IFNβ responses induced by intracellular bacteria or cytosolic DNA in different human cells do not require ZBP1 (DLM-1/DAI)

Cellular Microbiology (2008) 10(12), 2579–2588

doi:10.1111/j.1462-5822.2008.01232.x First published online 10 October 2008

IFNb responses induced by intracellular bacteria or cytosolic DNA in different human cells do not require ZBP1 (DLM-1/DAI) Juliane Lippmann,1 Stefan Rothenburg,2 Nikolaus Deigendesch,3 Julia Eitel,1 Karolin Meixenberger,1 Vincent van Laak,1 Hortense Slevogt,1 Philippe Dje N’Guessan,1 Stefan Hippenstiel,1 Trinad Chakraborty,4 Antje Flieger,5 Norbert Suttorp1 and Bastian Opitz1* 1 Department of Internal Medicine/Infectious Diseases and Pulmonary Medicine, Charité Universitätsmedizin Berlin, Augustenburger Platz 1, 13353 Berlin, Germany. 2 National Institutes of Health, National Institute of Child and Human Development, Bethesda, MD 20892-2427, USA. 3 Institute of Immunology, University Hospital Hamburg-Eppendorf, Hamburg, Germany. 4 Institute for Medical Microbiology, Justus-Liebig University of Giessen, Frankfurter Strasse 107, 35392, Giessen, Germany. 5 Robert Koch-Institute, Research Group NG5 Pathogenesis of Legionella Infections, Nordufer 20, 13353 Berlin, Germany. Summary Intracellular bacteria and cytosolic stimulation with DNA activate type I IFN responses independently of Toll-like receptors, most Nod-like receptors and RIG-like receptors. A recent study suggested that ZBP1 (DLM-1/DAI) represents the long anticipated pattern recognition receptor which mediates IFNa/b responses to cytosolic DNA in mice. Here we show that Legionella pneumophila infection, and intracellular challenge with poly(dA-dT), but not with poly(dGdC), induced expression of IFNb, full-length hZBP1 and a prominent splice variant lacking the first Za domain (hZBP1DZa) in human cells. Overexpression of hZBP1 but not hZBP1DZa slightly amplified poly(dA-dT)-stimulated IFNb reporter activation in HEK293 cells, but had no effect on IFNb and IL-8 production induced by bacteria or poly(dA-dT) in A549 cells. We found that mZBP1 siRNA impaired

Received 11 March, 2008; revised 4 August, 2008; accepted 22 August, 2008. *For correspondence. E-mail bastian.opitz@ charite.de; Tel. (+49) 30 450 553383; Fax (+49) 30 450 553992. © 2008 The Authors Journal compilation © 2008 Blackwell Publishing Ltd

poly(dA-dT)-induced IFNb responses in mouse L929 fibroblasts at a later time point, while multiple hZBP1 siRNAs did not suppress IFNb or IL-8 expression induced by poly(dA-dT) or bacterial infection in human cells. In contrast, IRF3 siRNA strongly impaired the IFNb responses to poly(dA-dT) or bacterial infection. In conclusion, intracellular bacteria and cytosolic poly(dA-dT) activate IFNb responses in different human cells without requiring human ZBP1. Introduction The innate immune system serves as a first-line defence against invading pathogens including bacteria or viruses. It senses microbial components and endogenous dangerassociated molecules by so-called pattern recognition receptors (PRRs) such as the Toll-like receptors (TLRs), the Nod (nucleotide-binding oligomerization domain protein)-like receptors (NLRs), or the RIG (retinoicacid-inducible protein)-like receptors (RLRs), which, for example, mediate transcriptional or post-translational regulation of inflammatory mediators (Meylan et al., 2005; Akira et al., 2006; Fritz et al., 2006; Yoneyama and Fujita, 2007). Interferon (IFN)-a and -b constitute the type I IFN family, and were originally identified as humoral factors that confer an antiviral state on cells (Katze et al., 2002). The expression of IFNa/b can be induced by RLRs and by some TLRs through activation of IFN regulatory factor (IRF)-3/7 transcription factors. In contrast, most NLRs might not be able to regulate IRF3/7 activation and IFNa/b expression, except NLRX1 which negatively controls RLR-mediated IFNa/b responses (Moore et al., 2008). After expression and secretion, type I IFN binds to the IFNa/b receptor, which via signalling to the JAK/STAT pathway induces expression of so-called IFN-stimulated genes (ISGs), many of which have antiviral activities (Levy and Darnell, 2002). Besides its role in viral infections, several recent studies demonstrated that type I IFNs either contribute or detract from innate immune responses to various bacterial infections in vivo (Auerbuch et al., 2004; Carrero et al., 2004; O’Connell et al., 2004; Mancuso et al., 2007; Stanley et al., 2007). While some bacteria might trigger IFNa/b responses through TLR4 or TLR9, strong evidence for the

