Characterization of putative STAT2 phosphorylation sites that regulate the cellular responses to type I interferons

128

Abstracts / Cytokine 48 (2009) 91–137

The E3 ubiquitin ligase Triad3A negatively regulates the RIG-I/MAVS signaling pathway by targeting TRAF3 for degradation Peyman Nakhaei 1,2, Thibault Mesplede 1,3, Mayra Solis 1,2, Qiang Sun 1, Tiejun Zhao 1, Long Yang 1, Tsung-Hsien Chuang 4, Carl F. Ware 5, Rongtuan Lin 1,3, John Hiscott 1,2,3, 1 The Terry Fox Molecular Oncology Group, Lady Davis Institute for Medical Research, McGill University, Montreal, Que., Canada, 2 Department of Microbiology & Immunology, McGill University, Montreal, Que., Canada, 3 Department of Medicine, McGill University, Montreal, Que., Canada, 4 Department of Immunology, The Scripps Research Institute, La Jolla, CA, USA, 5 The Division of Molecular Immunology, La Jolla Institute for Allergy and Immunology, San Diego, CA, USA The primary role of the innate immune response is to limit the spread of infectious pathogens, with, activation of Toll-like receptor (TLR) and RIG-like receptor (RLR) pathways resulting in a pro-inflammatory response required to combat infection. Limiting the activation of these signaling pathways is likewise essential to prevent tissue injury in the host. Triad3A is an E3 ubiquitin ligase that interacts with several components of TLR signaling and modulates TLR activity. In the present study, we demonstrate that Triad3A negatively regulates the RIG-I RNA sensing pathway through Lys48-linked, ubiquitin-mediated degradation of the tumor necrosis factor receptor-associated factor 3 (TRAF3) adapter. Triad3A was induced following dsRNA exposure or virus infection and decreased TRAF3 levels in a dose-dependent manner; moreover, Triad3A expression blocked IRF-3 activation by Ser-396 phosphorylation and inhibited the expression of type 1 interferon and antiviral genes. Lys48-linked ubiquitination of TRAF3 by Triad3A increased TRAF3 turnover, whereas reduction of Triad3A expression by stable shRNA expression correlated with an increase in TRAF3 protein expression and enhancement of the antiviral response following VSV or Sendai virus infection. Triad3A and TRAF3 physically interacted together, and TRAF3 residues Y440 and Q442 — previously shown to be important for association with the MAVS adapter — were also critical for Triad3A. Point mutation of the TRAF-Interacting-Motif (TIM) of Triad3A abrogated its ability to interact with TRAF3 and modulate RIG-I signaling. TRAF3 appears to undergo sequential ubiquitin ‘‘immuno-editing” following virus infection that is crucial for regulation of RIG-I dependent signaling to the antiviral response. Thus, Triad3A represents a versatile E3 ubiquitin ligase that negatively regulates RIG-like receptor signaling by targeting TRAF3 for degradation following RNA virus infection.

