody and a tartrate-resistant acid phospha-

511

tase (TRAP) kit were purchased from Sigma. Recombinant mouse RANKL and GST-RANKcd were purified essentially as described [7,14,15]. Plasmids and transfections. Expression vectors for FLAG-TRAF2, FLAG-TRAF5, and FLAG-TRAF6 have been described previously [7,15–17]. Transfection of HEK293 cells was performed essentially as described [7,16]. Osteoclast differentiation. For mouse bone marrow-derived monocytes (BMM), tibiae and femurs were excised from 6- to 8-week-old C57BL/6 mice and the bone marrow cells were flushed out with incomplete a-MEM, centrifuged at 1200 rpm for 5 min, and resuspended in 5 ml incomplete a-MEM. BMM cells were incubated in 20 ml Red Blood Cell lysis buffer (8.3 g/l NH4Cl, 1 g/l sodium bicarbonate, 0.4 g/l EDTA) at room temperature for 1–2 min after which a-MEM supplemented with 10% FBS and 100 U/ml penicillin (complete a-MEM) was added and BMM cells were centrifuged for 5 min at 1200 rpm. BMM cells were then plated in 10 cm dishes with 10 ml of complete a-MEM and cultured for 24 h. Nonadherent cells were collected and centrifuged for 5 min at 1200 rpm, counted, and seeded at a concentration of 2.5 · 104 cells/well in a 96-well plate for cytotoxicity assay or at 5.0 · 104 cells/well (24-well), 2.5 · 104 cells/well (48-well), or 5 · 103 cells/well (96-well) for osteoclast differentiation assays. Cells were cultured for 3 days in the presence of M-CSF (10 ng/ml) before they were washed and used for further experiments. Peptides were added at the indicated concentration with RANKL (100 ng/ ml) and M-CSF (10 ng/ml) and after 2 days media was supplemented with fresh M-CSF and RANKL until osteoclasts appeared (approximately on day 4 or 5) and then stained for TRAP. The mice were housed in a temperature-controlled environment with free access to food and water. All the protocols and procedures were approved by the Institutional Animal Care and Use Committee at The University of MD Anderson Cancer Center. GST-fusion pull-down assays. Bacterial purified GST or GST-RANKcd bound to glutathione–agarose beads was mixed HEK293 cell lysates that were previously transfected with FLAG-TRAF2, TRAF5 or TRAF6 in buffer A (20 mM Tris, pH 7.4, 250 mM NaCl, 1 mM DTT, 1 mM sodium orthovanadate, 2 mM EDTA, 1% Triton X-100, 2 mg/ml leupeptin, and 2 mg/ml aprotinin). Peptides (100 lM) were added and samples were allowed to rotate end-over-end for 1 h at 4 oC followed by three washes in buffer A and two washes in low salt buffer (20 mM Tris pH 7.4, 25 mM NaCl, and 1 mM DTT). Bound proteins were eluted in SDS-sample buffer and boiled for 5 min, subjected to SDS–PAGE, and immunoblotted with anti-FLAG. Peptide synthesis. In brief, peptides with and without palmitate containing putative TRAF-binding sequences were chemically synthesized with an N-terminal acetylation and C-terminal amidation. All peptides were synthesized in an automated peptide synthesizer (Symphony, Rainin Instruments, Co). Five milliliters of 100 mM palmitic acid with DIC (1,3diisopropylcarbodiimide) was activated, added to reaction vessel, mixed for 4 h and drained. The reaction was washed 3· with DMF for 30 s each and steps are repeated for a second coupling cycle. The molecular mass of each peptide was verified by MALDI-TOF mass spectrometry. Cytotoxicity assay. Cytotoxicity of TRAF decoy peptides on BMM was performed in 96-well flat-bottom tissue culture dishes. Briefly, BMMs (2.5 · 104) were plated and treated as described above. TRAF peptides were diluted in culture media and added to the wells in 2-fold serial dilutions. Cells were incubated for 72 h and the remaining adherent cells were stained with crystal violet (0.5% in 20% methanol) and solubilized with Sorenson’s buffer (0.1 M sodium citrate, pH 4.2, in 50% ethanol). Absorbance was measured at 630 nm using an EL800 universal microplate reader. Bone-resorption assay. BMM cells (5 · 103) were seeded in a 96-well OsteoAssay plate from Cambrex (Rockland, ME) and initially treated with RANKL (100 ng/ml) and M-CSF (10 ng/ml) with or without 40 lM TRAF peptides. Cell cultures were supplemented with fresh M-CSF at day 3 and a 20 ll aliquot of the supernatant was used to evaluate the bone resorption from the OsteoAssay plate at days 2, 4, 5, and 7. Bone resorption was evaluated by measuring collagen I release from the OsteoAssay plate using CrossLaps for Culture ELISA kit from Nordic