2580 J. Lippmann et al. existence of an additional cytosolic sensing mechanism has emerged. Studies by the groups of Portnoy and Decker showed that host cells infected with Listeria monocytogenes produced IFNb (O’Riordan et al., 2002; Stockinger et al., 2002). These IFN responses were dependent on the cytosolic localization of the bacteria (O’Riordan et al., 2002; Stockinger et al., 2002). Moreover, we and others found that also Legionella pneumophila or Mycobacterium tuberculosis, which replicate within cellular vacuoles, activated IFNb, an effect that was dependent on bacterial secretion systems potentially capable of injecting microbial molecules into the host cell cytosol (Opitz et al., 2006; Stetson and Medzhitov, 2006; Stanley et al., 2007). Recent studies indicated that this ‘cytosolic response’ does not only include type I IFNs but some 80 genes (many of which were ISGs) that were only induced by the cytosolic surveillance pathway (McCaffrey et al., 2004; Leber et al., 2008). In addition, IFNb produced by host cells infected with Legionella appears to mediate a cell-autonomous defence against the bacteria (Opitz et al., 2006; Coers et al., 2007). Microbial, endogenous or synthetic DNA which gained access to the host cell cytosol stimulated a similar transcriptional response including IFNb expression (Okabe et al., 2005; Ishii et al., 2006; Stetson and Medzhitov, 2006; Leber et al., 2008), which led to the suggestion that DNA represents the bacterial molecule which is sensed in the host cell cytosol of bacterial infected cells and activates this type of gene expression (Stetson and Medzhitov, 2006; Leber et al., 2008). The IFNb responses to intracellular bacteria or cytosolic DNA were dependent on TBK1 and IRF3, but independent of the TLRs, RLRs and most NLRs tested so far (O’Riordan et al., 2002; McCaffrey et al., 2004; O’Connell et al., 2005; Okabe et al., 2005; Ishii et al., 2006; Opitz et al., 2006; Stetson and Medzhitov, 2006; Stanley et al., 2007; Leber et al., 2008). Z-DNA-binding protein (ZBP)1 (also called DLM-1 or DAI) was identified as a gene that was upregulated in mouse tumour stromal cells (Fu et al., 1999), or in mouse macrophages by IFNg or cytosolic DNA stimulation (Fu et al., 1999; Ishii et al., 2006). ZBP1 contains two N-terminal Za domains and a C-terminus of unknown function (Rothenburg et al., 2002). In vitro, both Za domains of human ZBP1 have been shown to bind Z-DNA independently, in a comparable manner, and that binding to Z-DNA was greatly enhanced when both Za domains were present in the same molecule (Deigendesch et al., 2006). As a result of alternative splicing and the use of alternative transcriptional start and stop sites, human ZBP1 mRNA is very heterogeneous. Exon 2, coding for the first Za domain, is spliced out in ~50% of ZBP1 mRNA (Rothenburg et al., 2002). The two major human ZBP1 variants (full-length hZBP1 and hZBP1DZa) showed strikingly different subcellular localizations (Deigendesch

et al., 2006). Recently, it was suggested that murine ZBP1 senses cytosolic DNA and activates TBK1- and IRF3dependent IFNa/b responses (Takaoka et al., 2007). In our study we tested the hypothesis that human ZBP1 or its major splicing variant hZBP1DZa mediates IFNb responses to cytosolic DNA or infection with intracellular bacteria.

Results L. pneumophila and poly(dA-dT) induced IFNb, hZBP1 and hZBP1DZa expression in lung epithelial cells First, we found that Legionella infection as well as poly(dA-dT)-DNA, but not poly(dG-dC)-DNA transfection into A549 lung epithelial cells induced IFNb mRNA expression (Fig. 1A). Moreover, THP1 cells infected with L. monocytogenes or transfected with poly(dA-dT) also produced IFNb (data not shown). Next, we assessed expression of hZBP1 and hZBP1DZa in A549 cells. Quantitative PCR (Q-PCR) using primers which targeted exon 4 and 5 of hZBP1 and thus amplified both major hZBP1 isoforms demonstrated that L. pneumophila, poly(dA-dT) and IFNb induced hZBP1 expression (Fig. 1B–D). Furthermore, specific primers were used targeting exon 1 and exon 2 for hZBP-1 or exon 1 and exon 4 for hZBP1DZa respectively. L. pneumophila infection, poly(dA-dT) transfection and IFNb treatment stimulated a strong upregulation of both hZBP1 splice variants, while poly(dG-dC) did not induce hZBP1 or hZBP1DZa expression (Fig. 1B–D). In accordance with a recent study which detected multiple ZBP1 splice variants in addition to the two major mRNAs (Rothenburg et al., 2002), we also observed amplification of additional hZBP1 transcription variants in some settings (Fig. 1C, bottom panels, and data not shown).

Overexpression of hZBP1 only slightly amplified the poly(dA-dT)-stimulated IFNb reporter gene activation in HEK293 cells Many PRRs (or their truncated variants) and signalling molecules involved in type I IFN responses including melanoma-differentiation-associated gene 5 (MDA5), RIG-I lacking the helicase domain, IFNb promoter stimulator 1 (IPS-1), TRIF, TBK1 and IRF3 have been shown to stimulate IFNb, IRF3 and nuclear factor-kB (NF-kB) reporter gene activation when overexpressed (Andrejeva et al., 2004; Yoneyama et al., 2004; Kawai et al., 2005; Opitz et al., 2007). In contrast, our data show that neither full-length hZBP1, hZBP1DZa, nor the truncated derivatives of hZBP1 consisting of exons 1–4 or of exons 5–10 stimulated IRF-3, NF-kB and IFNb reporter genes when transiently overexpressed (data not shown). In addition,

© 2008 The Authors Journal compilation © 2008 Blackwell Publishing Ltd, Cellular Microbiology, 10, 2579–2588

Human ZBP1 and IFNb responses 2581

Fig. 1. L. pneumophila infection or cytosolic poly(dA-dT) stimulation induced expression of IFNb, full-length hZBP1 and hZBP1DZa in A549 cells. A. A549 cells were infected with wild-type L. pneumophila (L. p.) strain Corby for 9 h or were transfected with poly(dA-dT) or poly(dG-dC) and incubated for 9 h. IFNb and GAPDH expressions were analysed by RT-PCR. B–D. A549 cells were infected with wild-type L. pneumophila strain Corby (B) or were transfected with 1 mg of poly(dA-dT) or poly(dG-dC) (C) or were treated with 500 U ml-1 IFNb (D) and incubated for the indicated time intervals. Expression of hZBP1, hZBP1DZa and GAPDH mRNA was analysed by RT-PCR. Relative hZBP1 mRNA expression (RQ, relative quantification) was analysed by Q-PCR with quantification of total hZBP1 mRNA normalized to GAPDH mRNA and compared with untreated cells. Results shown are representative of three independent experiments performed in duplicates.