doi:10.1016/j.cyto.2009.07.540

PP2-163 Activation of ID3 expression by TGFb in primary human macrophage reveals involvement of Smad1/5 mediated signaling Dinara Nurgazieva, Ming Wen, Julia Kzhyshkowska, Sergij Goerdt, Alexei Gratchev, Poster Presentation II Activation of ID3 expression by TGFb in primary human macrophage reveals involvement of Smad1/5 mediated signaling Dinara Nurgazieva, Ming Wen, Julia Kzhyshkowska, Sergij Goerdt, Alexei Gratchev, Department of Dermatology, Medical Faculty Mannheim, Ruprecht-Karls-University of Heidelberg, Mannheim, Germany Macrophages encounter TGFb in various normal and pathological situations. The effect of TGFb on macrophages is, however, poorly studied. Recently we demonstrated that glucocorticoids are necessary to maintain the ability of macrophages to respond to TGFb stimulation. Expression profiling of TGFb-treated macrophages revealed activation of the expression of ID3 — a members of the ID family of helix-loop-helix (HLH) proteins. To understand mechanism of ID3 expression regulation its promoter was cloned and analysed using luciferase reporter assay. Interestingly stimulation of HepG2 cells transfected with reporter constructs, containing ID3 promoter, with TGFb did not induce an increase of luciferase activity. Also addition of enhancers identified using ECR Browser to the reporter constructs did not provide TGFb reactivity. Since ID3 mRNA expression was shown to be bone morphogenetic protein (BMP) dependent, we tested whether BMP signalling is activated in TGFb-treated macrophages. In contrast to TGFb signalling that is initiated by TGFb receptor I ALK5 and involves phosphorylation of Smad2/3, BMP signalling is initiated by ALK1/2/3/6 type I receptors and involves phosphorylation of Smad1/5. We established that stimulation of macrophages by TGFb leads to rapid phosphorlyation of Smad1/5. Inhibition of ALK5 receptor by SB431542 inhibitor efficiently blocked this phosphorylation. Analysis of concentration dependency of observed effect revealed, that different concentrations of SB431542 are needed to block Smad1/5 and Smad2/3 phosphorlyation. Since it was suggested that Smad1/5 phosphorlyation may be mediated via MAPK/ ERK pathway we tested whether treatment of macrophages with PD98059 — a Ras/ MAPK inhibitor will block Smad1/5 activation. PD98059 used in a wide range of concentrations did not have any effect on Smad1/5 or Smad2/3 phosphorlyation. Taken together we established that in primary human macrophages TGFb activates both TGFb and BMP signalling pathways. In contrast to published data in macrophages TGFb-mediated phosphorylation of Smad1/5 is not mediated by MAPK/ERK pathway. doi:10.1016/j.cyto.2009.07.541

PP2-164 Discovering and circumventing pitfalls analyzing fluorescence resonance energy transfer between proteins in cells: application to IFN receptors Christopher D. Krause, Lara S. Izotova, Barbara Schwartz, Chinkuei Kuo, Gina DiGioia, Sidney Pestka, Poster Presentation II Over the last several years, we used fluorescent resonance energy transfer (FRET) to demonstrate interactions between several cytokine receptor pairs tagged on their intracellular domains with fluorescent proteins (FPs) in their cellular environments. Complicating this analysis are cell-to-cell variations in the observed FRET between protein pairs tagged to donor and acceptor fluorescent proteins. Characterizing these variations will aid an accurate determination of FRET between protein pairs. To this end, we employed a multi-faceted approach involving (1) mathematic separation of fluorescent components seen in an complex emission spectrum, (2) advanced analysis of populations of mathematically-analyzed spectra, (3) use of newer FPs arising from different species or exhibiting different colors, (4) alternate fluorescence technologies. We found previously that lower FRET values between a given pair of proteins were seen when tagged to red-shifted FPs; this corresponded to lower levels of endogenous fluorescence present in the observed emission spectra. To evaluate quenching of FPs by endogenous fluorophores, we employed fluorescence lifetime methods. We found that fluorescence from blue-shifted FPs in cells is more strongly quenched than of red-shifted FPs. We discovered that use of donor:acceptor FP pairs derived from the same species can lead to artefactual protein:protein interactions mediated by the fluorescent proteins themselves. Using optimized FP pairs arising from distinct species, we report that while various interferon-gamma receptors interact with varying relative affinities, interactions between interferon-alpha receptor chains are only seen at very low levels. Importantly, biological activity of reconstituted interferon-alpha receptor chains is seen only when both receptor chains are expressed at low levels and when there is significant preassociation between the chains. This supports our previous hypothesis that preassociation of the receptor complex is a prerequisite for biological activity. Mechanisms explaining the different observed structural behavior of the interferon-gamma and interferon-alpha receptor complexes will be discussed. doi:10.1016/j.cyto.2009.07.542

