Search tips
Search criteria 


Logo of molcellbPermissionsJournals.ASM.orgJournalMCB ArticleJournal InfoAuthorsReviewers
Mol Cell Biol. 1999 October; 19(10): 6665–6672.

TANK Potentiates Tumor Necrosis Factor Receptor-Associated Factor-Mediated c-Jun N-Terminal Kinase/Stress-Activated Protein Kinase Activation through the Germinal Center Kinase Pathway


Tumor necrosis factor (TNF) receptor-associated factors (TRAFs) are mediators of many members of the TNF receptor superfamily and can activate both the nuclear factor κB (NF-κB) and stress-activated protein kinase (SAPK; also known as c-Jun N-terminal kinase) signal transduction pathways. We previously described the involvement of a TRAF-interacting molecule, TRAF-associated NF-κB activator (TANK), in TRAF2-mediated NF-κB activation. Here we show that TANK synergized with TRAF2, TRAF5, and TRAF6 but not with TRAF3 in SAPK activation. TRAF2 and TANK individually formed weak interactions with germinal center kinase (GCK)-related kinase (GCKR). However, when coexpressed, they formed a strong complex with GCKR, thereby providing a potential mechanism for TRAF and TANK synergy in GCKR-mediated SAPK activation, which is important in TNF family receptor signaling. Our results also suggest that TANK can form potential intermolecular as well as intramolecular interactions between its amino terminus and carboxyl terminus. This study suggests that TANK is a regulatory molecule controlling the threshold of NF-κB and SAPK activities in response to activation of TNF receptors. In addition, CD40 activated endogenous GCKR in primary B cells, implicating GCK family proteins in CD40-mediated B-cell functions.

The tumor necrosis factor receptor (TNFR) family, consisting of over 20 known distinct members, plays important roles in controlling lymphocyte activation, acute-phase responses, and tumor progression (1, 24, 36, 52). Most are type I transmembrane proteins with characteristic cysteine-rich pseudorepeats in the extracellular region. They can be further classified, based on their cytoplasmic tails, into three groups. The death domain-containing receptor group, which includes TNFRI and Fas, contains a homologous region of about 80 amino acids (aa) called the death domain (21, 55). The members of the non-death domain-containing group, which includes TNFRII, CD40, and CD30, share little homology in their cytoplasmic tails. Members of the third group of this family, including TRID and osteoprotegerin, do not contain functional cytoplasmic tails and are believed to function as inhibitors by competing for ligands with the other two groups of the TNFR family (38, 49, 51). Most of the TNFR family proteins are able to send signals activating two distinct signal transduction pathways, nuclear factor κB (NF-κB) and the stress-activated protein kinase (SAPK; also known as c-Jun N-terminal kinase [JNK]), which lead to TNF-mediated transcriptional regulation for cell survival, cell proliferation and inflammatory responses.

Significant progress has recently been made in the study of TNF-mediated NF-κB and SAPK activation, with identification of the TNFR-associated factors (TRAF) (45). TRAF proteins either directly bind to the cytoplasmic tails of non-death domain-containing TNFRs or indirectly bind to death domain-containing TNFRs through intermediate molecules such as TRADD and RIP (4, 6, 16, 17, 32, 40, 45). In addition, TRAF proteins are involved in signaling through non-TNFR family proteins, including Epstein-Barr virus latent membrane protein 1, which plays an essential role in virus-mediated cell transformation, and the Toll family proteins, which belong to a family of receptors responding to various inflammatory stimuli such as interleukin-1 and lipopolysaccharide (32, 58). To date, six distinct TRAF proteins, TRAF1 through TRAF6, have been isolated. TRAF2, TRAF3, and TRAF6 are ubiquitously expressed, whereas TRAF1 and TRAF5 are preferentially expressed in the spleen, thymus, and lung, and TRAF4 expresses at low levels in normal tissues but is abundant in breast cancer cells (4, 6, 18, 20, 33, 45, 48). All TRAF family proteins have related TRAF-N and TRAF-C domains in their carboxyl-terminal regions (6). In addition, all TRAF family members except TRAF1 have two amino-terminal zinc-binding domains, including a ring finger and five zinc fingers (6). Overexpression of TRAF2, TRAF5, and TRAF6, but not TRAF3, in human embryonic kidney 293 cells activates both the NF-κB and SAPK pathways (2). With the recent identification of NF-κB-inducing kinase (NIK) and the I-κB kinases (IKK), a cascade of signals in the TNF-α-mediated NF-κB activation pathway, from TNF to TNFRs, TRAFs, NIK, IKKs and then to IκB and NF-κB/Rel family transcription factors, has been proposed (12, 28, 29, 41, 56, 60). In JNK activation, TNF-α is thought to signal through a three-tiered mitogen-activated protein kinase (MAPK) pathway, from the MAPK kinase kinase MEKK1, to the dual specificity MAPK kinase MKK/SEK/JNKK proteins, to phosphorylation of JNK, leading to the activation of the AP-1 transcription factor (11, 27, 30, 34, 47). Less clear, however, are the direct links between TNFR and MEKK1. Furthermore, the regulation of the NF-κB and SAPK pathways in response to the activation of various TRAF-associated receptors remains to be elucidated.

We previously isolated a TRAF-associated NF-κB activator (TANK) by yeast two-hybrid screening, using TRAF3 as a bait. TANK binds all known TRAF proteins except the predominantly nuclear TRAF4 and synergistically activates TRAF2-mediated NF-κB activation in a biphasic manner with increasing levels of TANK (5, 8). Here we report that TANK is also a mediator of TRAF-induced SAPK activation, which synergistically activated SAPK with TRAF2, TRAF5, and TRAF6 but not with TRAF3. This robust SAPK activation was potentially due to synergistic association of TANK and TRAF proteins with germinal center kinase (GCK) family kinases. A dominant-negative form of GCK-related kinase (GCKR) abrogated CD40 activation of SAPK in vivo, while endogenous GCKR was activated by the B-lymphocyte coreceptor, CD40. In addition, we suggest that TANK, which may form both inter- and intramolecular associations, can act as a regulator by controlling the activities of both the TRAF-mediated NF-κB and SAPK pathways.


Tissue culture and transfection.

