We are taking biochemical and genetic approaches to dissect two signaling pathways: ubiquitin-mediated activation of protein kinases and antiviral innate immunity. We have uncovered a novel function of ubiquitin in activating protein kinases in the nuclear factor κB (NF-κB) signaling cascade through a proteasome-independent mechanism. Our current effort is focused on elucidating the biochemical mechanism underlying the regulatory function of ubiquitin. In viral signaling, we are particularly interested in dissecting the signaling pathway by which a host cell mounts an immune response to RNA virus infection.
Ubiquitin-Mediated Activation of Protein Kinases in the NF-κB Pathway
The transcription factor NF-κB is activated by a large variety of agents, including proinflammatory cytokines and microbial pathogens. NF-κB is normally retained in the cytosol through its association with the inhibitory proteins of the IκB family. Stimulation of cells leads to the rapid activation of the IκB kinase (IKK) complex, which phosphorylates IκB and targets this inhibitor for polyubiquitination and subsequent degradation by the proteasome. This allows NF-κB to enter the nucleus to activate a battery of genes involved in inflammation, immunity, and survival of cells.
In the course of studying the regulation of IKK by TRAF6, a protein essential for NF-κB activation by interleukin-1 (IL-1) receptor (IL-1R) and Toll-like receptors (TLRs), we identified two factors that mediate IKK activation by TRAF6. One of these factors is a ubiquitin-conjugating enzyme (E2) complex consisting of Ubc13 and Uev1A, and the other is a protein kinase complex containing TAK1, TAB1, and TAB2. We found that TRAF6 functions as a ubiquitin ligase (E3) that, in conjunction with the Ubc13-Uev1A complex, catalyzes the synthesis of a unique polyubiquitin chain linked through lysine-63 of ubiquitin. This K63 polyubiquitination activates the TAK1 kinase complex through a proteasome-independent mechanism.
Further studies showed that TAB2 and its homolog TAB3 contain a highly conserved zinc finger domain known as the NZF domain, which binds preferentially to K63 polyubiquitin chains, and that this binding is required for TAK1 activation. After TAK1 is activated, it phosphorylates IKKβ at two serine residues within the activation loop, leading to activation of the IKK complex. We also identified a novel ubiquitin-binding domain (NUB) in NEMO, the essential regulatory subunit of the IKK complex. This domain and another zinc finger-type ubiquitin-binding domain at the carboxyl terminus of NEMO mediate specific binding to K63 polyubiquitin chains, which recruit IKK to the TAK1 complex, facilitating the phosphorylation of IKK by TAK1. We have recently developed a ubiquitin replacement strategy in human cells and demonstrated that K63 polyubiquitination is essential for IKK activation by IL-1β.
To understand how TAK1 is activated by K63 polyubiquitination, we used purified proteins to reconstitute TAK1 activation in vitro. This reconstitution system led to the discovery that unanchored K63 polyubiquitin chains, which are not conjugated to any other cellular protein, directly activate the TAK1 kinase complex through binding to TAB2. This binding leads to the oligomerization of TAK1, resulting in its autophosphorylation and activation. These results suggest that unanchored K63 polyubiquitin chains may function as "second messenger"–like signaling molecules to regulate protein kinases.
The RIG-I Antiviral Immunity Pathway
The RNA helicase proteins RIG-I (retinoic acid–inducible gene I) and MDA5 (melanoma differentiation antigen) detect infection of RNA viruses in the cytoplasm. RIG-I and MDA5 bind to viral double-stranded RNA. In addition, RIG-I specifically recognizes viral single-stranded RNA containing 5'-triphosphate, which is distinguished from the host RNA that is normally 5'-capped or contains other modifications. The binding of the viral RNA induces a conformational change that exposes the amino-terminal CARD domains of RIG-I and MDA5. These CARD domains interact with the CARD domain of another protein that we named MAVS (mitochondrial antiviral signaling). We found that MAVS is anchored to the mitochondrial outer membrane through a carboxyl-terminal transmembrane domain and that this mitochondrial localization is essential for its signaling function. MAVS activates the cytosolic kinases IKK and TBK1, which in turn activate the transcription factors NF-κB and IRF3, respectively. NF-κB and IRF3 then function together in the nucleus to turn on the production of type-I interferons and other antiviral molecules to suppress viral replication and assembly.
In addition to RNA, double-stranded DNA in the cytoplasm can also trigger robust production of interferons. DNA can be introduced into the cytoplasm by infection with DNA viruses or intracellular bacteria, or under some pathological conditions such as lupus. We recently identified DNA-dependent RNA polymerase III (Pol-III) as a sensor that detects some forms of cytosolic DNA and converts them into RNA containing 5'-triphosphate, which then activates the RIG-I pathway to induce type-I interferons. Pol-III is important for immune defense against some but not all DNA viruses and intracellular bacteria. Thus, other DNA sensors exist in different cells to mediate immune defense against a variety of pathogens. We are interested in dissecting the cytosolic DNA signaling pathways triggered by Pol-III and other DNA sensors.
Ubiquitin in RIG-I Signaling
We have recently reconstituted the RIG-I signaling cascade from viral RNA sensing to the activation of NF-κB and IRF3. RIG-I activation requires not only viral RNA but also K63 polyubiquitination. We found that the tandem CARD domains of RIG-I bind specifically to K63 polyubiquitin chains and that this binding in full-length RIG-I depends on RNA and ATP. Remarkably, unanchored K63 ubiquitin chains containing just three or four ubiquitin moieties are sufficient to cause full activation of RIG-I. We devised a strategy to isolate endogenous unanchored K63 ubiquitin chains from human cells and demonstrated that these chains are a potent ligand of RIG-I. Our results suggest a model in which viral RNA binds to RIG-I, causing a conformational change that recruits the ubiquitin ligase TRIM25 to synthesize K63 polyubiquitin chains, which then bind to the CARD domains of RIG-I, resulting in its full activation.
We are now further dissecting the RIG-I pathway, using the in vitro reconstitution system together with cell-based functional assays. Our goal is to understand the mechanism of signal transduction at each step of the RIG-I signaling cascade. Our recent results suggest that K63 polyubiquitination is also important for the activation of IKK and TBK1 by MAVS.
This research is supported in part by grants from the National Institutes of Health and the Robert Welch Foundation.
As of May 06, 2010