Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M703422200 on June 21, 2007

J. Biol. Chem., Vol. 282, Issue 33, 24246-24254, August 17, 2007
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
282/33/24246    most recent
M703422200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhong, J.
Right arrow Articles by Kyriakis, J. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhong, J.
Right arrow Articles by Kyriakis, J. M.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Dissection of a Signaling Pathway by Which Pathogen-associated Molecular Patterns Recruit the JNK and p38 MAPKs and Trigger Cytokine Release*

Jian Zhong and John M. Kyriakis1

From the Molecular Cardiology Research Institute, Department of Medicine, Tufts-New England Medical Center and the Department of Medicine, Tufts University School of Medicine, Boston, Massachusetts 02111

Received for publication, April 24, 2007 , and in revised form, June 13, 2007.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Pathogen-associated molecular patterns (PAMPs), molecular moieties produced by invading microbial pathogens, initiate innate immune responses by binding to pattern recognition receptors (PRRs). Engagement of PRRs elicits a wide variety of responses, including the production and release of cytokines and chemokines. These responses require the activation of several parallel signaling pathways, including NF-{kappa}B, the interferon regulatory factors, and the MAPKs. The JNK and p38 MAPK groups are major PRR effectors and are key to the PRR-dependent induction and release of proinflammatory cytokines such as tumor necrosis factor and interleukin-8. The mammalian Ste20 orthologue germinal center kinase (GCK) is required for the activation of JNK by a subset of PAMPs; however, the mechanisms by which GCK couples to downstream events remain unclear. Here we show that GCK is required for JNK and, unexpectedly, p38 activation by three bacterial PAMPs, lipopolysaccharide, peptidoglycan, and flagellin (FliC). We show that these same PAMPs, in a GCK-dependent manner, activate mixed lineage kinases-2 and -3, MAPK kinase kinases upstream of JNK, and p38. We also show that MLK2 and -3 are required for activation of JNK and p38 by ectopically expressed GCK. Finally, we show that MLK2 and -3 are required for lipopolysaccharide, peptidoglycan, and FliC recruitment of JNK and p38 as well as for PAMP recruitment of the transcription factor c-Jun, and for the induction of interleukin-8. Our results define a signaling pathway whereby PAMPs can trigger MAPK activation and gene expression.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Multicellular organisms must constantly maintain a vigilant defense against an environment rich in pathogenic microorganisms. These defense mechanisms are remarkably conserved among multicellular organisms and are referred to as innate immunity (1-6). In contrast to acquired immunity, which is based on the generation, by complex somatic gene rearrangement, of a large number of receptors, innate immunity relies on a small number of pattern recognition receptors (PRRs)2 that are responsive to pathogen-associated molecular patterns (PAMPs), conserved molecular moieties produced by invading bacteria, viruses, fungi, and parasites (1-6). PRRs can also be activated by some host cell proteins (e.g. oxidized low density lipoprotein in atherogenesis) and by the metabolic by-products of dying cells (e.g. deposition of crystalline sodium urate, as occurs in gout) (4, 7). Recruitment of PRRs elicits a dramatic and often self-reinforcing proinflammatory process that includes cytokine and chemokine release, fever, and, in extreme systemic cases, shock, disseminated intravascular coagulation, and death. Triggering of the innate immune response also elicits the activation of the acquired immune system (1-6).

Two broad classes of PRRs have been identified as follows: the transmembrane toll-like receptors (TLRs) and the cytosolic PRRs, which include the nucleotide-binding oligomerization domain/leucine-rich repeat proteins (NOD-LRRs) and the retinoic acid-inducible gene-like helicases (RLHs) (1, 4). LPS produced by Gram-negative bacteria recruits TLR4. TLR2 is activated by bacterial peptidoglycans (PGNs), whereas TLR5 responds to bacterial flagellin. The NOD-LRRs and RLHs are thought to recognize PAMPs produced by microbial pathogens that invade the cytosol (1, 4). Nucleotide binding/oligomerization domain-2 (NOD2), mutations that have been linked to Crohn disease, is a NOD-LRR protein that recognizes cytosolic PGN and, along with TLR2, contributes significantly to the innate immune response to PGNs. Naip5 (neuronal inhibitor of apoptosis-5) is a second NOD-LRR protein that is likely a cytosolic flagellin receptor. A subset of the RLHs recognizes viral DNA and RNA (1, 4, 5).

Neither the PRRs, the NOD-LRRs, nor the RLHs possess ligand-induced enzymatic activity that is relevant to signaling. Instead, these receptors signal via the signal-dependent recruitment of additional cytosolic polypeptides. The intracellular extensions of TLRs and the IL-1 receptor contain a TLR/IL-1 receptor (TIR) motif; and upon receptor occupancy, these TIR motifs bind to TIR motifs on a suite of cytosolic adapter proteins that relay the signals to downstream components. For example, myeloid differentiation factor-88 (MyD88) and TIR-associated protein (TIRAP) are two such adapters that are necessary for optimal mitogen-activated protein kinase (MAPK) and nuclear factor-{kappa}B (NF-{kappa}B) activation by TLR4 (1-6). MyD88, at the receptor complex, binds to proteins of the IL-1 receptor-associated kinase (IRAK) family. IRAKs, in turn, bind to another adapter, TNF receptor-associated factor-6 (TRAF6). TRAF6 also recruits proximal elements in MAPK and NF-{kappa}B pathways. Cytosolic PRR signaling is much more poorly understood; however, adapters of the receptor-interacting protein, IRAK, and TRAF family have been linked to these receptors (1-6).