A.T. Poblenz et al. / Biochemical and Biophysical Research Communications 359 (2007) 510–515

From our previous work, we had identified a novel TRAF6-binding domain in RANK (residues 342–349), which binds to only TRAF6, and not TRAF2 or TRAF5 [7,15] and was subsequently co-crystallized with the TRAF-C domain of TRAF6 and its structure solved [8]. We previously developed cell-permeable TRAF6 decoy peptides (T6DP), which specifically targets the interaction of TRAF6 with RANK and demonstrated that L-T6DP-1 inhibits RANKL-mediated signaling and osteoclast differentiation of RAW264.7 cells and primary mouse bone marrow-derived monocytes [8]. Although L-T6DP-1 was shown to be effective, it consists of 34 amino acids, which makes the peptide very costly and perhaps sensitive to endopeptidase, therefore unsuitable as a possible drug. Therefore, to facilitate peptide uptake by cultured cells, we developed a peptide with an N-terminal linkage to palmitic acid thus making a smaller scaffold for intracellular delivery. Furthermore, in an effort to define a minimal T6DP scaffold, we synthesized various N- and C-terminal mutants surrounding the TRAF6 consensus sequence (Fig. 1A). Additionally, to investigate the potential role of TRAF2 and TRAF5 in osteoclast differentiation, we constructed a decoy peptide containing the TRAF2/5 consensus sequence (Fig. 1B). As negative controls for these experiments, we also synthesized scrambled versions of binding sequences.

A

TRAF6 Consensus P-X-E-X-X-Ar/Ac

T6DP1

J- RKIPTEDEYTDRPSQ

T6DP2 T6DP3 T6DP4 T6DP5

J- RKIPTEDEYTDR J- RKIPTEDEY JPTEDEY JPTEDEYTDR

T6DP3sc J- EPYKDTREI

B

Defining a minimal scaffold of T6DP for inhibiting osteoclast differentiation In the next set of experiments, we investigated the effect of the different decoy peptides on RANKL-mediated osteoclast differentiation. Mouse BMMs were cultured with RANKL and M-CSF with different concentrations of decoy peptides and after 5 days the cells were fixed, stained for TRAP, and the number of multi-nucleated osteoclasts was counted. The minimal region of T6DP that effectively prevented osteoclast differentiation was T6DP3 with the core motif RKIPTEDEY, whereas peptides lacking the first three amino acids RKI (T6DP4 and T6DP5) were unable to significantly inhibit osteoclast differentiation (Fig. 2). Peptides with a longer core motif (T6DP1

C

GSTGST RANKcd TRAF2

IB: FLAG

TRAF5 TRAF6

peptide:

T2/5DP

Generation of TRAF6 and TRAF2/5 decoy peptides

To verify that the TRAF decoy peptides are specific for their respective TRAF, we performed a GST pull-down assay using the RANK cytoplasmic domain, which binds TRAFs [6]. HEK293 cells were transfected with FLAGtagged TRAF2, TRAF5, or TRAF6 and the cells harvested and lysed. Equal protein lysates were incubated together with GST or GST-RANKcd in the presence or absence of the decoy peptides T6DP3 or T2/5DP followed by immunoblotting with anti-FLAG. T6DP3 prevented the interaction of RANKcd with TRAF6, but had no effect on the interaction of TRAF2 and TRAF5 with RANK (Fig. 1C). Conversely, T2/5DP completely blocked the interaction between RANK and TRAF2 or TRAF5, while not affecting the interaction of TRAF6 with RANK (Fig. 1C). These data confirm the specificity of these TRAF decoy peptides.

