PerspectiveCell Biology

Kinasing and Clipping Down the NF-κB Trail

See allHide authors and affiliations

Science  01 Apr 2005:
Vol. 308, Issue 5718, pp. 65-66
DOI: 10.1126/science.1110902

Cells of the innate and adaptive immune system recognize and respond to pathogenic insults with a diverse array of cell surface receptors. These distinct receptor complexes share one commonality: activation of the master transcription factor NF-κB. The basic mechanism of NF-κB activation involves the phosphorylation and ubiquitin-mediated degradation of its inhibitor IκB. Phosphorylation of IκB is carried out by a large kinase complex termed IKK, which integrates signals from multiple pathways. Some of the pathways leading to IKK activation are quite sophisticated, as exemplified by the T cell activation pathway (1). T cells express a cell surface receptor termed the T cell receptor (TCR) that recognizes foreign peptides embedded in host major histocompatibility complex (MHC) molecules. Research over the past decade has uncovered more than a dozen proteins that link the TCR to IKK, yet two recent papers in Science (2, 3), including one on page 114 of this issue, teach us that our understanding of the NF-κB pathway is far from complete. These studies add two more players, a kinase and a caspase, to the NF-κB pathway that is set in motion by TCR activation. The new studies underscore how exquisitely the adaptive immune system is regulated.

The recognition of suitable peptide-MHC ligands by TCR activates a sequence of intracellular protein tyrosine phosphorylation events. These events are coordinated by three families of protein tyrosine kinases: Src, Syk, and Tec (4). Upon TCR ligation, Src protein tyrosine kinases (Lck in T cells) selectively phosphorylate two tyrosine residues present in the immunoreceptor tyrosine-based activation motifs (ITAMs) of the TCR subunits (see the figure). In a biphosphorylated state, the ITAMs are bound in a complex with the tandem Src homology 2 (SH2) domains of a Syk protein tyrosine kinase called ZAP-70. Catalytic activation of ZAP-70 depends on its autophosphorylation and transphosphorylation by Lck. Activated ZAP-70 then phosphorylates tyrosine residues on an essential adaptor protein termed LAT (linker of activated T cells). LAT is a type II transmembrane protein that is localized in the plasma membrane by addition of palmitoyl groups to its two cysteine residues. The phosphorylated tyrosine residues in LAT provide key docking sites for a number of SH2-containing adaptor and effector proteins, including phospholipase C-γ, SLP76, and guanine nucleotide exchange factors (GEFs) such as Vav (see the figure).

The NF-κB activation pathway in T cells.

Stimulation of a T cell receptor (TCR) by a peptide-MHC complex and costimulation of CD28 by ligands on antigen-presenting cells activate a cascade of tyrosine phosphorylation. This cascade results in the activation of PI3K, PDK1, and PKCθ. PDK1 recruits IKK to the lipid rafts of the plasma membrane through PKCθ, and recruits BCL10 and MALT1 through CARMA1. PKCθ may also regulate the CARMA1-BCL10-MALT1 (CBM) complex through phosphorylation of its members. MALT1 mediates polyubiquitination of the IKK subunit NEMO directly or indirectly through the TRAF6 ubiquitin ligase. Caspase-8 is also required for IKK activation by facilitating the binding of IKK to the CBM complex in response to TCR stimulation. Membrane translocation and ubiquitination of IKK lead to its activation. Activated IKK phosphorylates IκB and targets this NF-κB inhibitor for ubiquitination and subsequent degradation by the proteasome. Degradation of IκB enables NF-κB to enter the nucleus and switch on expression of genes that are essential for the proliferation and function of activated T cells.


Through an unknown mechanism, SLP76 and Vav lead to the activation of the serine-threonine kinase PKCκ and the recruitment of this kinase to lipid rafts in the plasma membrane, where TCR bound to its ligand resides. According to the current model, PKCθ activates IKK through additional proteins including CARMA1 (also known as CARD11), BCL10, and MALT1, which form a complex [referred to as CBM, as suggested in (2)]. Each component of this complex is essential for NF-κB activation in T cells (5). MALT1 activates IKK by promoting the ubiquitination of the IKK subunit NEMO directly (6), or indirectly through the RING domain ubiquitin ligase TRAF6 (7, 8).

To elucidate the mechanism of PKCθ activation in T cells, Lee et al. (3) investigated the phospholipid-dependent kinase PDK1. As noted previously (9), PKCθ contains a putative PDK1 phosphorylation site at threonine 538 in its activation loop. PDK1 is a serine-threonine kinase regulated by phosphatidylinositol 3-kinase (PI3K), an enzyme that phosphorylates inositol phospholipids at the 3' position of the inositol ring to produce phosphatidylinositol 4,5-bisphosphate (PIP2) and phosphatidylinositol 3,4,5-trisphosphate (PIP3). These phospholipids recruit selected effector proteins (including PDK1) that contain a pleckstrin-homology (PH) domain. The mechanism of PI3K activation through TCR requires further investigation. However, PI3K is known to be activated by a second cell surface molecule termed CD28, which provides a critical costimulatory signal for T cell activation by interacting with ligands expressed on antigen-presenting cells. Inhibitors of PI3K block the translocation of PKCθ to the lipid rafts in stimulated T cells, whereas constitutively active PI3K or overexpression of PDK1 enhance translocation of PKCθ to the lipid rafts (10). Thus, it is likely that the PI3K pathway is involved in PKCθ activation. One study, however, found no signal-dependent translocation of PDK1 to the lipid rafts in T cells (10), so the link between PI3K and PKCθ remained unclear.