The MAPKs represent key signaling intermediates in the cellular responses to PAMPs. Of particular note, the extracellular signal-regulated kinase (ERK), c-Jun NH2-terminal kinase (JNK), and p38 MAPK groups collaborate to recruit the activator protein-1 (AP-1) transcription factor (1, 8). AP-1 is a heterodimer that consists of c-Jun coupled with either c-Fos, other members of the Jun family (JunB and D), or members of the activating transcription factor (ATF) family (9). AP-1 activation is critical to PAMP-induced cytokine and chemokine gene expression (TNF, IL-1, IL-8, IL-12, etc.), as well as the expression of cyclooxygenase-II (COX-II), and the up-regulation of cell adhesion molecules (E-selectin and vascular cell adhesion molecule-1 for example) (9-12). MAPKs act not only at the transcriptional level but, especially in the case of p38s, at the post-transcriptional level where they function in the signal-induced stabilization of cytokine and chemokine mRNAs that contain A/U-rich elements (13).

Gene disruption studies indicate that TRAF6 is key to the recruitment by PAMPs of the MAPKs (14). However, a clear picture of how TRAF6, or other TRAFs for that matter, couples to MAPKs is still elusive. MAPKs are activated by MAP3K -> MAPK kinase (MKK) -> MAPK core pathways. It is clear that MKK-3 and -6, upstream of p38, and MKK-4 and -7, upstream of the JNKs, as well as the ERK-specific MKKs MEK1 and -2 are key to PAMP signaling (1, 8). Proximal to these MKKs, several MAP3Ks have been variously linked to MAPK signaling. Several studies indicate that PAMP (lipopolysaccharide (LPS)) activation of ERK requires the MAP3K tumor progression locus-2 (Tpl2) (15). Some studies have linked transforming growth factor-beta-activated kinase-1 (TAK1) and MAPK kinase kinase-3 (MEKK3), MAP3Ks upstream of the JNKs and p38s, to PAMP signaling, at least in fibroblasts and lymphocytes (16-18). A role for TAK1 in the responses of critical innate immune cells such as macrophages, dendritic cells, and lymphocytes to PAMPs has not been established. MAP3Ks of the mixed lineage kinase (MLK) family, especially MLK3, are potent activators of MKK-4 and -7 upstream of JNK and MKK6 upstream of p38. MLK3 has been linked to cytokine recruitment of JNK and p38 (8, 19), but a role for the MLKs in PAMP signaling has also not yet been demonstrated. Moreover, the mechanisms by which these MAP3Ks are activated are still unknown.

Germinal center kinase (GCK) is the founding member of a group of protein kinases, the GCKs, whose amino-terminal kinase domains are distantly homologous to that of Saccharomyces cerevisiae STE20 (20, 21). The carboxyl-terminal regulatory regions of GCK are highly divergent; and genetic and biochemical studies implicate the GCKs in a broad range of functions, including inflammation, cell proliferation, and apoptosis (20-24). Experiments using transfection and overexpression indicated that GCK could selectively activate the JNKs, but not the p38s, ERKs, or NF-{kappa}B (25).

Our early RNAi studies indicated that although GCK was activated by PAMPs (LPS, PGN, and poly(I-C), a PAMP that mimics viral RNA), as well as by TNF, IL-1, and CD40, it was apparently only required for JNK activation by LPS and PGN (22). Recruitment of other MAPKs was not examined in detail (22).

GCK is constitutively active catalytically, and the kinase activity is required for GCK ubiquitination and proteasome-dependent proteolytic degradation. Regulation of GCK involves a transient inhibition of the degradation of ubiquitinated GCK. This process is mediated by the stimulus-dependent binding of GCK to TRAF6 and by the binding of GCK to the MAP3K MEKK1 (22). These events sequester GCK from the proteasome and enable accumulation of GCK polypeptide and consequent GCK-dependent signaling. Thus, although GCK can activate MEKK1 in vitro (26), MEKK1 is likely to function more as a stabilizer of GCK in situ (22). GCK can also activate recombinant MLK3 in vitro (26), but the physiologic significance of MLK3 to GCK signaling has, until now, remained unclear.

Accordingly, we sought to dissect further the signaling pathways by which PAMPs, through GCK, recruit MAPK pathways. Our findings reveal that GCK is activated by at least three PAMPs, LPS, PGN, and bacterial flagellin (FliC; in this paper FliC was derived from Shiga-toxigenic Escherichia coli), and that GCK signals through MLK2 and -3 to recruit the JNKs p38 and their effectors. Our results point to an important role for GCK and the MLKs in PAMP-stimulated MAPK pathway activation.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Cells and Treatments—Jurkat human T lymphoma cells were cultured in RPMI, 10% fetal calf serum as described (22). 293 cells were cultured in DMEM, 10% fetal calf serum. LPS (Calbiochem, highly purified from E. coli serotype 0111:B4) and PGN (Staphylococcus aureus) were each used at 1 µg/ml for the times indicated in the figures. Recombinant FliC (His-tagged, from O113:H21 Shiga-toxigenic E. coli strain 98NK2, purified on Ni2+-nitrilotriacetic acid-agarose) and heat-inactivated FliC (50 °C, 4 h, negative control) were kindly provided by Dr. Cheleste Thorpe (Tufts-New England Medical Center) and were used at 100 ng/ml each for the times indicated in the figures.

Plasmids, Antibodies, and Assays—Full-length FLAG-GCK as well as deletion constructs of FLAG-GCK have been described (26, 27). GST-tagged mammalian expression constructs of full-length and truncated MLK3 were prepared by PCR cloning into the pEBG vector. Constructs were verified by DNA sequencing. Transfection of 293 cells was as described (22, 26, 27). Phospho-specific and conventional antibodies for JNK, p38, and ERK were from Cell Signaling Technologies. Anti-GCK, MLK2, and MLK3 antibodies were from Santa Cruz Biotechnology. Anti-FLAG was from Sigma. Immunoprecipitation and immunoblotting were as described (22, 26, 27). Immunoblots were quantified with the Image J software package (Macintosh platform). Protein kinase assays for MLK3, JNK, and p38 have been described (19, 27); and the assays were quantitated by phosphorimaging.