T6DP3

Results and discussion

Specificity of the TRAF decoy peptides

none

Bioscience Diagnostics (Portsmouth, VA). Absorbance was measured at 450 nm using the reading at 650 nm as reference with an EL800 universal microplate reader.

none

512

TRAF1,2,3, & 5 Consensus P/S/T/A-X-Q/E-E T2/5DP

J- KASRPVQEQGG

T2/5DPsc J- QEKQSRPGVGA Fig. 1. Consensus TRAF-binding sequences and specificity of the TRAF decoy peptides. (A, B) TRAF decoy peptides used in this study. Schematic diagram of consensus sequence for binding TRAF6 (A) and the other TRAFs (B) (J, palmitic acid; sc, scrambled; Ar, aromatic; Ac, acidic). (C) Specificity of the TRAF decoy peptides. The indicated peptides (100 lM) were mixed with HEK293 cells lysates programmed to express FLAG-tagged TRAF2, TRAF5, or TRAF6. These lysates were then mixed with glutathione–agarose beads containing either GST or GST-RANKcd and rotated for 1 h at 4 C. The beads were washed, and bound proteins eluted in SDS-sample buffer, subjected to SDS–PAGE, and immunoblotted with anti-FLAG antibody.

A.T. Poblenz et al. / Biochemical and Biophysical Research Communications 359 (2007) 510–515

513

Fig. 2. T6DP3 inhibits RANKL-mediated osteoclast differentiation. (A–C) Mouse BMMs were co-stimulated with M-CSF (10 ng/ml) and RANKL (100 ng/ml) in the absence (A) or presence of the indicated concentration of the TRAF decoy peptides in triplicate wells. After 5 days, the cells were stained for TRAP and representative pictures using a 10· objective were taken (B) and the total numbers of TRAP+, multi-nucleated osteoclasts were counted (C). T6BP control is T6DP3 lacking the palmitic acid linkage.

and T6DP2) were able to inhibit osteoclastogenesis, although to a lesser extent than T6DP3 (Fig. 2). In contrast, a scrambled version of T6DP3 (T6DP3sc) or a TRAF6 peptide without a palmitic acid linkage (T6BPc) was unable to inhibit osteoclast differentiation (Fig. 2). Additionally, none of the peptides presented toxic effects on these cells at the concentrations used in this assay (data not shown). Interestingly, the core motif RKIPTEDEY of T6DP3 is identical to the peptide sequence that was cocrystallized with TRAF6 [8]. Blocking TRAF2 and TRAF5 has no Effect on osteoclast differentiation To determine if TRAF2 or TRAF5 are essential for osteoclast differentiation, we used a TRAF2/5 decoy peptide to block the binding of TRAF2 and TRAF5 to RANK. While a small reduction in the number of osteoclasts was observed at the highest concentration used of the T2/5DP, the reduction was not statistically significant in comparison to the scrambled T2/5DP (Fig. 2). Similar to mice lacking TRAF2 or TRAF5, it appears that TRAF2 and TRAF5 are not essential for osteoclast differentiation.

T6DP3 prevents bone resorption We next asked whether TRAF decoy peptides could prevent bone resorption by measuring collagen I release in a human bone culture system. Indeed, RANKL-mediated bone resorption was prevented with T6DP3, but not with the scrambled version (Fig. 3). Notably, we observed a significant decrease of collagen I release in cells treated with T2/ 5DP as compared to its respective scrambled peptide (Fig. 3). These results clearly indicate a requirement of TRAF6 for osteoclast differentiation and bone resorption and a potential role of TRAF2 or TRAF5 in bone resorption. Requirement of TRAF6 in early stages of osteoclast differentiation In a final experiment, we asked at what time interval TRAF6 is required for osteoclast differentiation. For this experiment, we treated BMMs with 40 lm of T6DP3 at the same time as RANKL (time 0) or the peptide was added at 12, 24, or 48 h after initiation of osteoclast differentiation with RANKL. T6DP3 effectively blocked osteoclastogenesis when administrated up to 12 h after

514

A.T. Poblenz et al. / Biochemical and Biophysical Research Communications 359 (2007) 510–515

Absorbance (450 nm)