Lee et al. now provide strong evidence that PDK1 is required for the activation of PKCθ and IKK in T cells. After stimulation of TCR and CD28, PDK1 moves to the lipid rafts and then binds to and phosphorylates (presumably directly) PKCθ at threonine 538 in its activation loop. Silencing of PDK1 expression in Jurkat T cells by RNA interference (RNAi) abrogated PKCθ phosphorylation, IKK and NF-κB activation, and interleukin-2 production in response to TCR stimulation. RNAi of PDK1 also prevented the binding of PKCθ to IKK and blocked the translocation of these two kinases to lipid rafts. With these results in hand, Lee and co-workers propose that PDK1 recruits PKCθ to plasma membrane lipid rafts and that PKCθ in turn recruits IKK.

If PKCθ's only job is to recruit IKK to the plasma membrane, then what are the members of the CBM complex (CARMA1-BCL10-MALT1) doing, given that they are clearly essential for IKK activation? It turns out that PDK1 also binds to CARMA1 and is required for the recruitment of CARMA1 and BCL10 to the lipid rafts. These findings prompted Lee et al. to propose a dual role for PDK1 in the TCR pathway. PDK1 recruits not only IKK to lipid rafts through PKCθ but also the CBM complex to lipid rafts, where MALT1 activates IKK through ubiquitination of NEMO.

The model proposed by Lee et al. is a substantial revision of the current linear model in which PKCθ operates upstream of CBM, which in turn recruits IKK to the lipid rafts. Although PKCθ is required for IKK activation in mature T cells, it is not clear whether it is also required for the recruitment of CBM. The requirement of CARMA1 for IKK recruitment needs further investigation. Lee et al. showed that CARMA1 deficiency did not affect the recruitment of IKK to PKCθ. However, two other groups—one using the same CARMA1-deficient cell line, the other using T cells derived from CARMA1-deficient mice—showed that CARMA1 is required for recruiting IKK to the lipid rafts (11, 12). Although the exact biochemical mechanism of IKK activation by PKCθ remains to be resolved, the establishment of PDK1 as an essential upstream kinase of PKCθ has filled an important gap in the TCR pathway.

Another player in the TCR pathway is caspase-8, a cysteine protease that initiates apoptosis through “clipping” various cellular proteins. Surprisingly, T cells from patients carrying mutations in caspase-8 not only are defective in apoptosis, but also fail to activate NF-κB after TCR stimulation (13). In a recent study, Su et al. further investigated the mechanism by which caspase-8 mediates NF-κB activation in T cells, and they make several important observations (2). First, the enzymatic activity, but not autoprocessing, of caspase-8 is required for restoring NF-κB activation in a caspase-8-deficient cell line. Second, caspase-8 binds to CBM and IKK upon TCR stimulation. In cells lacking caspase-8, BCL10 still binds to MALT1, but this complex no longer recruits IKK in response to signals. Third, FADD, an adaptor protein known to interact with caspase-8, also binds to BCL10 and MALT1 in a signal-dependent manner. Taken together, these results suggest that FADD and caspase-8 bind to CBM and facilitate the recruitment and activation of IKK. Currently, it is not clear where caspase-8 is positioned in the IKK pathway in T cells. Caspase-8 may work upstream of CBM to facilitate the oligomerization and activation of CBM, thus promoting the recruitment of IKK. Alternatively, caspase-8 could bridge CBM and IKK, facilitating binding between these two complexes.

The signaling capabilities of caspase-8 appear to be an evolutionarily conserved mechanism in innate and adaptive immunity. Indeed, the Drosophila caspase-8 homolog Dredd is essential for the antibacterial response mediated by the NF-κB homolog Relish (14). In cells treated with caspase inhibitors or in Dredd-deficient cells, Relish cannot be cleaved into a mature subunit in response to bacterial challenge. Although Dredd is a putative protease of Relish, so far there is no biochemical evidence that Dredd cleaves Relish directly. Instead, there is now evidence that Dredd mediates the activation of Drosophila IKK, which phosphorylates and targets Relish for cleavage (15). Thus, the caspase-8 family of proteins has evolved to acquire dual roles in apoptosis and immunity, both of which are beneficial to the hosts in their battle against pathogens.

The revelation that PDK1 and caspase-8 play key parts in TCR-mediated activation of IKK sets the stage for unraveling the biochemical mechanisms of T cell signaling. Key questions remain to be addressed: How are PI3K and PDK1 activated by early signaling events in the TCR pathway? What are the physiological substrates of PKCθ? Does PKCθ help to recruit CBM? Is CBM required for the recruitment of IKK? What are the physiological targets of caspase-8 in the IKK pathway? How is caspase-8 activated by TCR stimulation? Answers to these and other questions will undoubtedly add to a more complete picture of T cell activation, which has been a core subject of immunology research. Because PDK1, PKCθ, and caspase-8 are enzymes that are amenable to chemical intervention, specific inhibitors of these enzymes could potentially become a new generation of immunosuppressive agents. The recent work on PDK1 and caspase-8 is a timely reminder that perhaps there are more treasures hidden down the NF-κB trail.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.

Navigate This Article