Figure 1
View larger version (74K):
[in this window]
[in a new window]

  		FIGURE 1.
GCK is required for LPS, PGN, and FliC activation of JNK and p38 but not ERK. Jurkat cells were treated with the primary (A, C, and D) or alternative (B) GCK siRNAs ("Experimental Procedures"). After 16 h, cells were treated with LPS (A and B), PGN (C), or either FliC or heat-inactivated FliC (D) for the indicated times. Cell extracts were prepared and subjected to SDS-PAGE and immunoblotting with the indicated antibodies. Phospho-specific antibody immunoblots were quantitated with ImageJ as described ("Experimental Procedures").

 
RNA Interference—siRNA-dependent RNAi was performed on Jurkat cells as described (22). The primary GCK sense siRNA used was 5'-GGUGCAUAUGGGCGCCUGC-3'. The alternative sense GCK siRNA was 5'-CCCCUACACGGGUGCCACC-3'. The primary human sense MLK3 siRNA was 5'-GCGCGAGAUCCAGGGUCUC-3'. The alternative sense human MLK3 siRNA oligonucleotide was 5'-GCUGGAGAUUCAGCACAUG-3'. The primary human sense MLK2 siRNA oligonucleotide was 5'-GCUGGAGAUUCAGCACAUG-3'. The alternative sense MLK2 oligonucleotide was 5'-AAGCAGUGAUGUCUGGAGC-3'. All oligonucleotides were constructed with 3'-dTdT overhangs. In all instances, complementary strands (not shown), also with 3'-dTdT overhangs, were routinely synthesized for conventional siRNA-based RNAi approaches. Alternative siRNA sequences were used where indicated in the figures/legends (for LPS-treated cells). Otherwise, the primary sequences were used. For negative controls, a nonspecific control siRNA (for green fluorescent protein, Dharmacon, Inc.) was employed.

Reverse Transcriptase PCR—Reverse transcriptase PCR was performed as described (22). The human IL-8 primers were forward, ATGACTTCCAAGCTGGCCGTGGCT, and reverse, TCTCAGCCCTCTTCAAAAACTTCTC. The beta-actin control primers were forward, GTGGGGCGCCCCAGGCACCA, and reverse, CTCCTTAATGTCACGCACGATTTC.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
LPS, PGN, and FliC Recruitment of JNK and p38 but Not ERK Requires GCK—We have shown previously that the activation of JNK by LPS and PGN requires GCK (22). We wished to determine whether the activation of other MAPKs required GCK. Accordingly, Jurkat human T lymphoma cells were subjected to GCK RNAi, and the cells were treated with LPS (Fig. 1, A and B), PGN (Fig. 1C), or FliC (Fig. 1D) for the times indicated. MAPK activation was monitored by immunoblotting with phospho-specific antibodies. As a control for FliC signaling, we exploited a heat-inactivated preparation of FliC protein. As is evident from Fig. 1, all three PAMPs trigger the characteristic (22) time-dependent stabilization of GCK, which is typically maximal within 3 min. This stabilization is closely followed by activation of all three MAPKs, with PGN and FliC producing a particularly robust response. Activation of the MAPKs is maximal at between 10 and 30 min. Heat-inactivated FliC produces none of these responses. Silencing of gck substantially impairs JNK and p38 but not ERK activation by all three agonists. Thus, GCK is important for both JNK and p38 activation by these three PAMPs but is dispensable for ERK activation. A second gck-specific RNAi produces identical results (Fig. 1, A versus B), ruling out off-target effects.


Figure 2
View larger version (51K):
[in this window]
[in a new window]

  		FIGURE 2.
Activation of MLK2 and -3 by PAMPs requires GCK. Jurkat cells were subjected to gck RNAi using the primary GCK siRNA ("Experimental Procedures"). After 16 h, cells were treated with LPS (A and C) or FliC (B) for various times, as indicated. Extracts were prepared, and either MLK3 (A and B) or MLK2 (C) was immunoprecipitated from a portion of the extract and assayed in vitro using recombinant biotinylated MEK1 (bio-MEK1 in the figure) and [{gamma}-32P]ATP as substrates. MLK2/3 activity was quantitated by phosphorimaging. In parallel, the remainder of the extract was subjected to SDS-PAGE and immunoblotting with the indicated antibodies. KA, kinase assay; IP, immunoprecipitate; IB, immunoblot.

 
MLK2 and -3 Are Activated by PAMPs (LPS, FliC), Activation Requires GCK—The mechanisms by which GCK couples to MAPKs is unknown. Our previous studies indicated that GCK could activate the MAP3Ks MEKK1 and MLK3 but not MEKK3 in vitro (26). However, MEKK1 appears to function in the stabilization of GCK, rather than as a downstream effector (22). MLK3 and the closely related MLK2 contain amino-terminal Src homology 3 domains and can recruit both JNK and p38 (at least in overexpression studies). The carboxyl-terminal regulatory domain of GCK contains several proline-rich consensus SH3 interaction motifs and recombinant GCK and MLK3 interact in situ (20, 21, 26).

With these observations in mind, we asked if PAMPs could activate endogenous MLK3 in a GCK-dependent manner. Thus, gck was silenced as above, and cells were treated with LPS (Fig. 2, A and C) or FliC (Fig. 2B) for the indicated times. MLK3 (Fig. 2, A and B) or MLK2 (Fig. 2C) was immunoprecipitated and assayed for phosphorylation of a model substrate (biotinylated MEK1; see Ref. 19). From Fig. 2, it is evident that LPS and FliC trigger the characteristic (22) stabilization of GCK, first detectable at 1 min of stimulation and reaching a maximum by 3 min. Both MLK3 and -2 are activated by LPS, and MLK3 is also activated by FliC. In each case, activation is first detected at 1 min, and reaches a maximum at 10 min of stimulation. As with JNK and p38 activation, MLK3 and MLK2 activation by PAMPs is strikingly impaired upon silencing of gck. Thus, optimal MLK2/3 activation by LPS, PGN, and FliC requires GCK.