For transfection studies, 293T human embryonic kidney cells were maintained in Dulbecco’s modified Eagle medium (Gibco BRL, Grand Island, N.Y.) supplemented with 10% fetal bovine serum (Omega Scientific, Tarzana, Calif.) and 1% penicillin and streptomycin at 37°C in a 5% CO2 atmosphere in a tissue culture incubator. Cells were transfected in 10-cm-diameter tissue culture dishes at densities of approximately 40% confluency, using a calcium phosphate method. Briefly, 435 μl of sterile water containing plasmid DNA was mixed with 65 μl of 2 M CaCl2 and 500 μl of 2× HEPES-buffered saline solution (50 mM HEPES [pH 7.05], 10 mM KCl, 12 mM dextrose, 280 mM NaCl, 1.5 mM Na2HPO4). After thorough mixing, the DNA solution was added to the culture dish. After 12 h, the medium was changed, and the cells were incubated for another 24 h before harvesting. HS-Sultan and Ramos cells were maintained in RPMI medium (Mediatech) supplemented with 10% fetal bovine serum and 1% penicillin and streptomycin at 37°C in a 5% CO2 atmosphere.

Plasmid constructs.

The plasmid constructs used in the experiments were as follows. Hemagglutinin epitope (HA)-tagged TRAF2, TRAF3, TRAF5, and TRAF6 were all subcloned in pEBB vectors and contained full-length murine cDNA of these molecules; glutathione S-transferase (GST)-tagged pEBG-TANK and Flag-tagged pCMV TANK contained full-length murine TANK cDNA; pEBB HA- and GST-tagged pEBG-TANK-C contained the carboxyl-terminal portion of TANK (aa 190 to 413); GST-tagged pEBG- and Flag-tagged pCMV TANK-N contained the amino-terminal portion of TANK (aa 1 to 168); full-length human CD40 and murine CD40L were both subcloned in pBABE vectors; pEBB HA-tagged TRAF2-C contained the C-terminal portion of murine TRAF2 (aa 94 to 501), with deletion of the zinc ring finger; the kinase-deficient mutant GST-tagged pEBG-DN-MEKK1 contained an amino acid change of lysine to arginine at position 447; GST-tagged pEBG-DN-SEK1 contained a mutation from lysine to arginine at position 129 and was thus kinase inactive; pCMV2-Flag-GCKR contained Flag-tagged wild-type GCKR; pCMV2-Flag-DN-GCKR contained the dominant-negative form of GCKR, which has a lysine-to-aspartic acid mutation at amino acid position 178 (50); pCR3.1-Flag-GLK contained Flag-tagged wild-type GCK-like kinase (GLK); and pCR3.1-Flag-DN-GLK contained the dominant-negative form of GLK, which has a lysine-to-glutamic acid mutation at position 35 (13). The Fyn and Btk expression constructs were previously described (7).

Cell lysis.

At 36 to 48 h after calcium phosphate transfection, the cells were washed once in ice-cold phosphate-buffered saline and lysed with 800 μl of radioimmunoprecipitation assay buffer (50 mM Tris-HCl [pH 7.5], 150 mM NaCl, 1% NP-40, 0.25% sodium deoxycholate, 1 mM EGTA, 5 mM NaF, 1 μg each of aprotinin, leupeptin, and pepstatin per ml, 1 mM Na3VO4, 1 mM phenylmethylsulfonyl fluoride) for 30 min. The cell lysate was transferred to a microcentrifuge tube and then centrifuged at 15,000 × g for 10 min at 4°C to remove the cell pellet.

In vitro kinase assays.

Equal amounts of supernatants from various transfections were incubated at 4°C on a rotating mixer, first with 1 μg of monoclonal anti-HA antibody or 1 μg of polyclonal anti-JNK1 antibody C-17 (Santa Cruz Biotechnology, Inc.) for 1 h and then with protein A/G-agarose (Calbiochem, San Diego, Calif.) for 1 h. Solutions for the JNK assay were made as previously described (15). The agarose beads with bound JNK molecules were washed twice in radioimmunoprecipitation assay buffer and twice in HEPES binding buffer. Remaining supernatant above the agarose beads was carefully removed, and the kinase reaction was performed in 30 μl of kinase buffer containing 1.0 μg of purified GST-Jun fusion protein containing the amino-terminal 79 aa [GST-Jun(1-79)] of c-Jun (GST-Jun expression plasmid kindly provided by Dennis Templeton, Case Western Reserve University) and 2 μCi of [γ-32P]ATP (NEN Life Science Products Inc., Boston, Mass.) per sample. The kinase reaction was carried out at 30°C for 30 min. and then terminated by addition into the reaction mixture of equal volumes of 2× sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) sample buffer and heating in a boiling water bath for 5 min. After SDS-PAGE on an 11% gel, the gel was dried and exposed to a Kodak X-ray film. Fold activation quantitated was with an the Alpha Imager 2000 densitometer (Alpha Innotech Corporation). Human tonsil B and T cells were prepared from tonsillar mononuclear cells by rosetting with aminoethylisothiouronium bromide-treated sheep erythrocytes as described previously (22). The B- and T-cell populations were greater than 97% pure, as assessed by flow cytometry with CD19 and CD3 monoclonal antibodies, respectively. B cells were stimulated with an antibody to either CD40, G28.5 (ATCC, Rockville, Md.), or a soluble CD40 ligand fusion protein, CD8-gp39 (gift from Marilyn Kehry, Boehringer Ingelheim Pharmaceuticals Inc.). Myelin basic protein (MBP) in vitro kinase assays were performed as previously described (22, 39).


Coimmunoprecipitation was performed from cell extracts prepared 36 to 48 h after transient transfection of 293T cells by calcium phosphate precipitation. Cell extracts were incubated with either 3 μg of anti-HA monoclonal antibody, 3 μg of anti-Flag monoclonal antibody, or 2 μl of anti-GST polyclonal antibody for 3 to 4 h, followed by incubation with 20 μl of protein A/G beads (Oncogene Sciences) for 1 h. Beads were extensively washed in buffer (1% NP-40, 150 mM NaCl2, 50 mM Tris, 1 mM EDTA, protease inhibitors) before being boiled in sample buffer, and bound proteins were fractionated by SDS-PAGE and immunoblotted with the respective antibodies.

Western blot analysis.