A

1.00

0.75

M-CSF

A

M-CSF RL T6DP3 T6DP3sc T2/5DP T2/5DPsc

0.50 0h

12 h

24 h

48 h

0.25 1500

0.00 Day 2 M-CSF

T6DP3sc

Day 4 RL

T2/5DP

Day 5

Day 7

T6DP3

T2/5DPsc

TRAP+ OC/well

B

B

M-CSF + RL

1000

500

0 M-CSF + RL:

+

+ 0h

+ + 12 h 24 h

+ 48 h

T6DP3 (Time After RL) Fig. 3. T6DP3 prevents bone resorption. (A, B) Mouse BMMs were seeded in a 96-well OsteoAssay plate in triplicate and initially treated with RANKL (100 ng/ml) and M-CSF (10 ng/ml) with or without TRAF decoy peptides (40 lM). Bone resorption was evaluated by measuring collagen I release from the OsteoAssay plate using CrossLaps for Culture ELISA kit on the indicated days (A). Results are representative of one assay repeated three times in triplicate. In a parallel experiment, after 7 days, the cells were stained for TRAP and representative pictures using a 10· objective were taken (B).

RANKL treatment, whereas 24 or 48 h after RANKL treatment the addition of T6DP3 did not have such a pronounced effect on the number of osteoclasts generated (Fig. 4). Hence, TRAF6 appears to be required in the early stages of RANKL-mediated osteoclast differentiation.

Conclusions Similar to the lack of osteoclast differentiation in primary TRAF6-deficient monocytes, the data presented here have established that the interaction of TRAF6 with RANK is critical for the differentiation of primary monocytes into functional bone resorbing osteoclasts. The objective of this study was to identify a small TRAF6 decoy peptide that when linked to palmitate prevents RANKLmediated osteoclast differentiation. Through deletion mapping experiments, T6DP3 (RKIPTEDEY) was identified as the most efficient inhibitor of the TRAF6 decoy peptides to prevent osteoclast formation. Interestingly, this sequence is

Fig. 4. TRAF6 is required for the early stages of osteoclast differentiation. (A, B) Mouse BMMs were seeded in a 96-well plates in triplicate and treated with RANKL (100 ng/ml) and M-CSF (10 ng/ml) in the absence or presence of T6DP3 (40 lM), which was added at the indicated times after RANKL treatment. After 5 days, the cells were stained for TRAP and representative pictures using a 10· objective were taken (A) and the total numbers of TRAP+, multi-nucleated osteoclasts were counted (B).

identical to the peptide used for structure determination [8]. Notably, we have also provided evidence that linking palmitate to the N-terminus of a TRAF6-binding peptide (or possibly any small peptide) allows for efficient delivery into the cells where it exhibits functional activity. Furthermore, we observed that interaction of RANK with TRAF2 or TRAF5 does not appear to be required for osteoclast differentiation, but may have a functional role in bone resorption, which at this time is unclear. By using these decoy peptides, TRAF6 appears to be required for the initial programming of the monocytes into osteoclasts and may not be required once the initial transcription machinery have been induced. In summary, our results support the critical role for TRAF6 in osteoclast development and provide a potential new target for small molecule discovery for the treatment of unwanted bone destruction observed in metabolic bone disorders and metastatic cancer of the bone. Acknowledgment This work was supported in part by institutional startup funds from The University of Texas MD Anderson Cancer Center to B.G.D.

A.T. Poblenz et al. / Biochemical and Biophysical Research Communications 359 (2007) 510–515