Figure 3
View larger version (54K):
[in this window]
[in a new window]

  		FIGURE 3.
Activation of recombinant JNK or p38 by coexpressed GCK requires both MLK3 and MLK2. 293 cells were transfected with either HA-JNK3 (A) or HA-p38{alpha} (B) and, as indicated, either vector or FLAG-GCK. As indicated, cells were treated with siRNAs for either MLK2, MLK3, or both. After 16 h, extracts were prepared, and from a portion of the extract, JNK or p38 was immunoprecipitated with anti-HA and assayed either GST-c-Jun-(1-135) (JNK assay) or GST-ATF2-(1-91) (p38 assay) and [{gamma}-32P]ATP as substrates. Activity was quantitated as above. In parallel, the remainder of the extract was subjected to SDS-PAGE and immunoblotting with the indicated antibodies. KA, kinase assay; IB, immunoblot.

 
In further support of the idea that MLK2 and -3 are GCK effectors, we find that ectopic expression of GCK activates JNK and p38 in a manner completely ablated upon contemporaneous silencing of mlk2 and mlk3. Thus, 293 cells were subjected to siRNA-dependent RNAi of either mlk2, mlk3, or both. Cells were then transfected with either vector or GCK and either JNK or p38. JNK and p38 were then recovered by immunoprecipitation and assayed in vitro for phosphorylation of the substrate proteins c-Jun or activating transcription factor-2 (ATF2), respectively (Fig. 3). From Fig. 3A, it is evident that ectopically expressed GCK triggers a substantial activation of coexpressed JNK. This is only moderately affected upon silencing of mlk2 or mlk3 individually. However, contemporaneous silencing of both mlk2 and mlk3 abolishes GCK activation of coexpressed JNK. Activation of p38 by GCK also requires MLK2 and -3 (Fig. 3B). That ectopic GCK was able to activate p38 was surprising insofar as our previous studies (25) indicated that GCK was selective for the JNKs. However, subsequent work has shown that MLK3 and -2 can activate the p38 pathway (28). Of note, we consistently do not observe ERK or NF-{kappa}B activation upon overexpression of GCK.3


Figure 4
View larger version (54K):
[in this window]
[in a new window]

  		FIGURE 4.
In situ interactions between GCK and MLK3. A, endogenous Jurkat cell MLK3 and GCK interact in an LPS-stimulated manner. Jurkat cells were treated with LPS for the indicated times. MLK3 was immunoprecipitated, subjected to SDS-PAGE, and, to detect associated GCK, immunoblotting with anti-GCK as indicated. In parallel whole cell extracts (WCE in the figure) were subjected to SDS-PAGE and immunoblotting with the indicated antibodies. IP, immunoprecipitate; IB, immunoblot. B, schematic of GCK truncation mutants. CT, carboxyl-terminal domain; Leu-rich, leucine-rich; P1, P 2, P3, Pro-Glu-Ser/Thr (PEST) sequence. Numbers indicate amino acid positions. C, in situ interaction of MLK3 and GCK requires the third PEST sequence (P3 in the figure), which contains a consensus SH3-binding motif (20, 21). 293 cells were transfected with GST-MLK3 and the indicated FLAG-tagged GCK constructs. The GCK construct was immunoprecipitated and subjected to SDS-PAGE and immunoblotting with anti GST to detect associated GST-MLK3. wt, wild type. D, schematic of MLK3 truncation mutants. LeuZ, leucine zipper; C and CRIB, Cdc42/Rac interaction and binding domain; Pro-rich, proline-rich. E, MLK3 SH3 domain is necessary and sufficient for binding to GCK in situ. 293 cells were transfected with FLAG-GCK and the indicated GST-tagged MLK3 constructs. The MLK3 construct was isolated on glutathione-agarose and subjected to SDS-PAGE and immunoblotting with anti-FLAG to detect associated FLAG-GCK.

 
We have previously shown that purified, recombinant GCK can interact with and activate recombinant MLK3 in vitro (26); however, we have not examined whether the endogenous proteins interact in an agonist-dependent manner. It is also not known what domains of each protein are necessary for in situ interaction. From Fig. 4A, it is evident that endogenous GCK and MLK3 can interact in situ in an LPS-stimulated manner. We cannot detect this interaction until after 30 min of LPS stimulation. We suspect that this is because of the detection limits of the available GCK and MLK3 antibodies, coupled with the low abundance of GCK in resting and even stimulated cells. Taken together, these issues would make it likely that the GCK-MLK3 interaction would not become apparent until the stimulus-dependent accumulation of sufficient GCK enabled detection of endogenous MLK3 in GCK complexes.


Figure 5
View larger version (89K):
[in this window]
[in a new window]

  		FIGURE 5.
Activation of JNK and p38, but not ERK, by PAMPs requires both MLK2 and -3. Jurkat cells were treated with either the principal (A, C, and D) or alternative (B) siRNAs ("Experimental Procedures") for MLK2, MLK3, or both as indicated. After 16 h, cells were treated with LPS (A and B), PGN (C), or FliC (D) for the indicated times (alternative siRNAs were used for the LPS-treated cells in these studies). Cell extracts were prepared, and activation of MAPKs was determined by subjecting a portion of each extract to SDS-PAGE and immunoblotting with phospho-specific antibodies and Image J quantitation, as indicated and described (see "Experimental Procedures" and the legends to Figs. 1, 2, 3). The remaining extracts were subjected to SDS-PAGE and immunoblotting with the other indicated antibodies or used in the studies described in Fig. 6A.