PP2-165 Sting is an endoplasmic reticulum adaptor that facilitates innate immune signaling Hiroki Ishikawa, Glen N. Barber, Poster Presentation II Sting is an endoplasmic reticulum adaptor that facilitates innate immune signaling Hiroki Ishikawa, Glen N. Barber, Department of Medicine and Sylvester Comprehensive Cancer Center, University of Miami School of Medicine, Miami, FL 33136, USA Activation of host innate immune responses following virus infection is largely mediated by viral encoded RNA and DNA, the mechanisms of which remain to be fully determined. For example, the DExD/H box RNA helicases RIG-I/MDA5 are known to complex with intracellular viral RNA and to require the CARD containing molecule, IPS-1/MAVS/VISA/Cardif, to activate NF-jB and IRF3 and trigger type I interferon (IFN) production. A TLR-dependent pathway also exists to recognize viral RNA species, mainly in dendritic cells and macrophages, with TLR3 and TLR 7/8 facilitating activation of the IFN pathway in response to dsRNA and ssRNA, respectively. In contrast, little is known regarding the activation of the IFN pathway in response to intracellular viral DNA. Here we report the identification, following expression cloning, of a molecule (STING; stimulator of interferon genes) that appears essential for effective innate immune signaling processes. STING comprises five putative transmembrane regions, predominantly localizes in the endoplasmic reticulum and is able to activate both NF-jB and IRF3 transcription pathways to induce expression of type I interferon (IFN- and IFN-). In contrast, loss of STING rendered murine embryonic fibroblasts extremely susceptible to negative-stranded virus infection, including vesicular stomatitis virus. Interestingly, STING ablation abrogated the ability of intracellular Bform DNA, as well as members of the herpesvirus family, to induce IFN-. STING was found to interact with RIG-I and with TRAP, a member of the translocon-associated protein (TRAP) complex required for protein translocation across the endoplasmic reticulum membrane following translation. Ablation by RNA interference of both TRAP and translocon adaptor SEC61 was subsequently found to inhibit STING’s ability to stimulate expression of IFN-. Thus, as well as identifying a regulator of innate immune signaling, our results imply a potential role for the translocon in innate signaling pathways activated by select viruses as well as intracellular DNA. doi:10.1016/j.cyto.2009.07.543

PP2-166 Characterization of putative STAT2 phosphorylation sites that regulate the cellular responses to type I interferons Håkan C. Steen, Ana M. Gamero, Poster Presentation II

Abstracts / Cytokine 48 (2009) 91–137 Characterization of putative STAT2 phosphorylation sites that regulate the cellular responses to type I interferons Håkan C. Steen, Ana M. Gamero, Department of Biochemistry, Temple University, Philadelphia, PA, USA Type I interferons (IFN-a and -b) are used in the treatment of several diseases including multiple sclerosis, hepatitis C and certain malignancies. Type I IFNs primary mode of action in the treatment of cancer is the anti-proliferative and proapoptotic effects they exert on tumor cells. Yet the molecular mechanisms responsible for the pro-apoptotic effects of IFN-a/b are unclear. We have previously shown that STAT2 plays a vital role in the induction of apoptosis in certain cell lines treated with IFN-a/b alone or when in combination with the protein tyrosine phosphatase-inhibitor sodium orthovanadate. The latter observation suggests that kinase activity is essential in promoting the pro-apoptotic effects of IFN-a/b. Therefore, we hypothesized that additional signaling pathway(s) are involved in orchestrating the final outcome of the JAK/STAT-pathway by post-translationally modifying STAT2. We searched for putative phosphorylation sites in STAT2 by performing mass spectrometry analysis and using prediction programs (NetPhos 2.0 and Motif Scan). Six putative phosphorylation sites situated in the N-terminal, coiled-coil and DNA-binding domains were identified. Transient luciferase reporter assays showed that mutation of serine(S)-287, threonine(T)-387 or T404 of STAT2 to alanine (A) inhibited IFN-a/b mediated gene transcription. In addition, western blot analysis revealed that STAT2 T387>A was phosphorylated on Y690 in the absence of IFN-a/b stimulation thus suggesting that phosphorylated T387 may negatively regulate STAT2 activation. Possible serine/threonine kinases responsible for phosphorylating STAT2 are casein kinase II (CKII; S287, T404) and CDK5 (T387). Inhibition of CKII and CDK5 by DRB and Roscovitine, respectively, during IFN-atreatment inhibited the induction of several IFN stimulated genes. These results indicate that accessory signaling pathways influence the cellular response to type I IFNs and will therefore be important to characterize to improve IFN-based therapy. doi:10.1016/j.cyto.2009.07.544

PP2-167 Low toxicity signaling by monophosphoryl Lipid A Thomas C. Mitchell, Caglar Cekic, Carolyn R. Casella, Chelsea A. Eaves, Poster Presentation II Low toxicity signaling by monophosphoryl Lipid A Thomas C. Mitchell, Caglar Cekic, Carolyn R. Casella, Chelsea A. Eaves, University of Louisville, Department of Microbiology and Immunology, Institute for Cellular Therapeutics, Louisville, KY, USA Development of new vaccine adjuvants is constrained by a need for immunostimulatory compounds that generate very low rates of inflammatory side effects. An extreme example is LPS, whose endotoxicity stems from dual stimulation of the MyD88 and TRIF adaptor pathways downstream of Toll-like receptor 4 (TLR4), but whose monophosphorylated Lipid A (MPL) derivative is gaining regulatory acceptance as a vaccine additive. We reported previously that a commercially available form of MPL, MPLA, exerts TRIF-biased signaling effects in monocyte/macrophage cells that may explain MPL’s success as a safe immunostimulant. We now report several advances in our understanding of TRIF-biased signaling, including: (1) it is observed in dendritic cells, the cells responsible for T cell priming during immunization, (2) it occurs when comparing synthetic monophosphorylated Lipid A (sMLA) to synthetic diphosphate Lipid A (sDLA), showing that loss of a single phosphate is sufficient for differential adaptor stimulation, and (3) it is observed in IL1R-deficient cells, indicating that sMLA’s defective activation of the IL-1beta inflammasome is a result, not a cause, of TRIF-biased signaling activity. Quantitative signaling analyses show, moreover, that sMLA’s TRIF-biased pattern of gene expression is associated with robust activation of the mitogen-activated protein kinase (MAPK) p38 and weak activation of TAK1 and JNK, demonstrating that selective stimulation of some but not all MAPK’s is an attribute of sMLA’s low toxicity signaling through TLR4. These findings increase our ability to design and deploy next-generation TLR4 agonists as safe vaccine adjuvants.

129

The cellular protein P58IPK inhibits PKR, an interferon-induced, dsRNA-activated protein kinase activated during influenza virus infection. This inhibition aids influenza virus replication by reducing PKR-mediated eIF2a phosphorylation. Recently we used mutant mouse embryonic fibroblasts lacking P58IPK to show that influenza virus mRNA translation becomes efficient in the absence of P58IPK [Goodman AG et al. J Virol 2007;81:2221–30]. This decreased efficiency results from increased PKR activation and eIF2a phosphorylation and was independent of the ER stress-sensitive eIF2a kinase, PERK. To examine the connections within this pathway and to predict behavior of this biological system, we have created a mathematical model. This model complements our experiments and provides platform to identify key components and processes of influenza virus replication affected by P58IPK. In addition, this model can test conditions that cannot be tested in the lab. Using a system of ordinary differential equations, we created a model to investigate the experimental results of Goodman, et al. Our model suggests that another eIF2a kinase, also inhibited by P58IPK, is activated during virus infection. Setting the initial concentration of P58IPK to zero simulates the knockout conditions from our experiments and predicts a viral response consistent with the biological system devoid of P58IPK. The model presented here represents a first step in building a comprehensive model to explain influenza virus infection as it relates to P58IPK and viral replication. doi:10.1016/j.cyto.2009.07.546

PP2-169 Disparity in PTPN1 and PTPN2 modulation of IL-6 promoter activation by platelet-activating factor Geneviève Hamel-Côté, Steeve Veronneau, Simon Rollin, Marek Rola-Pleszczynski, Jana Stankova, Poster Presentation II Disparity in PTPN1 and PTPN2 modulation of IL-6 promoter activation by plateletactivating factor Geneviève Hamel-Côté, Steeve Veronneau, Simon Rollin, Marek Rola-Pleszczynski, Jana Stankova, Service d’Immunologie, Département de pédiatrie, Faculté de Médecine et Sciences de la Santé, Université deSherbrooke, Sherbrooke, Que., Canada Platelet-activating factor (PAF) is a potent pro-inflammatory phospholipid mediator that has been shown to trigger interleukin-6 (IL-6) production by binding a Gprotein coupled receptor, PAFR. This process is known to be dependent on tyrosine kinase activation, leading us to investigate if protein tyrosine phosphatases (PTP) are involved in modulation of PAF-induced IL-6 production. Results obtained by luciferase assays suggest opposing roles of PTPN1 and PTPN2 phosphatases in human IL-6 promoter activation triggered by PAF. The involvement of PTPN1 in PAFR signalization leading to IL-6 mRNA expression has been ascertained in primary monocytes using pharmacological inhibitors and PTPN1 siRNA. Our results indicate that PTPN1 down-regulates promoter activity by 50%, as seen in luciferase assays, in a way that is, at least in part, independent of G-protein activation but dependent on JAK2 activity. Alternatively, PTPN2 acts in an isoform-specific manner. Co-expression of the 48 kDa isoform of PTPN2, but not the 45 kDa, up-regulated IL-6 promoter by 40% activity after PAF stimulation. Preliminary results indicate that this up-regulation is dependent on PAFR internalization capacity by decreasing internalization after stimulation by around 64–77% at early time after stimulation. Together, our results suggest a differential role of these PTPs in modulating PAF-induced IL-6 promoter activity. doi:10.1016/j.cyto.2009.07.547

PP2-170 Biochemical monitoring of the early endocytic traffic of the type I interferon receptor Béatrice Payelle-Brogard, Sandra Pellegrini, Poster Presentation II Biochemical monitoring of the early endocytic traffic of the type I interferon receptor Béatrice Payelle-Brogard, Sandra Pellegrini, Institut Pasteur, Unit of Cytokine Signaling, CNRS URA 1961, Paris, France

doi:10.1016/j.cyto.2009.07.545

PP2-168 A mathematical model predicts the kinetics of P58IPK topology in response to influenza virus infection Alan G. Goodman, Bertrand C. W. Tanner, Wendy E. Thomas, Michael G. Katze, Poster Presentation II A mathematical model predicts the kinetics of P58IPK topology in response to influenza virus infection Alan G. Goodman 1, Bertrand C.W. Tanner 2, Wendy E. Thomas 1, Michael G. Katze 1, 1 University of Washington, Seattle, WA, USA, 2 University of Vermont, Burlington, VT, USA

The type I interferon (IFN) receptor consists of two transmembrane chains IFNAR1 and IFNAR2, associated with the tyrosine kinases Tyk2 and Jak1, respectively. Binding of IFN to this receptor complex induces activation of Jak/Stat and non-Stat signaling pathways. Ligand binding also drives receptor internalization and sorting towards degradation or recycling. To gain insights into receptor trafficking and its relation to signaling, we performed subcellular organelle fractionation from IFN-stimulated Daudi cells and defined biochemically an early endosomal antigen-1 (EEA1)positive compartment bearing the activated IFN receptor. Endosomes were thus purified by immunoaffinity isolation on anti-EEA1 antibodies-coated beads. The content of these purified endosomal fractions was analyzed by western blot and proteomics. Shortly after IFN stimulation, robustly ubiquitinated IFNAR1 and a small amount of IFNAR2 were found in this endosomal compartment, which also contained tyrosine