we observed stimulation of the reporter gene by poly (dA-dT) but not by poly(dG-dC) in control vectortransfected HEK293 cells (Fig. 2A). The poly(dA-dT)mediated IFNb promoter activation was slightly enhanced by transient transfection of hZBP1, but not of hZBP1DZa. Moreover, we made use of HEK293 cells stably expressing doxycycline-inducible eGFP, hZBP1 or hZBP1DZa. Immunoblot experiments demonstrated expression of the respective proteins after doxycycline treatment (Fig. 2B). Consistent with the aforementioned results, expression of hZBP1 but not of hZBP1DZa moderately enhanced the poly(dA-dT)-stimulated IFNb reporter gene activation (Fig. 2C). hZBP1 or hZBP1DZa overexpression did not enhance poly(dA-dT)-induced NF-kB reporter activation (data not shown). Overexpression of hZBP1 or hZBP1DZa had no effect on expression of IFNb or IL-8 induced by poly(dA-dT) stimulation or L. pneumophila infection in A549 cells As HEK293 cells are poorly equipped with many important PRRs and signalling molecules and are poorly invaded by Legionella, we next tested if overexpression of

hZBP1 affected IFNb responses in the more physiological lung epithelial A549 cells, which support intracellular Legionella replication (Vinzing et al., 2008). The transfection efficiency in these cells was approximately 60% in control experiments (data not shown). In accordance with earlier results, we observed an induction of IFNb mRNA by L. pneumophila infection and by poly(dA-dT) stimulation (Fig. 3A and B). In these cells, however, hZBP1 or hZBP1DZa overexpression did not amplify the bacteriaor DNA-mediated IFNb responses. Moreover, hZBP1 or hZBP1DZa overexpression had no effect on IL-8 production stimulated by L. pneumophila infection or poly(dA-dT) treatment (Fig. 3C and D). Different effects of ZBP1 siRNAs on IFNb expression in human and mouse cells Next, we performed a comprehensive analysis using RNAi technology in order to further determine if hZBP1 or hZBP1DZa have a role in IFNb responses to intracellular bacteria or poly(dA-dT) stimulation. First, three different siRNAs targeting exons 2, 5 or 6 of hZBP1 suppressed endogenous total hZBP1 mRNA expression in A549 cells

© 2008 The Authors Journal compilation © 2008 Blackwell Publishing Ltd, Cellular Microbiology, 10, 2579–2588

2582 J. Lippmann et al.

Fig. 2. hZBP1 overexpression slightly enhanced poly(dA-dT)-stimulated IFNb reporter activation. A. HEK293 cells were transfected with a control vector, 100 ng of hZBP1 or hZBP1DZa, respectively, together with 50 ng of the IFNb reporter plasmid and with or without 50 ng of poly(dA-dT) or poly(dG-dC) as indicated. Twenty-four hours post transfection, reporter activation was determined by luciferase assays. Luciferase values were normalized to protein concentrations and were compared with control-transfected cells that were set as 1. Assays were run in duplicates, and one representative experiment out of three independent sets of experiments is shown. B. HEK293 cells stably expressing either doxycycline-inducible eGFP, hZBP1 or hZBP1DZa were treated with doxycycline as indicated, and cell lysates were examined by immunoblotting using hZBP1-specific (RG7D12) or actin-specific antibodies. C. HEK293 cells stably expressing doxycycline-inducible eGFP, hZBP1 or hZBP1DZa were transfected with 50 ng of an IFNb reporter plasmid together with or without 50 ng poly(dA-dT) as indicated and were treated with doxycycline. Twenty-four hours post transfection, relative reporter activation was determined. Assays were run in duplicates, and one representative experiment out of three independent sets of experiments is shown.

Fig. 3. Overexpression of hZBP1 or hZBP1DZa had no effect on expression of IFNb or IL-8 induced by L. pneumophila infection or poly(dA-dT) stimulation. A549 cells were transfected with either a control vector, hZBP1 or hZBP1DZa or were left untreated, and after 24 h infected with L. pneumophila (L.p.) at an moi of 10 for 16 h (A and C) or were transfected with poly(dA-dT) for 8 h (B and D). Relative IFNb mRNA expression was assessed by Q-PCR (A and B) or supernatants were subjected to IL-8 ELISA (C and D). Data obtained in (A) and (B) represent the mean ⫾ SD of three independent experiments performed in duplicates with quantification of IFNb mRNA normalized to GAPDH mRNA and compared with untreated cells. Data obtained in (C) and (D) represent the mean ⫾ SEM of three independent experiments. n.s., non significant.

© 2008 The Authors Journal compilation © 2008 Blackwell Publishing Ltd, Cellular Microbiology, 10, 2579–2588

Human ZBP1 and IFNb responses 2583 by approximately 60–90% (Fig. 4A). Two out of three siRNAs (siZBP1b and c) led to a strong suppression of both major hZBP1 transcripts induced by poly(dA-dT) stimulation, and of the transiently overexpressed proteins (data not shown and Fig. 4B). An additional siRNA (siZBP1a) which targets exon 2 of ZBP1 was able to inhibit full-length hZBP1 but not hZBP1DZa (hZBP1 lacking exon 2), as expected. Our data show that knockdown of both hZBP1 transcription variants or of the fulllength hZBP1 only did not impair IFNb expression induced by Legionella infection or cytosolic DNA stimulation (Fig. 4D and F). In contrast, IRF3 siRNA significantly suppressed the IFNb responses as shown before (Opitz et al., 2006). All siRNAs targeting hZBP1 or IRF3 did not inhibit IL-8 production stimulated by L. pneumophila infection (Fig. 4E) or poly(dA-dT) stimulation (Fig. 4G). In addition, we used a different human cell line and L. monocytogenes as a prototype cytosolic bacterium. Again, different hZBP1 siRNAs efficiently suppressed their corresponding mRNA (data not shown). hZBP1 siRNAs had no effect on IFNb expression induced by L. monocytogenes infection or poly(dA-dT) transfection (data not shown), whereas IRF3 siRNA strongly abrogated the IFNb responses. Finally, we performed RNAi experiments in mouse L929 fibroblasts which have been used in the recently published study on mZBP1 (Takaoka et al., 2007). In agreement with this report, we found that siRNA-mediated gene silencing of mZBP1 reduced the poly(dA-dT)-induced IFNb expression after poly(dA-dT) transfection (Fig. 5). Significant reduction (approximately 50%) of the poly(dA-dT)-induced IFNb expression was observed only after 9 h while no significant effect was observed after 6 h. Overall, our data indicate that hZBP1 is not essential for mediating IFNb responses to intracellular bacteria or poly(dA-dT) stimulation in two human cell lines, whereas mZBP1 might be involved in poly(dA-dT)-induced IFNb expression in mouse L929 fibroblasts at a later time point. Discussion In this study we have examined the role of human ZBP1 and its major transcription variant in IFNb responses to two different intracellular bacterial infections and to cytosolic DNA stimulation. Our data demonstrate that Legionella infection and poly(dA-dT) stimulation induced IFNb, hZBP1 and hZBP1DZa expression in different human cells. Our overexpression data obtained with HEK293 cells suggest that hZBP1 might play a regulatory role in cytosolic DNA-triggered IFNb induction under certain conditions. Nonetheless, the gain-of-function and knock-down experiments in A549 and THP1 cells, which might better reflect the complex sensing and signalling mechanisms in innate immune cells, argue against an