 
The GCK polypeptide contains a central proline-rich domain (amino acids 428-434) that corresponds to a consensus interaction motif for polypeptides with SH3 domains (8, 20, 21). MLK3 and MLK2 are quite similar structurally, and both have an amino-terminal SH3 domain (28). Fig. 4B is a schematic illustration of the structure of GCK along with several truncation constructs that we have generated. We expressed in 293 cells full-length MLK3 (GST-tagged) with each of these truncation constructs (FLAG-tagged), and in coimmunoprecipitation experiments (Fig. 4C), we find that deletion of the region containing the putative SH3 interaction motif abolishes the GCK-MLK3 interaction. Fig. 4D is a schematic illustration of the structure of MLK3. MLK2 has a similar overall molecular architecture (28). Fig. 4E shows the results of an experiment reciprocal to that of Fig. 4C. Thus, we expressed in 293 cells full-length GCK (FLAG-tagged) with either full-length or truncated MLK3 (GST-tagged). It is evident that deletion of the MLK3 SH3 domain abolishes MLK3 binding to GCK, whereas the free MLK3 SH3 domain alone retains the ability to bind GCK in situ. Thus, the MLK3 SH3 domain is necessary and sufficient for the in situ interaction of MLK3 with GCK.


Figure 6
View larger version (48K):
[in this window]
[in a new window]

  		FIGURE 6.
MLK2 and -3 are required for LPS-stimulated c-Jun phosphorylation and induction of il-8. Jurkat cells were treated with the alternative sets of siRNAs to MLK2, MLK3, or both ("Experimental Procedures"), as indicated. MLK2/3 knockdown was assayed in parallel with the studies in Fig. 5B, and the levels of MLK2/3 remaining in the cell extracts are shown there. A, cell extracts were subjected to immunoblotting with the indicated antibodies; and c-Jun phosphorylation was measured using Image J ("Experimental Procedures"). B, total RNA was extracted from a portion of the cells treated as in A and Fig. 5B. IL-8 and actin RNA were amplified by reverse transcriptase PCR. IL-8 mRNA was quantitated using Image J ("Experimental Procedures"). The white lines in the gel images indicate the boundaries of individual agarose gels used in the study. IB, immunoblot.

 
MLK2 and -3 Are Necessary for LPS, PGN, and FliC Activation of JNK and p38 but Not ERK; MLK2 and -3 Are Also Required for LPS-stimulated c-Jun Phosphorylation and Induction of IL-8—The results in Figs. 1, 2, 3, 4 establish that GCK is required for JNK and p38 activation by a subset of PAMPs (LPS, PGN, and FliC). These results also demonstrate that MLK2 and -3 are key GCK effectors coupling to JNK and p38. As would be expected from the findings in Figs. 1, 2, 3, 4, Fig. 5 shows that MLK2 and -3 are required for optimal JNK and p38 activation by LPS (Fig. 5, A and B), PGN (Fig. 5C), and FliC (Fig. 5D). MLK2 and -3 appear to be dispensable for ERK activation by these PAMPs. Thus, Jurkat cells were treated with human MLK2, MLK3, or MLK2 plus MLK3 siRNAs. Cells were then treated with PAMPs for the indicated times, and MAPKs were analyzed for activation by immunoblotting with phospho-specific antibodies. Silencing of either mlk2 or mlk3 individually has at best a modest effect on JNK or p38 activation; however, contemporaneous silencing of both MLK2 and MLK3 substantially diminishes LPS, PGN, and FliC activation of JNK and p38 but not ERK. Identical results were obtained with either set of mlk2/mlk3-specific siRNAs, for LPS-treated cells (Fig. 5, A versus B). Thus, it is unlikely that these findings can be attributed to off-target effects.

The AP-1 transcription factor is a heterodimer that consists of c-Jun and either other members of the Jun family (JunB and JunD), c-Fos, or ATF2 (9). AP-1 activation coincides with phosphorylation of c-Jun at Ser-63 and -73, a reaction catalyzed by the JNKs. We have previously shown that LPS activation of c-Jun phosphorylation and induction of il-8 require GCK (22). Consistent with the idea that MLK2 and -3 are GCK effectors, we find that, as with LPS activation of JNK (Fig. 5, A and B), optimal LPS stimulation of c-Jun phosphorylation requires both MLK2 and MLK3 (Fig. 6A). Thus, Jurkat cells were subjected to MLK2, MLK3, or MLK2 plus MLK3 RNAi as above and treated with LPS for various times. c-Jun phosphorylation was first detected within 10 min of stimulation and was maximal at 60 min, consistent with the role of c-Jun as a distal effector in LPS signaling. Silencing of mlk2 or mlk3 individually had essentially no effect on LPS-stimulated c-Jun phosphorylation. By contrast, silencing of both mlk2 and mlk3 resulted in a striking reduction in LPS-stimulated c-Jun phosphorylation (Fig. 6A).

The induction of IL-8 is a well established response of human cells to LPS and other PAMPs. This induction involves both transcriptional and post-transcriptional events, the latter being dependent upon p38 (1-6, 13). Thus, the IL-8 mRNA contains an adenine/uracil-rich element in its 3'-untranslated region. This domain is required for constitutive degradation of the IL-8 message in a manner rapidly inhibited, in a signal-dependent manner, by the p38 substrate kinase MAPK-activated protein kinase-2 (MK2) (29). From Fig. 6B, it is evident that LPS stimulates the rapid accumulation of IL-8 mRNA, an accumulation that occurs with kinetics consistent with mRNA stabilization. Thus, increased IL-8 mRNA is apparent within 3 min of LPS stimulation. Contemporaneous silencing of both mlk2 and mlk3 dramatically reduces this LPS induction of Jurkat cell IL-8 mRNA, a result that fits well with the observation that GCK and MLK2 and -3 are required for optimal LPS activation of p38.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
GCK is a key component in PAMP signaling. PRR engagement triggers the accumulation of the GCK polypeptide, primarily through the suppression of its ubiquitin-dependent proteolysis. This process depends on the interactions of GCK with TRAF6 and MEKK1 (22). The results presented herein provide insight into the mechanism by which a subset of PAMPs, through GCK, recruits two MAPKs and their effectors. Our results show that GCK is required for optimal LPS, PGN, and FliC recruitment of JNK and p38 but not ERK. GCK recruits these MAPKs by fostering the activation of MLK2 and -3, and MLK2 and -3 are required for maximal JNK and p38 activation, c-Jun phosphorylation, as well as induction of il-8 by LPS, PGN, and FliC. These findings are summarized in Fig. 7.

