Like a finely fitted suit: IKKε tailors the antiviral response.

Every pathogen is different, so every immune response must be different. There is no “one size fits all” garment in immunology! This tailoring can occur at multiple levels. One systemic response is the development of “Th1” versus “Th2” CD4 T cells, which facilitate immune responses against (generally) intracellular and extracellular pathogens, respectively. On a more localized scale, different cell types, such as cardiac fibroblasts and myocytes, express different basal levels of immune components to dictate how the immune response behaves in particular tissues, such as the heart (1). Recent research by Sze-Ling Ng and colleagues has highlighted another level of tailoring, at the level of how individual cells coordinate their gene expression in response to distinct antiviral cytokines.

            Interferons are central components of the antiviral response. There are several types of these secreted proteins, including type I interferons (IFN-I), which help cells inhibit virus replication, and type II interferons (IFN-II), which are pro-inflammatory: they recruit immune cells to the site of infection. Both types act in a similar manner, by binding to receptors on the cell surface and inducing signals that culminate in a change in gene expression in the cell. Although IFN-I and IFN-II bind to different receptors, they share components of their signal transduction mechanism, including the transcription factor STAT1. Following IFN-I signaling, STAT1 becomes active and binds with two other proteins, STAT2 and IRF-9. Together, this complex binds to specific DNA sequences to mediate transcription of genes (interferon-stimulated genes, or ISGs) that mediate the antiviral state. Following IFN-II signaling, STAT1 also becomes active, but binds to itself to form a STAT1:STAT1 dimer. The STAT1 homodimer binds to DNA to mediate transcription of a distinct, but partially overlapping, set of genes that lead to the inflammatory response. A central question has been how cells balance these two responses, especially since both IFN-I and IFN-II are likely to be produced upon pathogen infection. Ng et al. now demonstrate that a cellular kinase, IKKε, phosphorylates STAT1, to alter its transcriptional response to IFNs.

            IKKε is best known for its role in inducing production of IFN-I in response to viral infection. However, genetic studies in mice have shown that a highly related kinase, TBK-1, is sufficient to induce IFN-I, raising the question of how IKKε contributes to the interferon response. One possibility is that IKKε is only required in a subset of specialized cells. Alternatively, because IKKε is inducible whereas TBK-1 is constitutively present, others speculated that IKKε functioned as a positive feedback mechanism. It was previously shown that IKK<ε could phosphorylate STAT1 on a serine at position 708 (2), and that this modification was necessary for an efficient immune response. Ng et al now show that this phosphorylation alters the capacity of STAT1 to form homodimers, leading to increased formation of the STAT1:STAT2:IRF-9 complex, and altering the balance of IFN-I vs. IFN-II transcription in cells.

            The authors started by examining production of IFN-dependent genes in influenza-infected cells that were either wild type or genetically deficient in IKKε (IKKε-/-). IFN-I-stimulated genes were present in decreased amounts in the IKKε-/- cells, whereas levels of IFN-II-stimulated transcription were increased. This corresponded to decreased binding of STAT1:STAT2:IRF-9 complexes to DNA, and increased binding of STAT1:STAT1 dimers. A previously solved crystal structure of STAT1 dimers bound to DNA showed that serine 708 is prominently located in the dimer interface, and contributes to the hydrogen bonds that hold the dimers together (3). Thus, Ng and colleagues reasoned that IKKε-dependent phosphorylation might inhibit STAT1 homodimer formation while still allowing STAT1:STAT2:IRF-9 interactions. Using mutant forms of STAT1 and IKKε, they demonstrated that this was indeed the case, as expression of active IKKε inhibited the association of STAT1 with itself but not with STAT1.

            Finally, the authors examined gene expression following IFN-I and IFN-II treatment of cells in the presence or absence of IKKε. To do this, they used RNA deep sequencing (RNAseq) and chromatin immunoprecipitation sequencing (ChIP-seq), powerful techniques for assessing the abundance of cellular mRNAs, and for determining the location of transcription factors on genomic DNA, respectively. The RNAseq analysis allowed them to classify genes according to their “beta-gamma mixture (BGM).” (IFN-I includes interferon-beta, whereas IFN-II is interferon-gamma, giving rise to this nomenclature). Since the pools of genes induced by IFN-I and IFN-II overlap, the authors could analyze the expression of genes that were primarily IFN-I-induced versus those that were primarily IFN-II-induced, or those that were induced by both. In the absence of IKKε, genes with a lower BGM (indicating stronger IFN-I responses) were expressed at lower levels; conversely, genes with a higher BGM were expressed at higher levels. Again, this corresponded with altered binding of STAT1 homodimers versus STAT1:STAT2:IRF-9 complexes to DNA, as determined by ChIP-seq. Thus, IKKε functions to alter the IFN-I and IFN-II responses in cells, by phosphorylating the common STAT1 component in the signaling pathway and leading to a bias towards induction IFN-I stimulated genes.

            So, what does all this mean in terms of the immune response? IKKε is activated in cells that are actively infected. Thus, IFN-I responses could be critical to induce the antiviral state in order to inhibit virus replication. If IFN-II is present, the pool of STAT1 may be limited, as IFN-II signaling might tie up all of the available STAT1 in STAT1:STAT1 homodimers. IKKε could thus prevent homodimer formation in these cells and facilitate induction of the IFN-I response. This explains the observation that IKKε-deficient mice are hypersusceptible to viral infection, despite production of normal levels of IFN, and provides a marvelous example of how the interrelationships between signaling pathways can finely tailor the immune response in response to pathogen infection.

Primary paper:

Ng SL, Friedman BA, Schmid S, Gertz J, Myers RM, Tenoever BR, and Maniatis T. 2011. IκB kinase ε (IKKε) regulates the balance between type I and type II interferon responses. Proc Natl Acad Sci U S A. 108(52):21170-5.

Other citations:

(1) Zurney J, Howard KE, and Sherry B. 2007. Basal expression levels of IFNAR and Jak-STAT components are determinants of cell-type-specific differences in cardiac antiviral responses. J. Virol. 81(24):13668-80.

(2) Tenoever BR, Ng SL, Chua MA, McWhirter SM, García-Sastre A, and Maniatis T. 2007. Multiple functions of the IKK-related kinase IKKepsilon in interferon-mediated antiviral immunity. Science. 315(5816):1274-8.

(3) Chen X, Vinkemeier U, Zhao Y, Jeruzalmi D, Darnell JE Jr, and Kuriyan J. 1998. Crystal structure of a tyrosine phosphorylated STAT-1 dimer bound to DNA. Cell. 93(5):827-39.