Equal amounts of proteins in cell extracts or immunoprecipitated proteins were separated by SDS-PAGE (9% gel) and blotted to nitrocellulose. Filters were incubated with the appropriate antibodies, either anti-HA monoclonal antibody 12CA5 (BAbCo, Richmond, Calif.), anti-Flag monoclonal antibody M2 (BAbCo), or an anti-GST polyclonal antibody, followed by horseradish peroxidase-labeled anti-mouse or anti-rabbit immunoglobulin G (Southern Biotechnology Associates, Inc.). The proteins were detected by enhanced chemiluminescence (Amersham). Endogenous GCKR was identified by using rabbit polyclonal antiserum as previously described (50), while endogenous JNK1 was identified with anti-JNK1 antibody C-17 (Santa Cruz Biotechnology).


TANK synergistically activated SAPK with TRAF2, TRAF5, and TRAF6 but not with TRAF3.

TANK (or I-TRAF) was originally isolated in a yeast two-hybrid screen as TRAF2- and TRAF3-associated proteins (5, 46). It binds to all known TRAF proteins except TRAF4, which is mostly insoluble in lysis buffer (5, 8, 46). When increasing amounts of TANK were coexpressed with a constant amount of TRAF2, a biphasic pattern of NF-κB activation was obtained with an initial dose-dependent enhancement, while a further increase in TANK leads to inhibition of TRAF2-mediated NF-κB activation (5). To determine the effect of TANK in TRAF-mediated SAPK activation, pEBB HA-tagged plasmids carrying TRAF2, TRAF3, TRAF5, and TRAF6 were cotransfected with an expression plasmid for JNK1 at constant levels, with increasing amounts of TANK-expressing constructs, into 293T cells. To detect SAPK activity, immunoprecipitated JNK1 was subjected to an in vitro kinase assay, using c-Jun(1-79) as the substrate. To control for the effect of TANK expressed alone in SAPK activation, 0, 1, 3, and 10 μg of pEBG-TANK was transfected into 293T cells with an expression plasmid for JNK1 at constant levels. No activation of SAPK was observed with increasing amounts of TANK (Fig. (Fig.1A).1A). Furthermore, no increase in SAPK activation was detected when 3 μg of pEBB-TRAF3 was transfected with 1, 3, or 10 μg of pEBG-TANK into 293T cells (Fig. (Fig.1B),1B), whereas overexpression of either TRAF2, TRAF5, or TRAF6 led to strong activation (Fig. (Fig.1B1B and C). However, the dose-dependent synergistic activation of SAPK varied among individual TRAF proteins. Strong synergistic SAPK activation was observed with TRAF2 and TANK in a dose-dependent manner, with higher SAPK activities obtained when more TANK was transfected. The synergy between TANK and TRAF5 or TRAF6 in activating SAPK displayed a biphasic pattern, with an initial increase at low dosages of TANK followed by decreased activation with larger dosages of TANK (Fig. (Fig.1C).1C). These results indicate the differences in the signaling capacities of various TRAF molecules and their abilities to cooperate with TANK.

FIG. 1
Synergy between TANK and TRAF proteins in activating SAPK. (A) TANK alone has no effect on SAPK activity. Human kidney 293T cells were transiently transfected with 0, 1, 3, or 10 μg of pEBG-TANK. Each plate received 2 μg of HA-tagged JNK1 ...

The carboxyl-terminal portion of TANK strongly inhibited TRAF- and TANK-mediated SAPK activation.

We previously mapped a 21-aa fragment in the middle of TANK as the TRAF family member-interacting motif in TANK (TIMtk) and defined the portion amino-terminal to TIMtk as TANK-N(1-168) and the portion carboxyl-terminal to TIMtk as TANK-C(190-413) (5). Synergistic NF-κB activation was obtained by coexpression of TANK-N and TRAF2, whereas TANK-C strongly inhibited CD40- and TRAF2-mediated NF-κB activation. To understand the role of TRAF2 and TANK in SAPK activation, we determined the effects by in vitro kinase assays of TANK-C and the dominant-negative form of TRAF2, which lacks the amino-terminal zinc ring domain of TRAF2, TRAF2-C, in CD40-mediated SAPK activation after cotransfection into 293T cells. Neither TANK-C nor TRAF2-C alone could activate SAPK (Fig. (Fig.2,2, lanes 8 and 9). However, TANK-C was able to completely inhibit SAPK activation induced by CD40 plus CD40L, TRAF2, or TRAF2 plus TANK (lanes 3, 6, and 13). TRAF2-C could fully abrogate SAPK activation by CD40, but its blockage of SAPK activation induced by TRAF2 or TRAF2 plus TANK was not as complete as that of TANK-C (lanes 4, 7, and 12). When TANK-N was coexpressed in 293T cells with TRAF2, it enhanced TRAF2-mediated SAPK activation (data not shown). These results suggest that TANK-C is a potent inhibitor of both the NF-κB and SAPK pathways, mediated through TNFRs, and that TANK-N can synergistically activate TRAF-mediated NF-κB and SAPK activation.

FIG. 2
Inhibition of SAPK activation by TANK-C. TANK-C and TRAF2-C inhibit SAPK activation by CD40 plus CD40L, TRAF2, and TRAF2 plus TANK. Human kidney 293T cells were transfected with 3 μg of pBABE-CD40 plus 3 μg of pBABE-CD40L, 6 μg ...

TANK may self-associate through both inter- and intramolecular interactions.

When coexpressed in 293T cells, full-length Flag-tagged TANK and GST-tagged TANK were associated with each other, suggesting that TANK can exist as a dimer (Fig. (Fig.3,3, lane 1). To further understand the nature of this interaction and the activating and inhibiting properties of the amino- and carboxyl-terminal regions of TANK, we performed a series of coimmunoprecipitations, using a dual-epitope tag strategy between various domains of TANK to explore the binding interactions. Antibodies to one epitope tag were used to immunoprecipitate protein complexes, while antibodies to a second epitope tag were used to detect the presence of an interacting protein by immunoblotting. TANK-N was found to strongly associate with full-length TANK and itself (Fig. (Fig.3,3, lanes 2 and 3), while TANK-C interacted very weakly with full-length TANK but did not associate with itself (lanes 5 and 6). In addition, coimmunoprecipitation between TANK-N and TANK-C was observed (lane 4). Thus, the dimerization of TANK can be explained in one of two ways, either through an intermolecular association of amino-termini or through the amino terminus of one TANK polypeptide binding to the carboxyl terminus of another TANK polypeptide. Based on the strength of the coimmunoprecipitation of the TANK amino termini and the relative weakness of the interaction between the carboxy terminus of TANK and the full-length protein, a TANK dimer is predicted to be formed through association between the amino termini. Furthermore, the weaker association between TANK-N and TANK-C may be significant in the formation of an intramolecular interaction between the amino terminus of TANK folding back to interact with its own carboxyl terminus. Supporting this prediction, the dimerization of full-length TANK is weaker by coimmunoprecipitation than dimerization of TANK-N with itself, although TANK-N expression is severalfold lower, suggesting that the potential intramolecular interaction between TANK-N and TANK-C may be inhibitory to TANK dimerization through the amino termini. Full-length TANK did not interact with the proteins Fyn and Btk, demonstrating binding specificity (Fig. (Fig.3,3, lanes 7 and 8). Thus, TANK may form both intermolecular and intramolecular interactions through either the TANK-N and TANK-N association or the TANK-N and TANK-C association.