Recombinant GCK can directly activate recombinant MLK3 in vitro (26), and we now find that in situ, GCK activation of MLK3 coincides with their interaction. Thus, endogenous and recombinant GCK and MLK3 interact in intact cells, in a stimulus-dependent manner, by a process that requires the MLK3 SH3 domain and a proline-rich region of the GCK polypeptide. Given the close structural similarity between MLK3 and MLK2 (28), a similar mechanism may underlie GCK activation of MLK2.


Figure 7
View larger version (17K):
[in this window]
[in a new window]

  		FIGURE 7.
Schematic model for PAMP signaling mediated by GCK. The diagram illustrates how GCK transduces signals from PRRs that are engaged by LPS, PGN, and FliC. In this mechanism, PRR recruitment of TRAF6 (1-8) triggers the binding of ubiquitinated (Ub)-GCK and sequestration from the proteasome (22). MEKK1 also suppresses Ub-GCK degradation (22). GCK accumulates and can then bind and activate MLK2 and -3, resulting in activation of the JNKs and p38s as well as consequent gene induction. Whether or not the relevant PRRs are at the cell surface or are cytosolic remains to be determined. For the sake of clarity, other pathways known to emanate from PRRs are not shown. These are reviewed elsewhere (1-8). For details see text.

 
We were surprised to observe that GCK was important to PAMP activation of p38. Our earlier studies indicated that GCK was a relatively specific JNK pathway activator (25). Yet our results (Fig. 3) indicate that ectopic GCK can indeed activate p38, in contrast to previous results (25). The reason for this discrepancy is unknown but may be due to the limitations of earlier methods for assaying p38 activation. Moreover, consistent with a role for GCK in p38 regulation, MLK2 and -3, (both shown here and elsewhere (26) to be GCK effectors) have been linked both in these studies and in studies using transfection/overexpression to recruitment of p38 (28).

Interestingly, our previous work had shown that MLK3 was pivotal to mitogen and, to a lesser extent, cytokine activation of ERK (19). By contrast, disruption of mlk3 at best modestly impaired the activation, by TNF, of mouse embryonic fibroblast JNK and was without effect on the activation of p38 and ERK by all stimuli tested (30). Given the structural similarities between MLK3 and MLK2 (28), it is possible that the mouse mlk3 knockout was compensated for by MLK2. On the other hand, there is evidence from a number of different cell types for a diverse array of MAP3Ks linking PRRs to JNK and p38 (16-18), and the importance of the MLKs to MAPK activation may be considerably cell- and stimulus-dependent. Indeed, in line with this, the present studies show that, in contrast to mitogen signaling, ERK activation by PAMPs is unimpaired by silencing of MLK3, MLK2, or both. The lack of a role for GCK or MLKs in PAMP recruitment of the ERKs is perhaps best explained by the observation that cytokine and PAMP activation of ERK involves the MAP3K Tpl-2 (15).

GCK is central to JNK and p38 recruitment by some PAMPs (LPS, PGN, and FliC) but not by other PAMPs (poly(I/C)), CD40, IL-1, or TNF. This observation raises the intriguing question of whether other members of the GCK family are linked to proinflammatory signaling. The GCK family is large and can be subdivided into several groups (20, 21). Many members of the GCK family that are most closely similar to GCK itself do seem to function as part of either inflammatory/immune signaling pathways or pathways that involve TRAF proteins. Thus, GCK-related has been linked to TNF/TRAF2 and, surprisingly, to B lymphocyte Wnt activation of JNK (31, 32). Nck-interacting kinase is a highly conserved enzyme. Misshapen, the Drosophila orthologue of Nck-interacting kinase, is involved in the dorsal closure pathway (20, 21, 33). Interestingly, Misshapen is an effector for Drosophila TRAF in this pathway (33). Hematopoietic progenitor kinase-1 is a negative regulator of T cell receptor recruitment of AP-1 and may regulate the T cell receptor complex via phosphorylation of the adapter protein SH2 domain-containing leukocyte protein of 76 kDa (SLP-76). In this capacity, hematopoietic progenitor kinase-1 is thought to down-modulate T cell receptor signaling as part of a negative feedback/rheostat mechanism that tunes T cell activation (34, 35).

PAMP recruitment of JNK and p38 through GCK and MLK2 and -3 adds an additional level of complexity to what is known of PRR signaling to MAPKs. Recent studies indicate that PAMPs and other agonists that signal through TRAF6 likely recruit MAPKs through a surprisingly large number of cell- and stimulus-dependent mechanisms. In mouse embryonic fibroblasts LPS signaling through TRAF6 requires MEKK3 for JNK and p38 activation (18). By contrast, IL-1 activation of JNK and p38, a process that also proceeds through TRAF6, requires TAK1 (16, 17). In addition, TAK1 is required in T lymphocytes for T cell receptor, IL-2, -7, and 15 recruitment of JNK, and in fibroblasts and B lymphocytes for PAMP and cytokine activation of JNK (16, 17). By contrast, we find that GCK is required for maximal JNK activation by LPS and PGN in HL-60 macrophages (22). Moreover, the present studies show that GCK as well as MLK2 and -3 are required for optimum Jurkat cell JNK and p38 activation by LPS, PGN and FliC. On the other hand, as noted above, GCK is not rate-limiting, in Jurkat cells, for JNK activation by poly(I-C) or CD40 (22). We do not observe regulation of either MEKK3 or TAK1 by GCK (Ref. 26).4

The reasons for the bewildering complexity in PAMP, cytokine, and by extension TRAF-dependent signaling are unknown. The kinetics of MEKK3 and TAK1 activation by proinflammatory agonists are also unclear. GCK is stabilized within 1 min of PAMP addition, and MLK2/3 activation occurs nearly as rapidly and dwindles to base line within 30-60 min. It is possible that the other MAP3Ks are recruited with different kinetics and enable complex positive and negative regulation of MAPK activation, depending on the physiologic context. Such regulation, especially negative regulation, may ensure that proinflammatory signaling does not unnecessarily proceed to excess (e.g. sepsis) or trigger autoimmune responses. Clearly, additional animal models and biochemical studies will be needed to arrive at a clearer picture of the relative in vivo contributions of the proximal elements that mediate MAPK activation in response to inflammatory stimuli.