essential role of hZBP1 in IFNb responses to intracellular bacteria or cytosolic DNA. In contrast to hZBP1 gene silencing, knock-down of mZBP1 in mouse fibroblasts reduced (a late) poly(dA-dT)-induced IFNb response in agreement with a recent study (Takaoka et al., 2007). While in this recent study by Takaoka et al. (2007) mZBP1 siRNA strongly inhibited poly(dA-dT)-triggered type I IFN expression in mouse cells, multiple siRNAs targeting human ZBP1 did not inhibit IFNb responses to poly(dA-dT) stimulation (or bacterial infection) in different human cells. While these negative results could theoretically be attributed to poorly working siRNAs or insufficient transfections, this appears less likely for several reasons. First, A549 and THP1 cells have often been successfully used for several RNAi studies by us (Opitz et al., 2006; 2007; Vinzing et al., 2008) and others (Seshadri et al., 2007). Second, different siRNAs strongly and differentially suppressed hZBP1 and hZBP1DZa mRNAs and proteins. Third, IRF3 siRNA clearly inhibited the IFNb expression induced by the bacteria or cytosolic DNA. The discrepancy between the RNAi data obtained in murine cells and human cells might reflect differences in the sensing mechanism or signalling leading to IFNb induction between humans and mice. This assumption might be supported by the observation that murine cells appeared to respond also to poly(dG-dC) (Ishii et al., 2006; Takaoka et al., 2007), whereas the human cells tested in our study showed almost no response to poly(dG-dC). Moreover, by using several different siRNAs targeting IPS-1/MAVS and the HCV NS3/4A protease which cleaves IPS-1/MAVS, we and others indicated a role of IPS-1/MAVS in mediating IFNb production activated by cytosolic DNA or intracellular bacteria in human cells (Ishii et al., 2006; Opitz et al., 2006; Cheng et al., 2007). In sharp contrast, IPS1/MAVS was clearly dispensable for IFNb responses to Listeria infection or cytosolic DNA stimulation in mice (Kumar et al., 2006; Soulat et al., 2006; Sun et al., 2006). On the other hand, Taniguchi and co-workers recently showed that mZBP1 siRNA abrogated poly(dA-dT)induced IFNb expression in mouse L929 fibroblasts but not in mouse embryonic fibroblasts (MEFs) (Wang et al., 2008). Moreover, a recent study indicated that ZBP1 knockout MEFs, dendritic cells and macrophages responded normally to poly(dA-dT) (Ishii et al., 2008). Thus, in addition or alternative to potential inter-species differences, the function of ZBP1 might also be cell typespecific or dependent on the functional status of the host cell. Different synthetic dsDNAs have been shown to vary greatly in their potential to stimulate an IFN response (Ishii et al., 2006). In this study poly(dA-dT) had by far the strongest effect on IFNb and IFNa expression followed by poly(dG-dC), which led to almost no IFNb induction. Five further synthetic dsDNAs had even weaker or no effect at

© 2008 The Authors Journal compilation © 2008 Blackwell Publishing Ltd, Cellular Microbiology, 10, 2579–2588

2584 J. Lippmann et al.

Fig. 4. Different hZBP1 siRNAs did not inhibit IFNb or IL-8 expression induced by L. pneumophila or cytosolic DNA stimulation in human lung epithelial cells. A. A549 cells were transfected with control siRNA (siC), siRNA targeting hZBP1 exon 2 (siZBP1a), exon 5 (siZBP1b) or exon 6 (siZBP1c) or were left untreated. After 2 days, cells were infected with L. pneumophila (moi 10) for 16 h to induce endogenous ZBP1 mRNA. Relative expression of hZBP1 was assessed by Q-PCR. Data obtained represent the mean ⫾ SD of one representative experiment out of three performed in duplicates with quantification of ZBP1 mRNA normalized to GAPDH mRNA and compared with untreated cells. B. A549 cells were transfected with siRNAs and hZBP1 or hZBPDZa expression vectors. After 24 h, expression of hZBP1 and hZBPDZa protein was examined by Western blot with the specific hZBP1 antibody RG7D12. Expression of actin was assessed as a loading control. C. A549 cells were transfected with control siRNA (siC) or IRF3 siRNA or were left untreated. After further 56 h, expression of IRF3 was assessed by RT-PCR. D–G. A549 cells were transfected with siRNAs as indicated, and infected with L. pneumophila (moi 10) for 16 h (D and E), or transfected with poly(dA-dT) and incubated for 8 h (F and G). Relative IFNb mRNA expression was analysed by Q-PCR (D and F), or IL-8 secretion was determined by ELISA (E and G). Data obtained in (D) and (F) represent the mean ⫾ SD of three independent experiments performed in duplicates with quantification of IFNb mRNA normalized to GAPDH mRNA and compared with untreated cells. Data obtained in (E) and (G) represent the mean ⫾ SEM of three independent experiments. *P < 0.05, ***P < 0.001; n.s., non significant. © 2008 The Authors Journal compilation © 2008 Blackwell Publishing Ltd, Cellular Microbiology, 10, 2579–2588

Human ZBP1 and IFNb responses 2585

Fig. 5. Knock-down of mZBP1 impaired IFNb expression induced by cytosolic DNA stimulation in murine fibroblasts. L929 murine fibroblasts were transfected with control siRNA (siC) or siRNA targeting mZBP1 exon 5 (siZBP1). After 24 h cells were transfected with poly(dA-dT) for the indicated time intervals or were left untreated. Expression of mZBP1 and mGAPDH was analysed by RT-PCR. Relative IFNb mRNA expression was analysed by Q-PCR with quantification of IFNb mRNA normalized to mGAPDH mRNA and compared with untreated cells. Results shown represent the mean ⫾ SD of three independent experiments performed in duplicates. ***P < 0.001; n.s., non significant.

all (Ishii et al., 2006). This clearly indicates that recognition of dsDNA is sequence- or structure-specific and that viewing poly(dA-dT) as the prototype for dsDNA is problematic. Indeed, poly(dA-dT) has been shown to adopt many different DNA conformations, including B-DNA, Z-DNA and cruciform DNA (Rich et al., 1984; Haniford and Pulleyblank, 1985; Ridoux et al., 1988). Because the DNA conformation is greatly dependent on the environment, it is very hard to determine which of these conformations is actually adopted within the cell and is responsible for the strong induction of the IFN response. Among other factors, proteins can directly influence DNA conformations. This is illustrated by the conversion of poly(dG-dC), which adopts a B-DNA conformation under low-salt conditions, into Z-DNA upon addition of proteins containing Za domains (Schwartz et al., 1999; Ha et al., 2004; Deigendesch et al., 2006). Moreover, the Za domain of ZBP1-related protein ADAR1 has been shown to stabilize (dT-dA)3 in the Z-conformation (Herbert et al., 1998). Comparison of ZBP1 from different mammalian species showed the existence of four conserved domains (CD), displaying sequence identities of 67%, 65%, 71% and 61%, respectively, between human and mouse ZBP1 (Rothenburg et al., 2002; N. Deigendesch, unpubl. obs.).

CD1 and CD2 comprise the two Za domains. The C-terminal part of ZBP1, which comprises CD3 and CD4, shows no homology with any known proteins and has been shown to be crucial for subcellular localization (Deigendesch et al., 2006) and activation of the IFN response (Takaoka et al., 2007) in murine cells. Using immunoprecipitation with poly(dA-dT), CD3 has been identified as a putative third DNA-binding domain, pending verification with more specific methods (Takaoka et al., 2007). In addition to ZBP1, two other Z-DNA-binding proteins have been characterized in vertebrates, which are induced after immunostimulation and contain two Za domains in the N-terminus: ADAR1, which is found in vertebrates (Herbert et al., 1997), and Z-DNA-binding protein kinase (PKZ) (Rothenburg et al., 2005), which likely evolved after a duplication of the antiviral doublestranded RNA-activated protein kinase PKR gene early in the evolution of bony fishes (Rothenburg et al., 2008). Like ZBP1, both the IFN-induced Za domain-containing isoform of ADAR1 (Yang et al., 2003) and PKZ (Bergan et al., 2008; N. Deigendesch, unpubl. obs.) display a cytoplasmic localization. The forth ZBP is the poxvirus virulence factor E3L, which contains one N-terminal Za domain and a C-terminal double-stranded RNA-binding domain. Both domains are essential for complete Vaccinia virus (VV) pathogenesis in a mouse model (Brandt and Jacobs, 2001). Substitution of the Za domain of E3L with that of ZBP1 and ADAR1 restored pathogenicity (Kim et al., 2003). Moreover, single-amino-acid substitutions that abolished Z-DNA binding in vitro severely impaired VV virulence (Kim et al., 2003). This indicates that E3L might act as an antagonist of cellular ZBPs by blocking their functions in the innate immune response as shown for ADAR1 (Liu et al., 2001) and as proposed for ZBP1 (Rothenburg et al., 2002). In agreement with important regulatory roles of E3L and ADAR1 in innate immune responses, E3L and ADAR1 overexpression has been recently shown to inhibit IFNb responses to poly(dA-dT), whereas ADAR1 knockout cells displayed an enhanced IFNb expression after cytosolic DNA stimulation (Wang et al., 2008). Overall, while ZBP1 and other Z-DNA-binding proteins probably play important roles in the innate immunity, further work is warranted to elucidate their exact functions in different infection models and in the response to cytosolic DNA stimulation.

Experimental procedures Bacterial strains The L. pneumophila wild-type serogroup 1 strain Corby was grown on buffered charcoal-yeast extract (BCYE) agar for 2 days

© 2008 The Authors Journal compilation © 2008 Blackwell Publishing Ltd, Cellular Microbiology, 10, 2579–2588

2586 J. Lippmann et al. at 37°C before use. The L. monocytogenes serotype 1/2a strain EGD was grown in brain heart infusion broth (Difco).

RNA interference Control non-silencing siRNA (sense UUCUCCGAACGUGU CACGUtt, antisense ACGUGACACGUUCGGAGAAtt), siRNA targeting exon 2 of hZBP1 (ZBP1a: sense CCAAGUCCUCUAC CGAAUGtt, antisense CAUUCGGUAGAGGACUUGGtt), hZBP1 exon 5 siRNA (ZBP1b: sense GCAAAGUCAGCCUCAAUUAtt; antisense UAAUUGAGGCUGACUUUGCtc), hZBP1 exon 6 siRNA (ZBP1c: sense GGUGAUUCCUCAACUUGGGtt; antisense CCCAAGUUGAGGAAUCACCtg), IRF3 siRNA (sense GGAGGAUUUCGGAAUCUUCtt; antisense GAAGAUUCCGAA AUCCUCCtg) and siRNA targeting exon 5 of mZBP1 (sense: GGAGCUCAGUACAUCUACAtt; antisense: UGUAGAUGUACU GAGCUCCgt) were from Ambion. A549 cells (DSMZ) and L929 cells were transfected by using Amaxa NucleofectorTM (Amaxa) according to the manufacturer’s protocol (Nucleofector™ Solution V, Nucleofector™ program G-16 and T-20) with 2 mg of siRNA per 106 cells.