FIG. 3
Self-association and potential intramolecular interaction of TANK. TANK can associate with itself, with interactions observed between the amino termini and between the amino and carboxyl termini. Various plasmids expressing Flag-TANK, Flag-TANK-N, GST-TANK, ...

The dominant-negative form of either GCKR, MEKK1, or SEK1, but not that of Rac1 or cdc42, strongly inhibited TRAF2- and TANK-mediated SAPK activation.

The upstream links where the stimulatory signals enter the SAPK pathway are complex and often depend on the nature of the stimuli and the cell type involved. Stimulation from growth factors, cytokines, and environmental stress may employ different entry points that connect to SEK1 (MKK4), which is an immediate upstream activator of SAPK. MEKK1 has been shown to be a common upstream regulator of SEK1 activity, which may be responsible for transmitting SAPK activation signals from various small GTPases such as Ras, Rac, cdc42, and Rho. On the other hand, some stimuli may enter the SAPK kinase cascade directly at the link of SEK1, as is the case with anisomycin (19), or through multiple pathways which cannot be effectively blocked by the MEKK1 dominant-negative mutant, as is the case with UV and hyperosmolarity (43).

To determine whether MEKK1, SEK1, and various small GTPases are in the path of signal transmission of SAPK activation induced by TRAF2 plus TANK stimulation, kinase-inactive dominant-negative mutants of MEKK1 and SEK1 were coexpressed with TRAF2 and TANK in 293T cells, and the lysates were assayed for SAPK activation by an in vitro kinase assay. As indicated in Fig. Fig.4B,4B, the dominant-negative forms of MEKK1 and SEK1 strongly blocked activation of SAPK activity induced by TRAF2 plus TANK (lanes 3 and 4), as well as by CD40 and TRAF2 alone (data not shown). On the other hand, overexpression of the dominant-negative forms of cdc42 and Rac1 did not affect TRAF2-plus-TANK-mediated SAPK activation (lanes 5 and 6). This finding suggests that TRAF2 and TANK activation of SAPK is mediated by MEKK1 and SEK1 but not by cdc42 and Rac1.

FIG. 4
Inhibition of SAPK activation with dominant-negative forms of GCKR, MEKK1, and SEK1 (DN-GCKR, DN-MEKK1, and DN-SEK1). (A) Dominant-negative GCKR inhibits TRAF2-mediated SAPK activation. Human 293T cells were transiently transfected with 6 μg of ...

It has been previously reported that a kinase-inactive dominant-negative form of GCKR inhibits SAPK activation mediated by TNF-α and TRAF2 (50). To determine whether GCKR is downstream of SAPK activation by TANK synergy with TRAF2 or CD40, dominant-negative GCKR was coexpressed with TRAF2 plus TANK and CD40 plus CD40L in 293T cells and assayed by in vitro kinase activity. As a positive control, we showed that the dominant-negative form of GCKR inhibited activation of SAPK by TNF-α and TRAF2 (Fig. (Fig.4A,4A, lanes 3 to 6). Furthermore, we found that the same dominant-negative form of GCKR also strongly inhibited SAPK activation induced by TRAF2 plus TANK and CD40 plus CD40L (lanes 7 to 10). According to sequence homology, GCKR is a member of the MAP4K family, whereas MEKK1 belongs to the MAP3K family. To determine the functional relationship between GCKR and MEKK1, we found that the dominant-negative GCKR did not affect MEKK1-mediated SAPK activity, whereas the dominant-negative MEKK1 abolished GCKR-mediated SAPK activity, indicating that GCKR is upstream of MEKK1 (Fig. (Fig.44C).

CD40 stimulation of primary tonsil B cells activated endogenous GCKR.

To further confirm the role of GCKR in the CD40 signaling pathway in more physiological conditions, we examined the activation of GCKR by CD40 in primary human tonsil B cells. Lysates from CD40-activated tonsil B cells were immunoprecipitated with a polyclonal antiserum to GCKR, and kinase activity was assayed by an in vitro kinase assay, using MBP as the substrate. The kinase activity of GCKR was upregulated within 5 min of CD40 stimulation and progressively increased through the 15-min time course (Fig. (Fig.5A).5A). These kinetics correlated with the kinase activity of endogenous JNK1 (Fig. (Fig.5B).5B). Expression of GCKR was also examined in primary cells and cell lines. Migrating as a 95-kDa protein, endogenous GCKR was similarly found to be expressed in HS-Sultan human plasmacytoma cells, Ramos human B cells, and human tonsil B and T cells (Fig. (Fig.5C).5C). Thus, GCKR is a downstream target of CD40 and is inducible by CD40 in primary B cells.

FIG. 5
Activation of endogenous GCKR in human tonsil B cells. (A) GCKR activity is enhanced by stimulation of CD40 in tonsil B cells. Lysates from human tonsil B cells were immunoprecipitated with an antiserum to GCKR and assayed for the ability to phosphorylate ...

TRAF2 and TANK synergistically interacted with GCKR.

To understand the mechanism of TRAF2 and TANK synergy in GCKR-mediated SAPK activation, we tested the possible interactions among TRAF2, TANK, and GCKR in a series of coimmunoprecipitation assays. When coexpressed in 293T cells, Flag-tagged GCKR was found in the complex coprecipitated with HA-tagged TRAF2 (Fig. (Fig.6A,6A, lane 4). Interestingly, overexpression of TANK strongly enhanced the association between TRAF2 and GCKR (Fig. (Fig.6A,6A, lane 6). Similarly, a weak association between Gst-tagged TANK and Flag-tagged GCKR was obtained in coimmunoprecipitation assays (Fig. (Fig.6B,6B, lane 5), but its association becomes much stronger when TRAF2 is coexpressed (Fig. (Fig.6B,6B, lane 6). As shown previously, TRAF2 and TANK form a strong interaction (Figure (Figure6C,6C, lane 6). These results suggest that TRAF2 and TANK individually interact with GCKR but together synergistically interact with GCKR to form a TRAF2-TANK-GCKR complex.

FIG. 6
Synergistic interaction of TRAF2 and TANK with GCKR. Individual TRAF2 and TANK interactions with GCKR are enhanced by coexpression of TRAF2 and TANK. Human 293T cells were transfected with pEBB-HA-TRAF2, pEBB-GST-TANK, or pcDNA3-Flag-GCKR, either alone ...

GLK is another GCK family protein involved in TRAF- and TANK-mediated SAPK activation.

A member of the GCK protein family, GLK, possesses 61% amino acid identity to GCKR. To explore the role of GLK in CD40-mediated activation of SAPK, we determined the effect of the kinase-inactive dominant-negative form of GLK in CD40-induced SAPK activation. As indicated in Fig. Fig.7,7, dominant-negative GLK strongly inhibited SAPK activation induced by CD40 plus CD40L or TRAF2 plus TANK, as shown by an in vitro kinase assay after coexpression in 293T cells (lanes 5 to 8). Similar to observations for GCKR, GLK-induced SAPK activation could be inhibited by dominant-negative forms of MEKK1 and SEK1, whereas MEKK1-induced SAPK could not be inhibited by the dominant-negative form of GLK (lanes 2 to 4, 9, and 10).

FIG. 7
Involvement of GLK in TRAF- and TANK-mediated SAPK activation. GLK-induced SAPK activity is inhibited by dominant-negative MEKK1 and SEK1 (DN-MEKK1 and DN-SEK1), while dominant-negative GLK (DN-GLK) inhibits SAPK activation by CD40 and TRAF2 plus TANK ...


We previously reported that point mutations in the amino-terminal zinc ring finger domain of TRAF5 abolished its ability to mediate NF-κB activation but did not affect its ability to activate SAPK. In addition, we found that TRAF2A, an alternative spliced form of TRAF2 with a 7-aa insertion in the zinc ring finger domain of TRAF2, activates the SAPK pathway but not the NF-κB pathway (10). We therefore suggested that TRAF proteins are the last common molecules between the NF-κB and SAPK pathways and may branch these two distinct pathways through different TRAF complexes. TRAF proteins have been reported to associate with many intracellular proteins, including TRADD, RIP, IRAK, NIK, TANK, TRIP, Peg3/Pw1, c-IAP1, c-IAP2, and A20 (4, 5, 16, 17, 26, 28, 42, 44, 54). These proteins potentially form large TRAF-associated complexes important in TRAF-mediated cell signaling. Nevertheless, the roles and the mechanisms of these molecules in TRAF-mediated NF-κB and SAPK activation are not known.

The ability of TANK to modulate TRAF2-mediated NF-κB activation suggested that TANK could potentially function as either an activator or an inhibitor of NF-κB activation induced by TNFR family proteins. In this report, we showed that TANK synergistically activated SAPK with TRAF proteins. TANK-mediated synergistic SAPK activation varies with individual TRAF proteins. It strongly enhanced TRAF2-mediated SAPK activation in a dose-dependent manner but did not synergize with TRAF3 in SAPK activation. As with TRAF2-mediated NF-κB activation, the effects of TANK upon TRAF5- and TRAF6-mediated SAPK activation were biphasic, with enhancement by lower levels of TANK and decreased enhancement by higher levels. These results suggest that TANK might function as a regulatory molecule controlling the threshold of TRAF-mediated NF-κB and SAPK activation.

This report also describes a potential mechanism of TANK function during TRAF-mediated NF-κB and SAPK activation. We previously defined a 21-aa peptide in the middle of TANK as the TRAF family member-interacting motif in TANK (TIMtk), and showed that the TIMtk can compete with the TRAF binding motif in the CD40 cytoplasmic tail for binding to the TRAF-C domain of TRAF proteins (5). In the work reported here, we predict that TANK forms a homodimer through an interaction in the TANK-N region, as well as forming an intramolecular association through the interaction between TANK-N and TANK-C, which may inhibit TANK dimerization. All these studies are consistent with our previously proposed model, suggesting that TANK in its ground state is autoinhibited by an intramolecular interaction of its carboxyl and amino termini. The binding of TRAF2 to the TIMtk motif in the middle of TANK may change TANK’s conformation, favoring the intermolecular TANK-N and TANK-N interaction to form a TANK dimer. Because TRAF binds TANK through the TIMtk, dimerization of TANK would lead to aggregation of TRAF2, which may trigger a signal for NF-κB and SAPK activation. Consistent with this model, we found that overexpression of TANK-C strongly inhibited NF-κB and SAPK activation induced by CD40, TRAF2 or TRAF2 plus TANK, presumably by competing away TANK dimerization sites at the amino terminus of endogenous TANK. Conversely, TANK-N overexpression synergistically activated TRAF-mediated NF-κB and SAPK activation, possibly by titrating away the autoinhibitory carboxyl-terminal portion of endogenous TANK.

Multiple pathways may be involved in SAPK activation. The SAPK activation initiated from many stress stimuli requires activation of a signaling cascade from the Rho group of small G proteins (such as Rac1 or cdc42) to mixed lineage kinases (such as MLK-2 and MLK-3), to MEKK1, to SEK1 or SEK2, to JNK family proteins, to c-Jun (3, 9, 11, 27, 30, 31, 37, 47, 57). Other stress inducers may activate JNK through a family of germinal center kinases (25). Recent studies suggested two potential signal transduction pathways for TRAF-induced SAPK activation (35, 50, 59). One is mediated by ASK1 (or MAPKKK5), which is a serine/threonine kinase at the level of the MAPK kinase kinases (35). Studies showed that ASK1 coprecipitates with TRAF2 upon TNF-α stimulation. The dominant-negative ASK1, which contains a lysine-to-alanine mutation at the ATP binding site, however, only partially inhibits TNF-α- or TRAF-induced SAPK activation (8, 35). Other studies suggest the involvement of GCK and GCKR in TRAF-induced SAPK activation (50, 59). In this report, we showed that although TRAF2 or TANK alone bound to GCKR weakly, together they formed a strong complex with GCKR. This synergistic binding of TRAF2 and TANK to GCKR may result in synergistic activation of SAPK by TRAF2 and TANK. Consistent with the role of GCKR in TRAF2- and TANK-mediated SAPK activation, we showed that the dominant-negative GCKR but not the dominant-negative Rac1 or cdc42 strongly inhibited SAPK activation induced by TRAF2 plus TANK. The GCK family proteins are kinases at the MAP4K level. Although their physiological targeting molecules in the kinase cascade are not yet confirmed, activation of some GCK family proteins is likely to lead to activation of MEKK1. We showed that the dominant-negative MEKK1 strongly inhibited SAPK activation induced by GCKR or GLK, whereas the dominant-negative GCKR or GLK did not affect MEKK-1-induced SAPK activation. Furthermore, coprecipitation between GCK and MEKK1 was recently reported (59). The dominant-negative SEK1 abolished SAPK activation induced by upstream activators such as CD40, TRAF2, and TRAF2 plus TANK, suggesting that SEK1 may be a potential kinase involved in the TRAF-mediated SAPK activation pathway (8). In addition to SAPK activation, TRAF proteins have also been reported to activate the p38 pathway, a stress-activated pathway parallel to the SAPK pathway (2). Further studies are necessary to determine the contributions of ASK1 and GCK family proteins in TRAF- and TANK-mediated SAPK and p38 activation.

We also reported the endogenous activation of GCKR by CD40 in primary B cells. Dominant-negative TRAF2 and GCKR completely abrogated CD40-mediated SAPK activation in vivo. Together, these data suggest that GCKR is a critical mediator in CD40 activation of SAPK. Interestingly, the kinase known as GCK derived its name by virtue of its high expression in the follicular germinal center, with putative roles in B-cell differentiation and selection (22, 39). In B cells, CD40 has been implicated in a variety of functions, including germinal center formation, immunoglobulin isotype switching, antibody affinity maturation, and generation and maintenance of memory B cells (14, 23). However, how GCKR-mediated SAPK activation contributes to CD40-mediated B-cell functions is a critical and open question which should be addressed further. In addition to GCKR, our and other studies suggest that GLK, another GCK family protein, and GCK may also be involved in TNFR-mediated and in TRAF- and TANK-mediated SAPK activation. With 11 known members of the GCK family, it will be of interest to determine whether additional members are involved in signal transduction mediated by CD40 as well as other TNFRs. Specificity of GCK members to distinct TNFRs in vivo will be important in delineating the function of GCK family-mediated signaling.


A. Chin, J. Shu, and C. Shi contributed equally to this work.

This work was supported by Boehringer Ingelheim Pharmaceuticals Inc. and National Institute of General Medical Sciences grant GM57559. A. Chin is a recipient of the UCLA Medical Science Training Program training grant GM 08042, and J. Shu is a postdoctoral fellow supported by Boehringer Ingelheim Pharmaceuticals Inc.

We thank R. Davis, M. Karin, and Z. Luo for various constructs, M. Kehry for soluble CD8-gp39, and T. Parks and K. Catron for helpful discussions.