References [1] S.L. Teitelbaum, F.P. Ross, Genetic regulation of osteoclast development and function, Nat. Rev. Genet. 4 (2003) 638–649. [2] H. Takayanagi, Mechanistic insight into osteoclast differentiation in osteoimmunology, J. Mol. Med. 83 (2005) 170–179. [3] J.Y. Chung, Y.C. Park, H. Ye, H. Wu, All TRAFs are not created equal: common and distinct molecular mechanisms of TRAFmediated signal transduction, J. Cell Sci. 115 (2002) 679–688. [4] H. Wajant, P. Scheurich, Analogies between Drosophila and mammalian TRAF pathways, Prog. Mol. Subcell. Biol. 34 (2004) 47–72. [5] H. Wu, J.R. Arron, TRAF6, a molecular bridge spanning adaptive immunity, innate immunity and osteoimmunology, Bioessays 25 (2003) 1096–1105. [6] B.G. Darnay, A. Besse, A.T. Poblenz, B. Lamothe, J.J. Jacoby, TRAFs in RANK signaling in TRAFs, in: H. Wu (Ed.), Landes Bioscience, 2005. [7] B.G. Darnay, J. Ni, P.A. Moore, B.B. Aggarwal, Activation of NFkappaB by RANK requires tumor necrosis factor receptor-associated factor (TRAF) 6 and NF-kappaB-inducing kinase. Identification of a novel TRAF6 interaction motif, J. Biol. Chem. 274 (1999) 7724–7731. [8] H. Ye, J.R. Arron, B. Lamothe, M. Cirilli, T. Kobayashi, N.K. Shevde, D. Segal, O.K. Dzivenu, M. Vologodskaia, M. Yim, K. Du, S. Singh, J.W. Pike, B.G. Darnay, Y. Choi, H. Wu, Distinct molecular mechanism for initiating TRAF6 signalling, Nature 418 (2002) 443–447. [9] M.A. Lomaga, W.C. Yeh, I. Sarosi, G.S. Duncan, C. Furlonger, A. Ho, S. Morony, C. Capparelli, G. Van, S. Kaufman, A. van der Heiden, A. Itie, A. Wakeham, W. Khoo, T. Sasaki, Z. Cao, J.M. Penninger, C.J. Paige, D.L. Lacey, C.R. Dunstan, W.J. Boyle, D.V. Goeddel, T.W. Mak, TRAF6 deficiency results in osteopetrosis and defective interleukin-1, CD40, and LPS signaling, Genes Dev. 13 (1999) 1015–1024.

515

[10] A. Naito, S. Azuma, S. Tanaka, T. Miyazaki, S. Takaki, K. Takatsu, K. Nakao, K. Nakamura, M. Katsuki, T. Yamamoto, J. Inoue, Severe osteopetrosis, defective interleukin-1 signalling and lymph node organogenesis in TRAF6-deficient mice, Genes Cells 4 (1999) 353–362. [11] A. Naito, H. Yoshida, E. Nishioka, M. Satoh, S. Azuma, T. Yamamoto, S. Nishikawa, J. Inoue, TRAF6-deficient mice display hypohidrotic ectodermal dysplasia, Proc. Natl. Acad. Sci. USA 99 (2002) 8766–8771. [12] Y. Kadono, F. Okada, C. Perchonock, H.D. Jang, S.Y. Lee, N. Kim, Y. Choi, Strength of TRAF6 signalling determines osteoclastogenesis, EMBO Rep. 6 (2005) 171–176. [13] J. Gohda, T. Akiyama, T. Koga, H. Takayanagi, S. Tanaka, J. Inoue, RANK-mediated amplification of TRAF6 signaling leads to NFATc1 induction during osteoclastogenesis, EMBO J. 24 (2005) 790–799. [14] T.C. Polek, M. Talpaz, B.G. Darnay, T. Spivak-Kroizman, TWEAK mediates signal transduction and differentiation of RAW264.7 cells in the absence of Fn14/TweakR. Evidence for a second TWEAK receptor, J. Biol. Chem. 278 (2003) 32317–32323. [15] B.G. Darnay, V. Haridas, J. Ni, P.A. Moore, B.B. Aggarwal, Characterization of the intracellular domain of receptor activator of NF-kappaB (RANK). Interaction with tumor necrosis factor receptor-associated factors and activation of NF-kappab and c-Jun Nterminal kinase, J. Biol. Chem. 273 (1998) 20551–20555. [16] B. Lamothe, A. Besse, A.D. Campos, W.K. Webster, H. Wu, B.G. Darnay, Site-specific Lys-63-linked tumor necrosis factor receptorassociated factor 6 auto-ubiquitination is a critical determinant of IkappaB kinase activation, J. Biol. Chem. 282 (2007) 4102–4112. [17] A. Wong, B. Lamothe, A. Lee, J. Schlessinger, I. Lax, FRS2 alpha attenuates FGF receptor signaling by Grb2-mediated recruitment of the ubiquitin ligase Cbl, Proc. Natl. Acad. Sci. USA 99 (2002) 6684– 6689.