Expression plasmids and cDNA transfection Construction of hZBP1 expression plasmids pcDNA3.1hZBP1 full and pcDNA3.1hZBP1DZa has been described (Deigendesch et al., 2006). For generating stably transfected inducible cell lines expressing hZBP1 or hZBP1DZa, Flp-In T-REx 293 cells were co-transfected with pCDNA5/FRT/TO containing inserts of hZBP1 full length and hZBP1DZa, and pOG44, a Flp recombinase expression vector, using polyethylenimine (jetPEI, Qbiogene). Stably transfected cells were selected and maintained with medium containing hygromycin B (200 mg ml-1) and blasticidin (15 mg ml-1) (Invitrogen). To induce expression of the gene of interest the medium was removed and fresh medium containing 1 mg ml-1 doxycycline (BD Biosciences) was added. Expression plasmids for MAVS and the IFNb luciferase reporter p125-luc were kindly provided by T. Wolff, Robert-Koch Institute Berlin, and by D. Golenbock, University of Massachusetts Medical School, MA. Reporter assays in HEK293 cells or HEK293 cells stably expressing hZBP1 variants were conducted by transiently co-transfecting HEK293 cells in 24-well plates with 50 ng of IFNb luciferase reporter plasmids with or without the indicated hZBP1 expression plasmids by the calcium phosphate method (Clontech). Luciferase activity was measured by using the luciferase reporter gene assay (Promega), and results were normalized to corresponding protein concentrations. A549 cells were transfected with the indicated expression vectors by using Amaxa Nucleofector™ (Amaxa) according to the manufacturer’s protocol (Nucleofector™ Solution V, Nucleofector™ program G-16).

(Invitrogen) in concentrations as indicated. Poly(dA-dT) and poly(dG-dC) were introduced into HEK293 cells by calcium phosphate transfection together with the plasmid DNA. In some experiments, cells were treated with IFNb (InvivoGen) as indicated.

RT-PCR analysis Total RNA from A549 or L929 cells was isolated with the RNeasy Mini kit (Qiagen) and reverse transcribed using MMLV reverse transcriptase (Invitrogen). The generated cDNA was amplified by semi-quantitative RT-PCR using specific primers (IFNb-sense 5′-GCTCTCCTGTTGTGCTTCTCCAC-3′; IFNb-antisense 5′-CA ATAGTCTCATTCCAGCCAGTGC-3′; hZBP1/ZBP1DZa-sense 5′-TCCGACTCCTTGCAGCTGCTGTC; hZBP1-antisense 5′-GG GCGGTAAATCGTCCATGCTTTGGAC-3′; hZBP1DZa-antisense 5′-ACAAGGCCAGCTCTGCAGGACC-3′; IRF3-sense 5′-TACG TGAGGCATGTGCTGA-3′; IRF3-antisense 5′-AGTGGGTGGC TGTTGGAAAT-3′; GAPDH-sense 5′-CCACCCATGGCAAATT CCATGGCA-3′; GAPDH-antisense 5′-TCTAGACGGCAGGTC AGGTCCACC-3′; mZBP1-sense 5′-AGGTCCAAGCAGCCAT TCTT-3′; mZBP1-antisense 5′-TCACCACAGGCTTTCTCT CTT-3′; mGAPDH-sense 5′-TGATGGGTGTGAACCACGAG-3′; mGAPDH-antisense 5′-TCAGTGTAGCCCAAGATGCC-3′).

Quantitative RT-PCR cDNAs obtained were subjected to real-time PCR on an ABI 7300 instrument (Applied Biosystems) using ZBP1, IFNb and GAPDH TaqMan Gene expression assays (Applied Biosystems). Input was normalized by the average expression of GAPDH. All PCR reactions were carried out in duplicates; relative IFNb expression in untreated cells was set as 1.

Western blot Cytoplasmatic extracts of A549 cells were separated by SDSPAGE, and blotted. Membranes were probed with rat monoclonal antibody RG7D12, which recognizes the Zb domain of hZBP1, encoded by exon 4 (N. Deigendesch and S. Rothenburg, unpubl. results) and antibodies specific for actin (Santa Cruz), and subsequently incubated with secondary antibodies (Cy5.5-labelled antirat, or IRDye 800-labelled anti-goat respectively). Proteins were detected by using an Odyssey infrared imaging system (LI-COR).

Il-8 ELISA IL-8 concentrations in supernatants were quantified using commercially available sandwich ELISA Kit (R&D systems).

Acknowledgements Infection/stimulation A549 cells were infected with L. pneumophila at different multiplicities of infection (moi) and incubated for additional time as indicated at 37°C. Poly(dA-dT):poly(dT-dA) [poly(dA-dT)] and poly(dG-dC):poly(dC-dG) [poly(dG-dC)] (GE Healthcare) were transfected into A549 or L929 cells using Lipofectamine 2000

We are grateful to A. Kühn, F. Schreiber, J. Hellwig and D. Stoll for excellent technical assistance and D. Freyer for generously providing L929 cells. This work was supported in part by the International Max Planck Research School for Infectious Diseases and Immunology to J.L., and by grants given by the Deutsche Forschungsgemeinschaft, the Deutsche Gesellschaft

© 2008 The Authors Journal compilation © 2008 Blackwell Publishing Ltd, Cellular Microbiology, 10, 2579–2588

Human ZBP1 and IFNb responses 2587 für Pneumologie und Beatmungsmedizin and the Jürgen Manchot Stiftung to B.O. Parts of this work will be included in the PhD thesis of J.L.

References Akira, S., Uematsu, S., and Takeuchi, O. (2006) Pathogen recognition and innate immunity. Cell 124: 783–801. Andrejeva, J., Childs, K.S., Young, D.F., Carlos, T.S., Stock, N., Goodbourn, S., and Randall, R.E. (2004) The V proteins of paramyxoviruses bind the IFN-inducible RNA helicase, mda-5, and inhibit its activation of the IFN-beta promoter. Proc Natl Acad Sci USA 101: 17264– 17269. Auerbuch, V., Brockstedt, D.G., Meyer-Morse, N., O’Riordan, M., and Portnoy, D.A. (2004) Mice lacking the type I interferon receptor are resistant to Listeria monocytogenes. J Exp Med 200: 527–533. Bergan, V., Jagus, R., Lauksund, S., Kileng, O., and Robertsen, B. (2008) The Atlantic salmon Z-DNA binding protein kinase phosphorylates translation initiation factor 2 alpha and constitutes a unique orthologue to the mammalian dsRNA-activated protein kinase R. FEBS J 275: 184–197. Brandt, T.A., and Jacobs, B.L. (2001) Both carboxy- and amino-terminal domains of the vaccinia virus interferon resistance gene, E3L, are required for pathogenesis in a mouse model. J Virol 75: 850–856. Carrero, J.A., Calderon, B., and Unanue, E.R. (2004) Type I interferon sensitizes lymphocytes to apoptosis and reduces resistance to Listeria infection. J Exp Med 200: 535–540. Cheng, G., Zhong, J., Chung, J., and Chisari, F.V. (2007) Double-stranded DNA and double-stranded RNA induce a common antiviral signaling pathway in human cells. Proc Natl Acad Sci USA 104: 9035–9040. Coers, J., Vance, R.E., Fontana, M.F., and Dietrich, W.F. (2007) Restriction of Legionella pneumophila growth in macrophages requires the concerted action of cytokine and Naip5/Ipaf signalling pathways. Cell Microbiol 9: 2344–2357. Deigendesch, N., Koch-Nolte, F., and Rothenburg, S. (2006) ZBP1 subcellular localization and association with stress granules is controlled by its Z-DNA binding domains. Nucleic Acids Res 34: 5007–5020. Fritz, J.H., Ferrero, R.L., Philpott, D.J., and Girardin, S.E. (2006) Nod-like proteins in immunity, inflammation and disease. Nat Immunol 7: 1250–1257. Fu, Y., Comella, N., Tognazzi, K., Brown, L.F., Dvorak, H.F., and Kocher, O. (1999) Cloning of DLM-1, a novel gene that is up-regulated in activated macrophages, using RNA differential display. Gene 240: 157–163. Ha, S.C., Lokanath, N.K., Van Quyen, D., Wu, C.A., Lowenhaupt, K., Rich, A., et al. (2004) A poxvirus protein forms a complex with left-handed Z-DNA: crystal structure of a Yatapoxvirus Zalpha bound to DNA. Proc Natl Acad Sci USA 101: 14367–14372. Haniford, D.B., and Pulleyblank, D.E. (1985) Transition of a cloned d(AT)n-d(AT)n tract to a cruciform in vivo. Nucleic Acids Res 13: 4343–4363. Herbert, A., Alfken, J., Kim, Y.G., Mian, I.S., Nishikura, K., and Rich, A. (1997) A Z-DNA binding domain present in the

human editing enzyme, double-stranded RNA adenosine deaminase. Proc Natl Acad Sci USA 94: 8421–8426. Herbert, A., Schade, M., Lowenhaupt, K., Alfken, J., Schwartz, T., Shlyakhtenko, L.S., et al. (1998) The Zalpha domain from human ADAR1 binds to the Z-DNA conformer of many different sequences. Nucleic Acids Res 26: 3486–3493. Ishii, K.J., Coban, C., Kato, H., Takahashi, K., Torii, Y., Takeshita, F., et al. (2006) A Toll-like receptor-independent antiviral response induced by double-stranded B-form DNA. Nat Immunol 7: 40–48. Ishii, K.J., Kawagoe, T., Koyama, S., Matsui, K., Kumar, H., Kawai, T., et al. (2008) TANK-binding kinase-1 delineates innate and adaptive immune responses to DNA vaccines. Nature 451: 725–729. Katze, M.G., He, Y., and Gale, M., Jr (2002) Viruses and interferon: a fight for supremacy. Nat Rev Immunol 2: 675– 687. Kawai, T., Takahashi, K., Sato, S., Coban, C., Kumar, H., Kato, H., et al. (2005) IPS-1, an adaptor triggering RIG-Iand Mda5-mediated type I interferon induction. Nat Immunol 6: 981–988. Kim, Y.G., Muralinath, M., Brandt, T., Pearcy, M., Hauns, K., Lowenhaupt, K., et al. (2003) A role for Z-DNA binding in vaccinia virus pathogenesis. Proc Natl Acad Sci USA 100: 6974–6979. Kumar, H., Kawai, T., Kato, H., Sato, S., Takahashi, K., Coban, C., et al. (2006) Essential role of IPS-1 in innate immune responses against RNA viruses. J Exp Med 203: 1795– 1803. Leber, J.H., Crimmins, G.T., Raghavan, S., Meyer-Morse, N.P., Cox, J.S., and Portnoy, D.A. (2008) Distinct TLR- and NLR-mediated transcriptional responses to an intracellular pathogen. PLoS Pathog 4: e6. Levy, D.E., and Darnell, J.E., Jr (2002) Stats: transcriptional control and biological impact. Nat Rev Mol Cell Biol 3: 651–662. Liu, Y., Wolff, K.C., Jacobs, B.L., and Samuel, C.E. (2001) Vaccinia virus E3L interferon resistance protein inhibits the interferon-induced adenosine deaminase A-to-I editing activity. Virology 289: 378–387. McCaffrey, R.L., Fawcett, P., O’Riordan, M., Lee, K.D., Havell, E.A., Brown, P.O., and Portnoy, D.A. (2004) A specific gene expression program triggered by Grampositive bacteria in the cytosol. Proc Natl Acad Sci USA 101: 11386–11391. Mancuso, G., Midiri, A., Biondo, C., Beninati, C., Zummo, S., Galbo, R., et al. (2007) Type I IFN signaling is crucial for host resistance against different species of pathogenic bacteria. J Immunol 178: 3126–3133. Meylan, E., Curran, J., Hofmann, K., Moradpour, D., Binder, M., Bartenschlager, R., and Tschopp, J. (2005) Cardif is an adaptor protein in the RIG-I antiviral pathway and is targeted by hepatitis C virus. Nature 437: 1167–1172. Moore, C.B., Bergstralh, D.T., Duncan, J.A., Lei, Y., Morrison, T.E., Zimmermann, A.G., et al. (2008) NLRX1 is a regulator of mitochondrial antiviral immunity. Nature 451: 573–577. O’Connell, R.M., Saha, S.K., Vaidya, S.A., Bruhn, K.W., Miranda, G.A., Zarnegar, B., et al. (2004) Type I interferon production enhances susceptibility to Listeria monocytogenes infection. J Exp Med 200: 437–445.

© 2008 The Authors Journal compilation © 2008 Blackwell Publishing Ltd, Cellular Microbiology, 10, 2579–2588

2588 J. Lippmann et al. O’Connell, R.M., Vaidya, S.A., Perry, A.K., Saha, S.K., Dempsey, P.W., and Cheng, G. (2005) Immune activation of type I IFNs by Listeria monocytogenes occurs independently of TLR4, TLR2, and receptor interacting protein 2 but involves TNFR-associated NF kappa B kinase-binding kinase 1. J Immunol 174: 1602–1607. O’Riordan, M., Yi, C.H., Gonzales, R., Lee, K.D., and Portnoy, D.A. (2002) Innate recognition of bacteria by a macrophage cytosolic surveillance pathway. Proc Natl Acad Sci USA 99: 13861–13866. Okabe, Y., Kawane, K., Akira, S., Taniguchi, T., and Nagata, S. (2005) Toll-like receptor-independent gene induction program activated by mammalian DNA escaped from apoptotic DNA degradation. J Exp Med 202: 1333–1339. Opitz, B., van Vinzing, M., Laak, V., Schmeck, B., Heine, G., Gunther, S., et al. (2006) Legionella pneumophila induces IFNbeta in lung epithelial cells via IPS-1 and IRF3, which also control bacterial replication. J Biol Chem 281: 36173– 36179. Opitz, B., Rejaibi, A., Dauber, B., Eckhard, J., Vinzing, M., Schmeck, B., et al. (2007) IFNbeta induction by influenza A virus is mediated by RIG-I which is regulated by the viral NS1 protein. Cell Microbiol 9: 930–938. Rich, A., Nordheim, A., and Wang, A.H. (1984) The chemistry and biology of left-handed Z-DNA. Annu Rev Biochem 53: 791–846. Ridoux, J.P., Liquier, J., and Taillandier, E. (1988) Raman spectroscopy of Z-form poly[d(A-T)]·poly[d(A-T)]. Biochemistry 27: 3874–3878. Rothenburg, S., Schwartz, T., Koch-Nolte, F., and Haag, F. (2002) Complex regulation of the human gene for the Z-DNA binding protein DLM-1. Nucleic Acids Res 30: 993– 1000. Rothenburg, S., Deigendesch, N., Dittmar, K., Koch-Nolte, F., Haag, F., Lowenhaupt, K., and Rich, A. (2005) A PKR-like eukaryotic initiation factor 2alpha kinase from zebrafish contains Z-DNA binding domains instead of dsRNA binding domains. Proc Natl Acad Sci USA 102: 1602–1607. Rothenburg, S., Deigendesch, N., Dey, M., Dever, T.E., and Tazi, L. (2008) Double-stranded RNA-activated protein kinase PKR of fishes and amphibians: varying number of double-stranded RNA binding domains and lineage specific duplications. BMC Biol 6: 12. Schwartz, T., Lowenhaupt, K., Kim, Y.G., Li, L., Brown, B.A., Herbert, A., and Rich, A. (1999) Proteolytic dissection of Zab, the Z-DNA-binding domain of human ADAR1. J Biol Chem 274: 2899–2906.

Seshadri, S., Duncan, M.D., Hart, J.M., Gavrilin, M.A., and Wewers, M.D. (2007) Pyrin levels in human monocytes and monocyte-derived macrophages regulate IL-1beta processing and release. J Immunol 179: 1274–1281. Soulat, D., Bauch, A., Stockinger, S., Superti-Furga, G., and Decker, T. (2006) Cytoplasmic Listeria monocytogenes stimulates IFN-beta synthesis without requiring the adapter protein MAVS. FEBS Lett 580: 2341–2346. Stanley, S.A., Johndrow, J.E., Manzanillo, P., and Cox, J.S. (2007) The Type I IFN response to infection with Mycobacterium tuberculosis requires ESX-1-mediated secretion and contributes to pathogenesis. J Immunol 178: 3143– 3152. Stetson, D.B., and Medzhitov, R. (2006) Recognition of cytosolic DNA activates an IRF3-dependent innate immune response. Immunity 24: 93–103. Stockinger, S., Materna, T., Stoiber, D., Bayr, L., Steinborn, R., Kolbe, T., et al. (2002) Production of type I IFN sensitizes macrophages to cell death induced by Listeria monocytogenes. J Immunol 169: 6522–6529. Sun, Q., Sun, L., Liu, H.H., Chen, X., Seth, R.B., Forman, J., and Chen, Z.J. (2006) The specific and essential role of MAVS in antiviral innate immune responses. Immunity 24: 633–642. Takaoka, A., Wang, Z., Choi, M.K., Yanai, H., Negishi, H., Ban, T., et al. (2007) DAI (DLM-1/ZBP1) is a cytosolic DNA sensor and an activator of innate immune response. Nature 448: 501–505. Vinzing, M., Eitel, J., Lippmann, J., Hocke, A.C., Zahlten, J., Slevogt, H., et al. (2008) NAIP and Ipaf control Legionella pneumophila replication in human cells. J Immunol 180: 6808–6815. Wang, Z., Choi, M.K., Ban, T., Yanai, H., Negishi, H., Lu, Y., et al. (2008) Regulation of innate immune responses by DAI (DLM-1/ZBP1) and other DNA-sensing molecules. Proc Natl Acad Sci USA 105: 5477–5482. Yang, J.H., Nie, Y., Zhao, Q., Su, Y., Pypaert, M., Su, H., and Rabinovici, R. (2003) Intracellular localization of differentially regulated RNA-specific adenosine deaminase isoforms in inflammation. J Biol Chem 278: 45833–45842. Yoneyama, M., and Fujita, T. (2007) RIG-I family RNA helicases: cytoplasmic sensor for antiviral innate immunity. Cytokine Growth Factor Rev 18: 545–551. Yoneyama, M., Kikuchi, M., Natsukawa, T., Shinobu, N., Imaizumi, T., Miyagishi, M., et al. (2004) The RNA helicase RIG-I has an essential function in double-stranded RNA-induced innate antiviral responses. Nat Immunol 5: 730–737.

© 2008 The Authors Journal compilation © 2008 Blackwell Publishing Ltd, Cellular Microbiology, 10, 2579–2588