   FOOTNOTES
 
* This work was supported by United States Public Health Service NIGMS Grant R01-GM46577 (to J. M. K.) and by NHBLI Training Fellowship T32-HL069770 (to J. Z.) from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

1 To whom correspondence should be addressed: Molecular Cardiology Research Institute, Tufts-New England Medical Center, 750 Washington St., Box 8486, Boston, MA 02111. Tel.: 617-636-5190; Fax: 617-636-5204; E-mail: jkyriakis{at}tufts-nemc.org.

2 The abbreviations used are: PRR, pattern recognition receptor; AP-1, activator protein-1; ERK, extracellular signal-regulated kinase; FliC, bacterial flagellin; GCK, germinal center kinase; IL, interleukin; IRAK, IL-1 receptor-associated kinase; JNK, c-Jun NH2-terminal kinase; LPS, lipopolysaccharide; MAPK, mitogen-activated protein kinase; MEKK, MAPK/ERK kinase kinase; MKK, MAPK kinase; MAP3K, MKK kinase; MLK, mixed lineage kinase; MyD88, myeloid differentiation factor-88; NOD-LRR, nucleotide-binding oligomerization domain/leucine-rich repeat protein; NF-{kappa}B, nuclear factor-{kappa}B; PAMP, pathogen-associated molecular pattern; PEST, Pro/Glu-Asp/Ser/Thr-rich domain; PGN, peptidoglycan; RLH, retinoic acid-inducible gene-like helicase; RNAi, RNA interference; siRNA, small interfering RNA; TAK1, transforming growth factor-beta-activated kinase-1; TIR, TLR/IL-1 receptor; TLR, toll-like receptor; TNF, tumor necrosis factor; TRAF, TNF receptor-associated factor; GST, glutathione S-transferase; HA, hemagglutinin; SH, Src homology; ATF, activating transcription factor. Back