1. Anderson D, Maraskovsky E, Billingsley W, Dougall W, Tometsko M, Roux E, Teepe M, DuBose R, Cosman D, Galibert L. A homologue of the TNF receptor and its ligand enhance T-cell growth and dendritic-cell function. Nature. 1997;390:175–179. [PubMed]
2. Arch R H, Gehrich R W, Thompson C B. Tumor necrosis factor receptor-associated factors (TRAFs)—a family of adaptor proteins that regulates life and death. Genes Dev. 1998;12:2821–2830. [PubMed]
3. Bagrodia S, Derijard B, Davis R J, Cerione R A. Cdc42 and PAK-mediated signaling leads to Jun kinase and p38 mitogen-activated protein kinase activation. J Biol Chem. 1995;270:27995–27998. [PubMed]
4. Cao Z, Xiong J, Takeuchi M, Kurama T, Goeddel D V. TRAF6 is a signal transducer for interleukin-1. Nature. 1996;383:443–446. [PubMed]
5. Cheng G, Baltimore D. TANK, a co-inducer with TRAF2 of TNF- and CD40L-mediated NF-κB activation. Genes Dev. 1996;10:963–973. [PubMed]
6. Cheng G, Cleary A M, Ye Z, Hong D I, Lederman S, Baltimore D. Involvement of CRAF1, a relative of TRAF, in CD40 signaling. Science. 1995;267:1494–1498. [PubMed]
7. Cheng G, Ye Z, Baltimore D. Binding of Bruton’s tyrosine kinase to Fyn, Lyn, or Hck through a Src homology 3 domain-mediated interaction. Proc Natl Acad Sci USA. 1994;91:8152–8155. [PubMed]
8. Chin, A., J. Shu, and G. Cheng. Unpublished data.
9. Coso O A, Chiariello M, Yu J C, Teramoto H, Crespo P, Xu N, Miki T, Gutkind J S. The small GTP-binding proteins Rac1 and cdc42 regulate the activity of the JNK/SAPK signaling pathway. Cell. 1995;81:1137–1146. [PubMed]
10. Dadgostar H, Cheng G. An intact zinc ring finger is required for tumor necrosis factor receptor-associated factor-mediated nuclear factor-kappa B activation but is dispensable for c-Jun N-terminal kinase signaling. J Biol Chem. 1998;273:24775–24780. [PubMed]
11. Derijard B, Raingeaud J, Barrett T, Wu I H, Han J, Ulevitch R J, Davis R J. Independent human MAP kinase signal transduction pathways defined by MEK and MKK isoforms. Science. 1995;267:682–685. [PubMed]
12. DiDonato J A, Hayakawa M, Rothwarf D M, Zandi E, Karin M. A cytokine-responsive IκB kinase that activates the transcription factor NF-κB. Nature. 1997;388:548–554. [PubMed]
13. Diener K, Wang X S, Chen C, Meyer C F, Keesler G, Zukowski M, Tan T-H, Yao Z. Activation of the c-Jun N-terminal kinase pathway by a novel protein kinase related to human germinal center kinase. Proc Natl Acad Sci USA. 1997;94:9687–9692. [PubMed]
14. Foy T M, Laman J D, Ledbetter J A, Aruffo A, Claassen E, Noelle R J. gp-39-CD40 interactions are essential for germinal center formation and the development of B cell memory. J Exp Med. 1994;180:157–163. [PMC free article] [PubMed]
15. Hibi M, Lin T, Smeal T, Minden A, Karin M. Identification of an oncoprotein- and UV-responsive protein kinase that binds and potentiates the c-Jun activation domain. Genes Dev. 1993;7:2135–2148. [PubMed]
16. Hsu H, Huang J, Shu H, Baichwal V, Goeddel D. TNF-dependent recruitment of the protein kinase RIP to the TNF receptor-1 signaling complex. Immunity. 1996;4:387–396. [PubMed]
17. Hsu H, Shu H, Pan M, Goeddel D. TRADD-TRAF2 and TRADD-FADD interactions define two distinct TNF receptor 1 signal transduction pathways. Cell. 1996;94:299–308. [PubMed]
18. Hu H M, O’Rourke K, Boguski M S, Dixit V M. A novel RING finger protein interacts with the cytoplasmic domain of CD40. J Biol Chem. 1994;269:30069–30072. [PubMed]
19. Iordanov M S, Pribnow D, Magun J L, Dinh T-H, Pearson J A, Chen S L-Y, Magun B E. Ribotoxic stress response: activation of the stress-activated protein kinase JNK1 by inhibitors of the peptidyl transferase reaction and by sequence-specific RNA damage to the α-sarcin/ricin loop in the 28S rRNA. Mol Cell Biol. 1997;17:3373–3381. [PMC free article] [PubMed]
20. Ishida T, Tojo T, Aoki T, Kobayashi N, Ohishi T, Watanabe T, Yamamoto T, Inoue J. TRAF5, a novel tumor necrosis factor receptor-associated factor family protein, mediates CD40 signaling. Proc Natl Acad Sci USA. 1996;93:9437–9442. [PubMed]
21. Itoh N, Nagata S. A novel protein domain required for apoptosis. J Biol Chem. 1993;268:10932–10937. [PubMed]
22. Katz P, Whalen G, Kehrl J H. Differential expression of a novel protein kinase in human B lymphocytes. Preferential localization in the germinal center. J Biol Chem. 1994;269:16802–16809. [PubMed]
23. Kawabe T, Naka T, Yoshida K, Tanaka T, Fujiwara H, Suematsu S, Yoshida N, Kishimoto T, Kikutani H. The immune responses in CD40-deficient mice: impaired immunoglobulin class switching and germinal center formation. Immunity. 1994;1:167–178. [PubMed]
24. Kwon B, Tan K, Ni J, Oh K, Lee Z, Kim K, Kim Y, Wang S, Gentz R, Yu G, Harrop J, Lyn S, Silverman C, Porter T, Truneh A, Young P. A newly identified member of the tumor necrosis factor receptor superfamily with a wide tissue distribution and involvement in lymphocyte activation. J Biol Chem. 1997;272:14272–14276. [PubMed]
25. Kyriakis J M. Signaling by the germinal center kinase family of protein kinases. J Biol Chem. 1999;274:5259–5262. [PubMed]
26. Lee S, Choi Y. TRAF-interacting protein (TRIP): a novel component of the tumor necrosis factor receptor (TNFR)- and CD30-TRAF signaling complexes that inhibits TRAF2-mediated NF-kappaB activation. J Exp Med. 1997;185:1275–1285. [PMC free article] [PubMed]
27. Lin A, Minden A, Martinetto H, Claret F X, Lange-Carter C, Mercurio F, Johnson G L, Karin M. Identification of a dual specificity kinase that activates the Jun kinases and p38-Mpk2. Science. 1995;268:286–290. [PubMed]
28. Malinin N L, Boldin M P, Kovalenko A V, Wallach D. MAP3K-related kinase involved in NF-κB induction by TNF, CD95 and IL-1. Nature. 1997;385:540–544. [PubMed]
29. Mercurio F, Zhu H, Murray B, Shevchenko A, Bennett B, Li J, Young D, Barbosa M, Mann M, Manning A, Rao A. IKK-1 and IKK-2: cytokine-activated IkappaB kinases essential for NF-kappaB activation. Science. 1997;278:860–866. [PubMed]
30. Minden A, Lin A, Claret F X, Abo A, Karin M. Selective activation of the JNK signaling cascade and c-Jun transcriptional activity by the small GTPases Rac and cdc42HS. Cell. 1995;81:1147–1157. [PubMed]
31. Minden A, Lin A, McMahon M, Lange-Carter C, Derijard B, Davis R J, Johnson G L, Karin M. Differential activation of ERK and JNK mitogen-activated protein kinases by Raf-1 and MEKK. Science. 1994;266:1719–1723. [PubMed]
32. Mosialos G, Birkenbach M, Yalamanchili R, VanArsdale T, Ware C, Kieff E. The Epstein-Barr virus transforming protein LMP1 engages signaling proteins for the tumor necrosis factor receptor family. Cell. 1995;80:389–399. [PubMed]
33. Nakano H, Oshima H, Chung W, Williams-Abbott L, Ware C, Yagita H, Okumura K. TRAF5, an activator of NF-kappaB and putative signal transducer for the lymphotoxin-beta receptor. J Biol Chem. 1996;271:14661–14664. [PubMed]
34. Natoli G, Costanzo A, Ianni A, Templeton D, Woodgett J, Balsano C, Levrero M. Activation of SAPK/JNK by TNF receptor 1 through a noncytotoxic TRAF2-dependent pathway. Science. 1997;275:200–203. [PubMed]
35. Nishitoh H, Saitoh M, Mochida Y, Takeda K, Nakano H, Rothe M, Miyazono K, Ichijo H. ASK1 is essential for JNK/SAPK activation by TRAF2. Mol Cell. 1998;3:389–395. [PubMed]
36. Nocentini G, Giunchi L, Ronchetti S, Krausz L, Bartoli A, Moraca R, Migliorati G, Riccardi C. A new member of the tumor necrosis factor/nerve growth factor receptor family inhibits T cell receptor-induced apoptosis. Proc Natl Acad Sci USA. 1997;94:6216–6221. [PubMed]
37. Olson M F, Ashworth A, Hall A. An essential role for Rho, Rac, and Cdc42 GTPases in cell cycle progression through G1. Science. 1995;269:1270–1272. [PubMed]
38. Pan G, Ni J, Wei Y, Yu G, Gentz R, Dixit V. An antagonist decoy receptor and a death domain-containing receptor for TRAIL. Science. 1997;277:815–818. [PubMed]
39. Pombo C, Kehrl J, Sanchez I, Katz P, Avuch J, Zon L, Woodgett J, Force T, Kyriakis J. Activation of the SAPK pathway by the human STE20 homologue germinal centre kinase. Nature. 1995;377:750–754. [PubMed]
40. Regnier C, Tomasetto C, Moog-Lutz C, Chenard M, Wendling C, Basset P, Rio M. Presence of a new conserved domain in CART1, a novel member of the tumor necrosis factor receptor-associated protein family, which is expressed in breast carcinoma. J Biol Chem. 1995;270:25715–25712. [PubMed]
41. Regnier C H, Song H Y, Gao X, Goeddel D V, Cao Z, Rothe M. Identification and characterization of an IκB kinase. Cell. 1997;90:373–383. [PubMed]
42. Relaix F, Wei X, Wu X, Sassoon D. Peg3/Pw1 is an imprinted gene involved in the TNF-NFkappaB signal transduction pathway. Nat Genet. 1998;18:287–291. [PubMed]
43. Rosette C, Karin M. Ultraviolet light and osmotic stress: activation of the JNK cascade through multiple growth factor and cytokine receptors. Science. 1996;274:1194–1197. [PubMed]
44. Rothe M, Pan M, Henzel W, Ayres T, Goeddel D V. The TNFR2-TRAF signaling complex contains two novel proteins related tobaculoviral inhibitor of apoptosis proteins. Cell. 1995;7:1243–1252. [PubMed]
45. Rothe M, Wong S C, Henzel W J, Goeddel D V. A novel family of putative signal transducers associated with the cytoplasmic domain of the 75 kDa tumor necrosis factor receptor. Cell. 1994;78:681–692. [PubMed]
46. Rothe M, Xiong J, Shu H, Williamson K, Goddard A, Goeddel D. I-TRAF is a novel TRAF-interacting protein that regulates TRAF-mediated signal transduction. Proc Natl Acad Sci USA. 1996;93:8241–8246. [PubMed]
47. Sanchez I, Hughes R T, Mayer B J, Yee K, Woodgett J R, Avruch J, Kyriakis J M, Zon L I. Role of SAPK/ERK kinase-1 in the stress-activated pathway regulating transcription factor c-Jun. Nature. 1994;372:794–798. [PubMed]
48. Sato T, Irie S, Reed J C. A novel member of the TRAF family of putative signal transducing proteins binds to the cytosolic domain of CD40. FEBS Lett. 1995;358:113–118. [PubMed]
49. Sheridan J, Marsters S, Pitti R, Gurney A, Skubatch M, Baldwin D, Ramakrishnan L, Gray C, Baker K, Wood W, Goddard A, Godowski P, Ashkenazi A. Control of TRAIL-induced apoptosis by a family of signaling and decoy receptors. Science. 1997;277:818–821. [PubMed]
50. Shi C, Kehrl J. Activation of stress-activated protein kinase/c-Jun N-terminal kinase, but not NF-kappaB, by the tumor necrosis factor (TNF) receptor 1 through a TNF receptor-associated factor 2- and germinal center kinase related-dependent pathway. J Biol Chem. 1997;272:32102–32107. [PubMed]
51. Simonet W, Lacey D, Dunstan C, Kelley M, Chang M, Luthy R, Nguyen H, Wooden S, Bennett L, Boone T, Shimamoto G, DeRose M, Elliott R, Colombero A, Tan H, Trail G, Sullivan J, Davy E, Bucay N, Renshaw-Gegg L, Hughes T, Hill D, Pattison W, Campbell P, Boyle W. Osteoprotegerin: a novel secreted protein involved in the regulation of bone density. Cell. 1997;89:309–319. [PubMed]
52. Smith C A, Farrah T, Goodwin R G. The TNF receptor superfamily of cellular and viral proteins: activation, costimulation, and death. Cell. 1994;76:959–962. [PubMed]
53. Song H, Regnier C, Kirschning C, Goeddel D V, Rothe M. Tumor necrosis factor (TNF)-mediated kinase cascades: bifurcation of nuclear factor-kappaB and c-jun N-terminal kinase (JNK/SAPK) pathways at TNF receptor-associated factor 2. Proc Natl Acad Sci USA. 1997;94:9792–9796. [PubMed]
54. Song H, Rothe M, Goeddel D V. The tumor necrosis factor-inducible zinc finger protein A20 interacts with TRAF1/TRAF2 and inhibits NF-kappaB activation. Proc Natl Acad Sci USA. 1996;93:6721–6725. [PubMed]
55. Tartagilia L A, Ayres T M, Wong G H W, Goeddel D V. A novel domain within the 55 kd TNF receptor signals cell death. Cell. 1993;74:845–853. [PubMed]
56. Woronicz J, Gao X, Cao Z, Rothe M, Goeddel D V. IkappaB kinase-beta: NF-kappaB activation and complex formation with IkappaB kinase-alpha and NIK. Science. 1997;278:866–869. [PubMed]
57. Yan M, Dai T, Deak J C, Kyriakis J M, Zon L I, Woodgett J R, Templeton D J. Activation of stress-activated protein kinase by MEKK1 phosphorylation of its activator SEK1. Nature. 1994;372:798–800. [PubMed]
58. Yang R, Mark M, Gray A, Huang A, Xie M, Zhang M, Goddard A, Wood W, Gurney A, Godowski P. Toll-like receptor-2 mediates lipopolysaccharide-induced cellular signalling. Nature. 1998;395:284–288. [PubMed]
59. Yuasa T, Ohno S, Kehrl J, Kyriakis J. Tumor necrosis factor signaling to stress-activated protein kinase (SAPK)/Jun NH2-terminal kinase (JNK) and p38. Germinal center kinase couples TRAF2 to mitogen-activated protein kinase/ERK kinase kinase 1 and SAPK while receptor interacting protein associates with a mitogen-activated protein kinase kinase kinase upstream of MKK6 and p38. J Biol Chem. 1998;273:22681–22692. [PubMed]
60. Zandi E, Rothwarf D, Delhase M, Hayakawa M, Karin M. The IkappaB kinase complex (IKK) contains two kinase subunits, IKKalpha and IKKbeta, necessary for IkappaB phosphorylation and NF-kappaB activation. Cell. 1997;91:243–252. [PubMed]

Articles from Molecular and Cellular Biology are provided here courtesy of American Society for Microbiology (ASM)