3 J. Zhong and J. M. Kyriakis, unpublished observations. Back

4 J. M. Kyriakis, unpublished observations. Back


   ACKNOWLEDGMENTS
 
We thank Cheleste Thorpe (Tufts-New England Medical Center) for FliC and discussions and Michael Mendelsohn for continued support.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Creagh, E. M., and O'Neill, L. A. J. (2006) Trends Immunol. 27, 352-357[CrossRef][Medline] [Order article via Infotrieve]
  2. Doyle, S. L., and O'Neill, L. A. J. (2006) Biochem. Pharmacol. 72, 1102-1113[CrossRef][Medline] [Order article via Infotrieve]
  3. Dunne, A., and O'Neill, L. A. J. (2003) Sci. STKE 2003, RE3
  4. Meylan, E., Tschopp, J., and Karin, M. (2006) Nature 442, 39-44[CrossRef][Medline] [Order article via Infotrieve]
  5. Akira, S., Uematsu, S., and Takeuchi, O. (2006) Cell 124, 783-801[CrossRef][Medline] [Order article via Infotrieve]
  6. Iwasaki, A., and Medzhitov, R. (2004) Nat. Immun. 5, 897-995
  7. Michelsen, K. S., and Arditi, M. (2006) Curr. Opin. Hematol. 13, 163-168[Medline] [Order article via Infotrieve]
  8. Kyriakis, J. M., and Avruch, J. (2001) Physiol. Rev. 81, 807-869[Abstract/Free Full Text]
  9. Karin, M., Liu, Z.-G., and Zandi, E. (1997) Curr. Opin. Cell Biol. 9, 240-246[CrossRef][Medline] [Order article via Infotrieve]
  10. Beyaert, R., Cuenda, A., Vanden Berghe, W., Plaisance, S., Lee, J. C., Haegeman, G., Cohen, P., and Fiers, W. (1996) EMBO J. 15, 1914-1923[Medline] [Order article via Infotrieve]
  11. Ahmad, M., Theofanidis, P., and Medford, R. M. (1998) J. Biol. Chem. 273, 4616-4621[Abstract/Free Full Text]
  12. Read, M. A., Whitley, M. Z., Gupta, S., Pierce, J. W., Best, J., Davis, R. J., and Collins, T. (1997) J. Biol. Chem. 272, 2753-2761[Abstract/Free Full Text]
  13. Dean, J. L. E., Sully, G., Clark, A. R., and Saklatvala, J. (2004) Cell. Signal. 16, 1113-1121[CrossRef][Medline] [Order article via Infotrieve]
  14. Lomaga, M. A., Yeh, W.-C., Sarosi, I., Duncan, G. S., Furlonger, C., Ho, A., Morony, S., Capparelli, C., Van, G., Kaufman, S., van der Heiden, A., Itie, A., Wakeham, A., Khoo, W., Sasaki, T., Cao, Z., Penninger, J. M., Paige, C. J., Lacey, D. L., Dunstan, C. R., Boyle, W. J., Goeddel, D. V., and Mak, T. W. (1999) Genes Dev. 13, 1015-1024[Abstract/Free Full Text]
  15. Dumitru, C. D., Ceci, J. D., Tsatsanis, C., Kontoyiannis, D., Stamakis, K., Lin, J.-H., Patriotis, C., Jenkins, N. A., Copeland, N. G., Kollias, G., and Tsichlis, P. N. (2000) Cell 103, 1071-1083[CrossRef][Medline] [Order article via Infotrieve]
  16. Sato, A., Sanjo, H., Takeda, K., Ninomiya-Tsuji, J., Yamamoto, M., Kawai, T., Matsumoto, K., Takeuchi, O., and Akira, S. (2005) Nat. Immun. 6, 1087-1095[CrossRef]
  17. Wan, Y. Y., Chi, H., Xie, M., Schneider, M. D., and Flavell, R. A. (2006) Nat. Immun. 7, 851-858[CrossRef]
  18. Huang, Q., Yang, J., Lin, Y., Walker, C., Cheng, J., Liu, Z. G., and Su, B. (2004) Nat. Immun. 5, 98-103[CrossRef]
  19. Chadee, D. N., and Kyriakis, J. M. (2004) Nat. Cell Biol. 6, 770-776[CrossRef][Medline] [Order article via Infotrieve]
  20. Kyriakis, J. M. (1999) J. Biol. Chem. 274, 5259-5262[Free Full Text]
  21. Dan, I., Watanabe, N. M., and Kusumi, A. (2001) Trends Cell Biol. 11, 220-230[CrossRef][Medline] [Order article via Infotrieve]
  22. Zhong, J., and Kyriakis, J. M. (2004) Mol. Cell. Biol. 24, 9165-9175[Abstract/Free Full Text]
  23. O'Neill, E., Rushworth, L., Baccarini, M., and Kolch, W. (2004) Science 306, 2267-2270[Abstract/Free Full Text]
  24. Kyriakis, J. M. (2007) Biochim. Biophys. Acta doi 10.1016/j.bbamcr.2006.11.003
  25. Pombo, C. M., Kehrl, J. H., Sánchez, I., Katz, P., Avruch, J., Zon, L. I., Woodgett, J. R., Force, T., and Kyriakis, J. M. (1995) Nature 377, 750-754[CrossRef][Medline] [Order article via Infotrieve]
  26. Chadee, D. N., Yuasa, T., and Kyriakis, J. M. (2002) Mol. Cell. Biol. 22, 737-749[Abstract/Free Full Text]
  27. Yuasa, T., Ohno, S., Kehrl, J. H., and Kyriakis, J. M. (1998) J. Biol. Chem. 273, 22681-22692[Abstract/Free Full Text]
  28. Gallo, K. A., and Johnson, G. L. (2002) Nat. Rev. Mol. Cell Biol. 3, 663-672[CrossRef][Medline] [Order article via Infotrieve]
  29. Frevel, M. A. E., Bakheet, T., Silva, A. M., Hissong, J. G., Khabar, K. S. A., and Williams, B. R. G. (2003) Mol. Cell. Biol. 23, 425-436[Abstract/Free Full Text]
  30. Brancho, D., Ventura, J.-J., Jaeschke, A., Doran, B., Flavell, R. A., and Davis, R. J. (2005) Mol. Cell. Biol. 25, 3670-3681[Abstract/Free Full Text]
  31. Shi, C.-S., and Kehrl, J. H. (1997) J. Biol. Chem. 272, 32102-32107[Abstract/Free Full Text]
  32. Shi, C. S., Huang, N. N., Harrison, K., Han, S. B., and Kehrl, J. H. (2006) Mol. Cell. Biol. 6, 6511-6521
  33. Liu, H., Su, Y.-C., Becker, E., Treisman, J., and Skolnik, E. Y. (1999) Curr. Biol. 9, 101-104[CrossRef][Medline] [Order article via Infotrieve]
  34. Liou, J., Kiefer, F., Dang, A., Hashimoto, A., Cobb, M. H., Kurosaki, T., and Weiss, A. (2000) Immunity 12, 399-408[CrossRef][Medline] [Order article via Infotrieve]
  35. Di Bartolo, V., Montagne, B., Salek, M., Jungwirth, B., Carrette, F., Fourtane, J., Sol-Foulon, N., Michel, F., Schwartz, O., Lehmann, W. D., and Acuto, O. (2007) J. Exp. Med. 204, 681-691[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
K. Clark, L. Plater, M. Peggie, and P. Cohen
Use of the Pharmacological Inhibitor BX795 to Study the Regulation and Physiological Roles of TBK1 and I{kappa}B Kinase {epsilon}: A DISTINCT UPSTREAM KINASE MEDIATES SER-172 PHOSPHORYLATION AND ACTIVATION
J. Biol. Chem., May 22, 2009; 284(21): 14136 - 14146.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
J. Zhong, L. C. Gavrilescu, A. Molnar, L. Murray, S. Garafalo, J. H. Kehrl, A. R. Simon, R. A. Van Etten, and J. M. Kyriakis
GCK is essential to systemic inflammation and pattern recognition receptor signaling to JNK and p38
PNAS, March 17, 2009; 106(11): 4372 - 4377.
[Abstract] [Full Text] [PDF]

Home page
 J. Nutr.Home page
B. Radnai, Z. Tucsek, Z. Bognar, C. Antus, L. Mark, Z. Berente, F. Gallyas Jr, B. Sumegi, and B. Veres
Ferulaldehyde, a Water-Soluble Degradation Product of Polyphenols, Inhibits the Lipopolysaccharide-Induced Inflammatory Response in Mice
J. Nutr., February 1, 2009; 139(2): 291 - 297.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Pathol.Home page
Y. Yan, G. Dalmasso, H. T. T. Nguyen, T. S. Obertone, L. Charrier-Hisamuddin, S. V. Sitaraman, and D. Merlin
Nuclear Factor-{kappa}B Is a Critical Mediator of Ste20-Like Proline-/Alanine-Rich Kinase Regulation in Intestinal Inflammation
Am. J. Pathol., October 1, 2008; 173(4): 1013 - 1028.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
282/33/24246    most recent
M703422200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhong, J.
Right arrow Articles by Kyriakis, J. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhong, J.
Right arrow Articles by Kyriakis, J. M.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2007 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement