View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Biochimica et Biophysica Acta 1813 (2011) 575–583

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta

journal homepage: www.elsevier.com/locate/bbamcr

Review Mitochondrial localization of viral proteins as a means to subvert host defense☆

Céline Castanier, Damien Arnoult ⁎

INSERM U1014, Hôpital Paul Brousse, Bâtiment Lavoisier, 14 avenue Paul Vaillant Couturier, 94807 Villejuif cedex, France Université Paris-Sud

article info abstract

Article history: Viruses have developed a battery of distinct strategies to overcome the very sophisticated defense Received 29 June 2010 mechanisms of the infected host. Throughout the process of pathogen–host co-evolution, viruses have Received in revised form 18 August 2010 therefore acquired the capability to prevent host cell apoptosis because elimination of infected cells via Accepted 20 August 2010 apoptosis is one of the most ancestral defense mechanism against infection. Conversely, induction of Available online 31 August 2010 apoptosis may favor viral dissemination as a result of the dismantlement of the infected cells. Mitochondria have been long recognized for their key role in the modulation of apoptosis but more recently, mitochondria Keywords: Virus have been shown to serve as a crucial platform for innate immune signaling as illustrated by the Mitochondria identification of MAVS. Thus, it is therefore not surprising that this organelle represents a recurrent target for Apoptosis viruses, aiming to manipulate the fate of the infected host cell or to inhibit innate immune response. In this Innate immunity review, we highlight the viral proteins that are specifically targeted to the mitochondria to subvert host defense. This article is part of a Special Issue entitled Mitochondria: the deadly organelle. © 2010 Elsevier B.V. All rights reserved.

Detection of viruses by host cells relies on several families of pattern- pathogen encountered, the nature of the infected cell, and probably the recognition receptors (PRRs), and these sensors are responsible for the intensity of the infection, the cellular responses are tipped either triggering of innate immune responses. PRRs recognize structures towards cell death or cell survival [5]. Indeed, the final fate of an infected conserved among microbial species, which are called pathogen- cell is commonly the result of the opposing effects of pro-survival and associated molecular patterns (PAMPs) [1,2]. PRRs are also responsible pro-death pathways, and in a number of cases, depends on the opposite for recognizing endogenous molecules released from damaged cells, actions of the virus and the infected cell. Mitochondria represent a the damage (or danger) associated molecular patterns (DAMPs). At the critical organelle implicated in the induction of programmed cell death cellular level, three main families of PRRs have been characterized in the (PCD) and, consequently, play a central role in cellular host–virus past 15 years: the transmembrane Toll-like receptors (TLRs), the interactions. Therefore, it is not surprising that so many viral strategies cytosolic Nod-like receptors (NLRs) and the RIG-I-like receptors are aiming at targeting mitochondria to subvert PCD. (RLRs) [1,2]. In addition, a heterogenous class of cytosolic sensors of In addition to the well-documented role of mitochondria in the double-stranded DNA is represented by the proteins DAI/DLM1/ZBP1 induction of PCD following infection, an emerging concept is that this and AIM2/HIN200, which trigger type I interferon and caspase-1 organelle represents a convergent point of a number of cellular innate inflammasome pathways, respectively [3]. Upon detection of the viruses immune responses. This has been remarkably illustrated recently by by these PRRs, infected cells trigger a large repertoire of defense the discovery of MAVS (IPS-1/Cardif/VISA), a critical adaptor protein pathways that, together, contribute to cellular innate immunity. These downstream of the RLR proteins RNA sensors RIG-I and MDA-5 [6–9]. responses include factors such as NF-κB, IRF (IFN-Regulatory Factor), MAVS is anchored to the mitochondrial outer membrane, and this JNK, p38 and ERK, which affect inflammatory and defense mediators at sub-cellular localization is essential for the function of the protein [8]. the transcriptional level. PRRs also trigger the transcription-indepen- These recent observations suggest that the role of mitochondria in the dent recruitment of the caspase-1 inflammasome, resulting in the regulation of cellular responses to viruses may span much beyond the maturation of the critical pro-inflammatory mediators IL-1β, IL-18 and modulation of PCD. This article reviews how some viral proteins IL-33 [4]. Another critical axis of the host response to infection is the addressed at the mitochondria subvert host defense either by modulation of survival/death pathways. Depending of the nature of the modulating PCD or by inhibiting innate immunity.

1. Mitochondrion, a key player in apoptosis ☆ This article is part of a Special Issue entitled Mitochondria: the deadly organelle. ⁎ Corresponding author. INSERM U1014, Hôpital Paul Brousse, Bâtiment Lavoisier, 14 Apoptosis, a form of PCD, is a cell suicide program essential for avenue Paul Vaillant Couturier, 94807 Villejuif cedex, France. Tel.: +33 1 45 59 60 38; fax: +33 1 45 59 53 43. development and for adult tissue homeostasis in all metazoan E-mail address: [email protected] (D. Arnoult). animals. The misregulation (inhibition or exacerbation) of apoptosis

0167-4889/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.bbamcr.2010.08.009 576 C. Castanier, D. Arnoult / Biochimica et Biophysica Acta 1813 (2011) 575–583

is involved in several pathogenic mechanisms, including neurode- [14]. Anti-apoptotic proteins such as Bcl-2 or Bcl-XL inhibit the generative diseases, cancers and AIDS [10]. The stereotypical death function of pro-apoptotic proteins such as Bax or Bak. Bcl-2 family throes of a cell undergoing apoptosis include DNA fragmentation, members contain conserved domains (the Bcl-2-homology domains nuclear condensation, cell shrinkage, blebbing and phosphatidylser- named BH1 to BH4), and these domains are required for the pro- or ine externalization; these features are closely associated with the anti-apoptotic function of the protein. An important subgroup of pro- activity of a family of cysteine proteases called caspases [11]. apoptotic Bcl-2 members is the ‘BH3-only’ proteins, so named because Caspases are a family of cysteine proteases involved in the they only have a BH3 domain. The ‘BH3-only’ proteins (Bik, Bid, Bim, transduction and execution of the apoptotic program [10]. Caspases Bad, Puma, NoxA, Hrk, Bmf) have a pro-apoptotic function by either are produced as pro-enzymes that must be proteolytically cleaved to activating the pro-apoptotic members (Bax and Bak) or by inhibiting produce active caspases. Caspases are classified in two groups: the the anti-apoptotic members (Bcl-2, Bcl-XL, A1, Mcl-1) [15]. The initiator caspases such as caspase-9 and caspase-8 and the effector mechanisms by which the pro-apoptotic Bcl-2 family members caspases such as caspase-3. induce MOMP remain controversial and not fully understood [16]. Mitochondria are implicated in the two major apoptotic pathways. One proposed model focuses on the rupture of the outer mitochon- The death receptor-mediated pathway (“extrinsic pathway”) involves drial membrane as a consequence of mitochondrial matrix swelling mitochondria mainly as an amplification loop, whereas cellular following the opening of the permeability transition pore (PTP), a deprival and stress-mediated apoptosis is regulated predominantly putative pore composed, at least, by the channel VDAC, the adenine at the mitochondrial level (“intrinsic pathway”). Mitochondria have a nucleotide translocator ANT and cyclophilin D [17]. According to pivotal role in apoptosis by releasing apoptogenic factors such as another model, pro-apoptotic Bcl-2 proteins such as Bax or Bak induce cytochrome c, Smac/DIABLO and Omi/HtrA2 [12] (Fig. 1). In the a selective process of MOMP through the formation of channels or cytosol, cytochrome c triggers caspase activation following the pores, enabling the selective release of proteins that are soluble in the formation of a ternary complex called the apoptosome that is IMS such as cytochrome c [18]. composed of the adaptor protein Apaf-1, caspase-9 and cytochrome Besides apoptosis, a number of other cell death pathways have c [11]. In the apoptosome, active initiator caspase-9 processes effector been described [19]. These other cell death pathways are autophagic caspases such as caspase-3. Next, active caspase-3 cleaves several cell cell death (for review see [20]), necroptosis (a form of regulated substrates to induce apoptosis and cell death [11]. Natural inhibitors necrosis) and PARP-1-mediated necrotic death [19]. Necrosis, unlike of caspases are found in cells, the X-linked inhibitor of apoptosis apoptosis, is considered as an unregulated cell death caused by proteins (XIAP) [13], and the function of Smac/DIABLO and Omi/ overwhelming stress that is incompatible with cell survival. Again, HtrA2 once released into the cytosol is to antagonize XIAP to promote mitochondria have been reported to play a role in necroptosis and caspase activation. Bcl-2 family members regulate mitochondrial necrosis through the production of ROS following disruption of the outer membrane permeabilization (MOMP), resulting in the release of association of ADP-ATP translocase (ANT) with cyclophilin D, or cytochrome c, Smac/DIABLO and Omi/HtrA2 and subsequent caspase through the release into the cytosol of caspase-independent cell death activation. The Bcl-2 family includes pro- and anti-apoptotic proteins effectors such as AIF [12].

Fig. 1. The central role of mitochondrion in apoptosis and modulation by viral proteins. Pro-apoptotic Bcl-2 members (Bax and Bak) promote the release into the cytosol of mitochondrial apoptogenic factors (cytochrome c, Smac/Diablo, Omi/HtrA2) involved in caspase activation. The pro-apoptotic function of Bax and Bak is antagonized by the anti- apoptotic Bcl-2 members (Bcl-2 and Bcl-XL). In blue, the viral Bcl-2 homologues (vBcl-2) with sequence homology with Bcl-2 that inhibit apoptosis of infected cells. In purple, the vBcl-2 s with structure homology with Bcl-2. Abbreviations: ADV, adenovirus; EBV, Epstein–Barr virus; HHV-8, human herpesvirus 8; γHV-68, murine γ-herpesvirus 68; HVS, herpesvirus saimiri; PPVO, parapoxvirus ORF virus; FPV, fowlpox virus; ASFV, African swine fever virus; VACV, vaccinia virus; MXV, myxoma virus, CMV, cytomegalovirus. C. Castanier, D. Arnoult / Biochimica et Biophysica Acta 1813 (2011) 575–583 577

2. Viral control of mitochondrial apoptosis gene products leads to the potent mitogenic effect of EBV [32]. BHRF1 and BALF1 share sequence and structure homology with Bcl-2 and Elimination of infected cells through apoptosis is one of the most hence are vBcl-2 s [33,34]. Whereas BHRF1 clearly resembles Bcl-2 in its ancestral defense mechanisms against infections, hence neutralizing anti-apoptotic role [35], the function of BALF1 is controversial. Indeed, host cell apoptosis represents a required step for viral replication [21]. BALF1 is capable of interacting with Bax and Bak [34] but another study Throughout the process of pathogen–host co-evolution, viruses have suggests that BALF1 is a BHRF1 antagonist. However, BALF1 is actively therefore acquired distinct strategies to neutralize immunity in expressed in EBV-positive Burkitt lymphoma's cell lines and nasopha- infected hosts and they have developed the capability of inhibiting ryngeal carcinoma biopsies and renders cells independent from serum host cell apoptosis. Conversely, viruses may take advantages of suggesting a prominent anti-apoptotic (rather than pro-apoptotic) promoting apoptosis, either to induce the breakdown of infected cells function for BALF1 during EBV-driven tumorigenesis [36]. Importantly, thereby favoring viral spreading or to kill uninfected cells of the anti-apoptotic proteins appear to be essential in the early phases of the immune system [21]. herpesvirus life cycle, as for other DNA viruses. The C-terminal Several proteins encoded by viral genome are homologues of the hydrophobic domain of BHRF1 is required for its mitochondrial host regulators of apoptosis such as Bcl-2 family members but key localization at the outer membrane and it exhibits high levels of components of the apoptotic machinery are also specifically targeted homology with several Bcl-2 family members [37]. Furthermore, the by viral factors with no homology to host proteins. So viral proteins 3-dimensional structure of BHRF1 is very similar to that of Bcl-2 since it that regulate mitochondria-mediated apoptosis can be classified into contains two central hydrophobic α-helices that are surrounded by the following subgroups: anti-apoptotic proteins (1) that share several amphipathic α-helices. While initial studies have described that sequence and/or structural similarity to anti-apoptotic Bcl-2 family BHRF1 does not seem to be capable of binding and inhibiting BH3-only members (so-called viral Bcl-2 proteins or vBcl-2 s) or pro-apoptotic proteins, presumably because it lacks an exposed pre-formed BH3 modulators (2) that directly insert into mitochondrial membranes to binding groove [33,37], two recent papers have shown that BHRF1 can trigger MOMP. A number a viral factors also inhibit apoptosis via other actually bind to Bim or BH3 domain peptides [38,39]. BHRF1 efficiently mechanisms which do not directly involve the Bcl-2 system and prevents MOMP and ensuing apoptosis induced by several differ- mitochondria (for review see [22]). ent stimuli such as DNA damaging agent, death receptors ligation leading, c-myc activation, heterologous viral infection, granzyme B, γ 3. Viral inhibitors of apoptosis irradiation or chemotherapeutic drugs [22].Surprisingly,BHRF1(but not Bcl-2) overexpression has been reported to trigger a rapid transit 3.1. Viral Bcl-2 homologs (vBcl-2) with sequence homology with Bcl-2 through the cell cycle suggesting yet another mechanism by which this protein might contribute to EBV-induced tumorigenesis [40]. Prevention of apoptosis of the infected cells is a way for a virus to BHRF1 is highly conserved both at the sequence and functional levels sustain its replication but it is also required for the establishment and/ in EBV analogues infecting baboons (herpesvirus papio) and chimpan- or the maintenance of the latency or the persistence of the virus. So zees (herpesvirus pan) [41] as well as in different EBV isolates [42].This viruses have selected a battery of Bcl-2 homologs that mimic the anti- suggests therefore that BHRF1 plays a significant, evolutionarily apoptotic machinery of host cell. conserved function in vivo. In accord, herpesvirus papio BHRF1 The first indication that viruses encode Bcl-2 functional homologs (hpoBHRF1) is a 17-kDa protein that shares 64% sequence identity and that their manipulation of apoptosis is a requirement for normal and 79% sequence similarity with BHRF1 from EBV. HpoBHRF1 and productive infection came from observations made with adenovirus- BHRF1 similarly protected human keratinocytes against cisplatin- infected cells. Human adenoviruses (ADV) are DNA virus pathogens, induced apoptosis in functional assays [43]. Finally, herpesvirus pan- some of which are the cause of as much as 10% of all childhood encoded BHRF1 (hpnBHRF1) acts like EBV BHRF1 in protecting a human pneumonias, whereas other cause conjunctivitis and outbreaks of Burkitt lymphoma cell line against apoptosis induced by UV irradiation, acute respiratory disease. One of the oncogenes of adenovirus (ADV), etoposide and serum withdrawal [44]. E1B-19K, was shown to be required to prevent apoptosis and enhanced Other vBcl-2 s have been described in the herpesvirus family such cytopathic effects during productive virus infection suggesting that the as the Kaposi sarcoma-associated Bcl-2 (KSBcl-2), γHV-68 M11 and E1B-19K protein may function as an apoptosis inhibitor in produc- HVS ORF16 that were identified in human herpesvirus 8 (HHV-8), tively infected human cells [23]. E1B-19K and Bcl-2 can functionally murine γ-herpesvirus 68 (γHV-68) and herpesvirus saimiri (HVS), substitute for one another in the suppression of apoptosis during virus respectively [45–47]. HHV-8 ORF16 encodes the so-called Kaposi infection and oncogenic transformation, indicating that viruses did, sarcoma-associated Bcl-2 (KSBcl-2). Overexpression of KSbcl-2 blocks indeed, see fit to encode inhibitors of apoptosis [24–26]. Moreover, apoptosis as efficiently as Bcl-2, Bcl-XL or BHRF1 [45]. KSBcl-2 is a E1B-19K inhibits cell death induced by a plethora of stimuli, including protein sharing limited (about 15–20%) overall sequence identity

E1A-triggered activation of p53, growth factor deprivation, ligation of with other Bcl-2 family members such as Bcl-2, Bcl-XL, Bax and Bak as TNFR, Fas and TRAIL receptor 1 (TRAIL-R1) at the plasma membrane well as BHRF1 [45]. Significant amino acid identity is concentrated in and heterologous Bax expression [27]. While E1B-19K and Bcl-2 share the BH1 and BH2 domains but curiously not in the BH3 domain. KSBcl- a limited sequence similarity [24], E1B-19K possesses a BH1, BH2 and 2 structure is almost identical to that of Bcl-2/Bcl-XL but important BH3 domain as well as a transmembrane domain and the most differences were found in the lengths of helices and loops [48].In conserved residues in the dozen or so viral serotypes in which the consequence, KSBcl-2 is not capable of forming homodimers or sequence of E1B-19K is known are also conserved in Bcl-2 [21,24]. heterodimers with Bcl-2 family members that may be explained by Similar to Bcl-2, a large fraction of E1B-19K is located at mitochondria the dissimilarity in sequence and structure with Bcl-2 [45,48]. The where it inhibits apoptosis by sequestering pro-apoptotic members of observation that KSBcl-2 neither homodimerizes or heterodimerizes the Bcl-2 family such as Bax [28–30], Bak [28] and Bnip3 [31]. Notably, suggests that this protein may have evolved to escape any negative several Bcl-2 family proteins such as Bak, Bik and the putative BH3- regulatory effects mediated by host pro-apoptotic Bcl-2 members only proteins Bnip3 and Nip3L have been originally identified through [45,48]. their interaction with E1B-19K [21]. γHV-68 M11 is a protein that unlike most of the other vBcl-2 s, Epstein–Barr virus (EBV) is a prominent member of γ herpesviruses, which have both BH1 and BH2 domains conserved with respect to Bcl- invading primary resting B lymphocytes to establish a latent infection 2, has a BH1 domain, but apparently lacks a BH2 domain [47]. In spite that eventually culminates in cell transformation. Coordinated expres- of this, M11 has been reported to inhibit Fas and TNFα-induced sion of anti-apoptotic proteins BHRF1 and BALF1 with other several apoptosis in cell lines [47] as well as Bax toxicity in yeast [49]. Nuclear 578 C. Castanier, D. Arnoult / Biochimica et Biophysica Acta 1813 (2011) 575–583 magnetic resonance determination of M11 structure has revealed a and to inhibit Bax activity is dependent on the structurally conserved BH3 binding groove that binds BH3 domain peptides of Bax and Bak BH3 domain of FPV039, even though this domain possesses little via a molecular mechanism shared with host Bcl-2 family proteins, sequence homology to other BH3 domains. FPV039 also inhibits involving a conserved arginine in the BH3 peptide binding groove. apoptosis induced by the “BH3-only” proteins, upstream activators of Mutations of this conserved arginine within the BH3 binding groove Bak and Bax, despite interacting detectably with only two: Bim and results in a properly folded protein that lacks the capacity of the wild- Bik [56]. In UV-irradiated HeLa cells, ORFV125 fully inhibits DNA type M11 to bind Bax BH3 peptide and to block Bax toxicity in yeast fragmentation, caspase activation and cytochrome c release. The [49]. Importantly, viruses harboring the same mutation exhibit mitochondrial localization of ORFV125 is determined by a distinctive impaired persistent replication and reactivation from latency in vivo C-terminal domain, and is required for its anti-apoptotic function. As pointing therefore a major role of M11 for the establishment of a assessed by bioinformatic analyses, ORFV125 shares sequence and persistent viral pool and chronic infection [49] rather than for viral structure similarities with anti-apoptotic members of the Bcl-2 family replication and virulence during acute infection as initially proposed (i.e., Bcl-2, Bcl-XL and Bcl-w). Accordingly, ORFV125 inhibits the UV- [50]. Finally, it has been reported that M11 inhibits autophagy induced activation of Bax and Bak [55]. Furthermore, ORFV125 through the binding of its hydrophobic surface groove to the BH3 interacts with a range of “BH3-only” proteins (Bik, Puma, Hrk, Noxa domain of Beclin1, a key autophagy effector [51]. and Bim) and neutralizes their pro-apoptotic activity. In addition, HVS ORF16 protects cells from heterologous virus-induced ORFV125 binds to the active, but not the inactive, form of Bax, and apoptosis. Unlike KSBcl-2, ORF16 heterodimerizes with Bax and Bak. reduces the formation of Bax dimers. Mutation of specific amino acids The BH1 and BH2 homology domains are highly conserved in ORF16 in ORFV125 that are conserved and functionally important in but it lacks the core sequence of the conserved BH3 homology domain, mammalian Bcl-2 family proteins leads to loss of both binding and suggesting that this region is not essential for the anti-apoptotic inhibitory functions [57]. Altogether, these observations confirm the activity [46]. Like BHRF1 or KSBcl-2, ORF16 contains conserved idea that ORFV125 represents a new member of the vBcl-2 family. hydrophobic residues in its C-terminus required for its insertion To avoid premature apoptosis of the host cell, African swine fever into mitochondrial outer membrane [45]. Interestingly, while some virus (ASFV) expresses multiple anti-apoptotic proteins such as the mammalian anti-apoptotic Bcl-2 can be converted into pro-apoptotic vBcl-2 family member A179L (also called 5-HL) [58,59]. A179L codes factors by caspases during apoptosis, several herpesvirus-encoded for a 21-kDa protein that contains all known domains associated with vBcl-2 cannot, therefore failing to display any latent pro-apoptotic Bcl-2 structural and functional features, including those mediating function [52]. protein-binding (i.e., homo- and heterodimerization) and regulating In addition to KSBcl-2, HHV-8 also expresses another protein with cell death [58]. Thus, A179L has been shown to suppress apoptosis very limited homology to Bcl-2 [53]. K7 is a 16-kDa glycoprotein induced by multiple stimuli [58,59]. structurally related to survivin-DeltaEx3, a splice variant of human Nine vBcl-2 s sharing sequence homologies (specially at the BH survivin that protects cells from apoptosis by an undefined mechanism. domains) with Bcl-2 have been described so far (Fig. 1) [22]. While some Both K7 and survivin-DeltaEx3 contain a mitochondrial targeting of them have been reported to directly bind Bax and/or Bak (E1B-19K, sequence, an N-terminal region of a BIR (baculovirus IAP repeat) BALF1, ORF16 or FPV039), they presumably all act like anti-apoptotic domain and a putative BH2 domain. K7 efficiently inhibits apoptosis Bcl-2 members [39] by inhibiting and sequestering activator(s) of Bax or induced by several stimuli (death receptors activation, ER stress or Bax Bak like the “BH3-only” proteins Bim and tBid [14] to prevent MOMP expression) and anchors to intracellular membranes where Bcl-2 re- and ensuing apoptosis. sides. K7 does not associate with Bax, but does bind to Bcl-2 via its putative BH2 domain. In addition, K7 binds to active caspase-3 via its BIR 3.2. vBcl-2 with structure homology with Bcl-2 domain and thus inhibits the activity of caspase-3. The BH2 domain of K7 is crucial for the inhibition of caspase-3 activity and is therefore Other viral factors are also classified as vBcl-2, not because they essential for its anti-apoptotic function. Furthermore, K7 bridges Bcl-2 share sequence homologies with Bcl-2 (for instance presence of “BH and activated caspase-3 into a protein complex. K7 therefore appears to domains”) but rather because they fold like Bcl-2 and exert similar be an adaptor protein and part of an anti-apoptotic complex that anti-apoptotic properties [22]. presents effector caspases to Bcl-2, enabling Bcl-2 to inhibit caspase Human cytomegalovirus (CMV) is a member of the β herpes- activity. Due to its molecular structure, K7 can be considered either as a viruses and it is a major cause of mortality in immuno-compromised viral IAP (vIAP) or as a vBcl-2 [53]. individuals (organ transplant recipients and patients with AIDS). CMV Poxviruses cause several diseases of humans and domestic animals is also a major cause of congenital disease during pregnancy. Human including smallpox, cowpox, sheeppox, fowlpox and goatpox and CMV encodes several proteins that subvert host cell function to favor recently two novel vBcl-2 inhibitors of apoptosis encoded by poxvirus viral replication. The best characterized is vMIA (viral mitochondria- have been discovered: FPV039 from fowlpox virus (FPV) [54] and localized inhibitor of apoptosis), the product of the CMV gene UL37 ORFV125 from parapoxvirus ORF virus (PPVO) [55]. While FPV039 has (pUL37x1) [60]. vMIA is required for viral replication and it inhibits limited homology with Bcl-2, it inhibits apoptosis in response to a apoptosis induced by diverse stimuli such as death receptor ligation, variety of cytotoxic stimuli, including virus infection itself. Similar to treatment with cytotoxic agents as well as infection with a mutant other anti-apoptotic Bcl-2 proteins, FPV039 localizes predominantly ADV strain lacking the vBcl-2 E1B-19K [60,61]. vMIA exerts its anti- to the mitochondria in both human and chicken cells. In addition, apoptotic activity by inhibiting MOMP at the mitochondrial level FPV039 interacts constitutively with Bak. Moreover, FPV039 is capable although vMIA does not share any obvious similarity with Bcl-2 or of substituting for F1L (another vBcl-2 encoded by vaccinia virus, vBcl-2 [60]. vMIA is inserted into the mitochondrial outer membrane VACV, see below) in inhibiting Bak activation and apoptosis triggered and importantly vMIA selectively binds/sequesters and inactivates by staurosporine or VACV infection, confirming that FPV039 is a Bax but not Bak to prevent MOMP [61]. The observation that the anti- functional homolog of F1L and is therefore a vBcl-2 [54]. FPV039 apoptotic function of vMIA is lost in Bax-deficient cells confirms that inhibits apoptosis induced by Bax overexpression as well and vMIA selectively acts on Bax to prevent apoptosis [61]. It appears that prevents both the conformational activation of Bax and the subse- both mitochondrial targeting sequence and Bax-binding domain are quent formation of Bax oligomers at the mitochondria, two critical required and sufficient for vMIA to prevent Bax activation [61]. steps in the induction of apoptosis. Additionally, FPV039 interacts The structure–function relationship of the Bax/vMIA interaction has with activated Bax in the context of Bax overexpression and virus been addressed by mutational and computational analyses, suggest- infection. Importantly, the ability of FPV039 to interact with active Bax ing a model in which the overall fold of vMIA closely resembles that of C. Castanier, D. Arnoult / Biochimica et Biophysica Acta 1813 (2011) 575–583 579

Bcl-XL [62]. In contrast to Bcl-XL, however, it seems that vMIA does not for the same peptide [75]. For all these reasons, M11L represents a bind to the BH3 domain of Bax but rather engages electrostatic structural mimic of Bcl-2 and hence a bona fide vBcl-2 [74]. interactions involving a region of Bax between its BH2 and BH3 N1L is the most potent virulence factor of VACV and is required for domains [62]. This region is not conserved in Bak, explaining why productive replication in vivo [76]. During viral infection, N1L binds to vMIA (unlike Bcl-XL) fails to interact with Bak [61,62]. Besides its anti- endogenous pro-apoptotic Bcl-2 family members such as Bax, Bak and apoptotic function through the neutralization of Bax, vMIA also Bid [77]. Like M11L, N1L shares no sequence homology with cellular induces the release of ER Ca2+ stores into the cytosol [63] and triggers proteins however it exhibits a three-dimensional structure that closely fragmentation of the mitochondrial network [64].Interestingly, mimics that of Bcl-2 and Bcl-XL [78]. So the surface of N1L possesses a fragmentation of the mitochondrial network following vMIA expres- constitutively open groove common to other anti-apoptotic members of sion seems to be responsible for an inhibition of the innate anti-virale the Bcl-2 family, which mediates the interaction with BH3-containing immunity [65] (see below). proteins [77]. The examples provided by M11L and N1L illustrate the Murine CMV also expresses a vBcl-2, termed m38.5. m38.5 is in a importance of the conservation of structure, rather than of primary similar position in the murine cytomegalovirus (MCMV) genome than amino acid sequence, for the maintenance of protein function along vMIA in human CMV. While m38 .5 is unrelated to vMIA in its amino acid evolution. sequence, it also inhibits death receptor ligation-induced apoptosis. m38.5 associates with Bax, recruits it to mitochondria, and blocks Bax- 4. Viral pro-apoptotic factors mediated but not Bak-mediated MOMP [66]. Thus, primate and murine CMVs have evolved non-homologous but functionally similar cell death Hepatitis B virus (HBV) is one of the most prominent etiological suppressors selectively targeting the Bax-mediated branch of the agents of chronic liver disease and persistent infection is associated mitochondrial apoptotic signaling pathway [66]. While some viruses with hepatocellular carcinoma. HBV X protein (HBx) is involved in express vBcl-2 that inhibits apoptosis by preventing activation and/or viral replication and exhibits an oncogenic potential in animal models oligomerization of the pro-apoptotic proteins Bax and Bak, it seems that [79]. HBx is addressed at the mitochondria where it induces CMVs have adopted a different strategy. Indeed, a recent report aggregation of these organelles and where it interacts with VDAC3 demonstrates that they encode two separate mitochondrial proteins to promote ΔΨm loss and ensuing apoptosis [80]. Interestingly HBx that lack obvious sequence similarities to Bcl-2-family proteins and sensitizes cells to apoptosis induced by TNFα [81]. Mutagenesis importantly specifically counteract either Bax or Bak [67]. In addition of studies revealed that a putative transmembrane region (aa 54–70) m38.5, MCMV open reading frame (ORF) m41.1 encodes for a small rather than its transactivation domain is required for its mitochon- mitochondrion-localized protein which functions as a viral inhibitor of drial localization and ΔΨm loss [80]. Bak oligomerization (vIBO). vIBO blocks Bak-mediated cytochrome c Poliovirus (PLV) infection causes paralytic poliomyelitis, an acute release and Bak-dependent induction of apoptosis. It protects cells from disease resulting in flaccid paralysis associated with apoptosis of cell death-inducing stimuli together with vMIA. Similar vIBO proteins motor neurons. Similar to HBx, viroporin 2B of PLV localizes to are encoded by CMVs of rats, and possibly by other CMVs as well [67]. mitochondria, alters their morphology, induces a perinuclear redis- These results suggest a non-redundant function of Bax and Bak during tribution of mitochondria and promotes caspase-dependent cell death viral infection, and a benefit for CMVs derived from the ability to inhibit [82]. OrfC protein from the Walleye dermal sarcoma virus (WDSV) Bak and Bax separately with two viral proteins. and E1E4 from the human papillomaviruses (HPVs) also affect The genome of many poxviruses encodes for proteins that despite mitochondrial morphology and distribution followed by dissipation the lack of sequence similarity, fold like Bcl-2, prevent apoptosis and of ΔΨm and apoptosis [83,84]. WDSV is a retrovirus causing benign are therefore considered as vBcl-2 s. F1L from VACV is responsible for tumors in fish whereas HPVs are responsible for a broad range of the inhibition of apoptosis [68]. Indeed cytochrome c release and infectious diseases, ranging from anogenital warts (e.g., HPV-6 and ensuing caspase activation do not occur in VACV infected cells. In -11) to progressive dysplastic-neoplastic lesions of the genital mucosa contrast, a vaccinia virus deletion mutant, VV811, is unable to inhibit (e.g., HPV-16 and -18). Both OrfC and E1E4 localize to mitochondria apoptosis; however, the anti-apoptotic function was restored by but while E1E4 associates to mitochondria due to an N-terminal expression of the F1L ORF. Although F1L displays no homology to leucine-rich region, OrfC does not seem to possess any homology to members of the Bcl-2 family, it localizes to the mitochondria through known mitochondrial targeting sequence [83,84]. a C-terminal hydrophobic domain [68]. Importantly F1L directly Retroviruses like the human immunodeficiency virus 1 (HIV-1) or interacts with the BH3 domain of pro-apoptotic Bcl-2 members such the human T lymphotropic virus 1 (HTLV-1, involves in neurogen- as Bak and Bim [69,70]. While F1L inhibits mitochondrial transloca- erative and lymphoproliferative disorders, notably acute T cell tion and activation of Bax, F1L does not bind to Bax suggesting that F1L leukemia) also express proteins with pro-apoptotic activity. For acts upstream of Bax activation [70]. Finally, other poxviruses (variola HIV-1, this contributes to the depletion of CD4+ T lymphocytes that is virus, monkeypoxvirus and ectromelia virus) encode functional F1L observed during AIDS. The viral protein R (Vpr) from HIV-1 has a orthologs [68]. direct mitochondrial effect because it interacts with ANT and VDAC, Myxoma virus (MXV) is a leporipoxvirus causing myxomatosis in thereby promoting PTP opening leading to MOMP associated with rabbit. Productive infection by MXV requires the expression of ΔΨm loss, release of mitochondrial apoptogenic factors and caspase M11L, a protein with no defined structural motifs except a putative activation [85,86]. pl3(II) protein from HTLV1 is also targeted to C-terminal TM domain that resembles that of some members of the mitochondria where it induces a rapid flux of ions across inner Bcl-2 family [71]. From its mitochondrial localization, M11L sup- membrane together with swelling, ΔΨm dissipation and mitochon- presses apoptosis induced by treatment with staurosporine [71], drial fragmentation [87]. overexpression of Bak [72] and viral infection [73]. M11L constitu- PB1-F2 is a protein encoded by influenza A virus (IAV) that tively interacts with Bax and Bak to prevent MOMP [74]. Bax- determines virulence in murine models of infection. PB1-F2 inserts mediated apoptosis is blocked in MXV-infected cells lacking Bak into mitochondrial outer membrane via its C-terminus and there, PB1- expression, suggesting that M11L interacts with Bax independently F2 acquires a pore forming activity similar to the pore forming activity from Bak [73]. M11L may also inhibit PTP opening to prevent displayed by pro-apoptotic Bcl-2 members, such as Bax, to promote apoptosis [71]. M11L has no obvious sequence homology with Bcl-2 MOMP [88,89]. Moreover, PB1-F2 interacts with VDAC and isoform 3 or Bcl-XL, yet adopts a virtually identical three-dimensional fold [75] of ANT (ANT3) suggesting that PB1-F2 may also promote MOMP allowing this protein to associate with a peptide derived from the BH3 through PTP opening [90]. Finally, other viral proteins such as VP3 and domain of Bak with an affinity comparable to that of Bcl-2 and Bcl-XL 2C from the avian encephalomyelitis virus (AEV) and the non- 580 C. Castanier, D. Arnoult / Biochimica et Biophysica Acta 1813 (2011) 575–583 structural protein 4A (NS4A) from hepatitis C virus (HCV) are also Although this interaction results in propagation of downstream sig- found associated to the mitochondria and promote MOMP through an naling, the mechanism by which the signal is transduced remains unknown mechanism [91–93]. poorly elucidated. In summary, manipulation of mitochondrial-mediated apoptosis is Following the MAVS/RIG-I or MAVS/MDA-5 association, MAVS was a common theme for several viruses and it can determine the found to associate with TNF-receptor-associated factor (TRAF) 3, an outcome of the infectious process. Even though mitochondrial E3 ubiquitin ligase assembling lysine 63-linked polyubiquitin chains, dysfunction has been historically associated with apoptosis, it is through its C-terminal TRAF domain [97]. TRAF3−/− cells show actually a crucial checkpoint for different modalities of cell death, severely impaired production of type I IFNs in response to virus including necrosis and autophagy. infection [98]. TRAF3 recruits and activates two “noncanonical” IKK- related kinases, designated TANK-binding kinase 1 (TBK1) and 5. Mitochondrion, a recruitment platform in the RIG-I-like inducible IkB kinase (IKK-i), which phosphorylate IRF-3 and IRF-7 receptor pathway (Fig. 2). TBK1 and IKK-i (also named IKK-ε) interact with TRAF family member associated NF-kB activator (TANK) and NAK-associated In 2004, a new pathway and TLR-independent response to virus protein 1 (NAP1). Phosphorylation of IRF-3 and IRF-7 by these kinases was uncovered with the discovery of RIG-I-like receptors (RLR) induces the formation of homodimers and/or heterodimers, which proteins [1,94]. This family of cytosolic PRRs is composed of 3 translocate into the nucleus and bind to IFN-stimulated response molecules: retinoic acid inducible gene-I (RIG-I), melanoma differ- elements (ISREs), thereby resulting in the expression of type I IFN entiation-associated gene-5 (MDA-5) and laboratory of genetics and genes and a set of IFN-inducible genes. Finally, the NF-κB modulator physiology 2 (LGP2). RIG-I and MDA-5 are both prototypical PRR NEMO (IKKγ) has been shown to be essential for the signal factors whereas LGP2 acts likely as an activator of RIG-I [1,95]. transduction downstream of the complex formed by TBK1, IKK-i, RIG-I and MDA-5 are both cytosolic helicases with ATPase activity TANK and NAP1 (Fig. 2) [99]. and they contain a C-terminal regulatory domain required for the In co-immunoprecipitation experiments, MAVS has also been binding to viral RNA. Importantly, the N-termini of RIG-I and MDA-5, reported to interact with TRAF6 [8,9], which in a complex with but not LGP2, contain 2 tandem CARD domains [1]. The ATPase transforming growth factor beta-activated kinase 1 (TAK1) and TAK1- activity of both helicases as a result of binding to their ligands may be binding protein 2 and 3 (TAB2 and TAB3) leads to NF-κB activation critical not only for translocation along dsRNA [96], but also for [100]. In addition, FAS-associated death domain-containing protein conformational changes that lead to the exposure of the CARDs (FADD) is found in a complex with MAVS [6] and the FADD/caspase8- masked by the C-terminal regulatory domain. This conformational dependent pathway could be required for the activation of NF-κB change is required for the interaction between the 2 CARD domains downstream of MAVS [101]. Importantly, FADD have been demon- of RIG-I and MDA-5 with the CARD domain of mitochondrial anti- strated to be important in innate anti-viral immunity since FADD−/− viral signaling protein (MAVS; also known as CARDIF, IPS-1 or VISA) cells are defective in type I IFNs production [102]. Finally, TRADD, a [6–9], a protein inserted in the mitochondrial outer membrane crucial adaptor of tumor necrosis factor receptor (TNFRI), is recruited through a transmembrane domain in its C-terminal region [8]. to MAVS and orchestrated complex formation with TRAF3, TANK and

Fig. 2. Overview of the RIG-I-like receptors pathway. In the cytosol, viral RNAs are recognized by the helicases RIG-I or MDA-5. The N-termini of RIG-I and MDA-5 contain 2 CARD domain that interact with the CARD domain of the mitochondrial adaptor MAVS. MAVS, after recruitment of transactivators, then induces phosphorylation of IRF3 and IRF7 and NF-κB activation leading to the production of type I IFNs and pro-inflammatory cytokines, respectively. C. Castanier, D. Arnoult / Biochimica et Biophysica Acta 1813 (2011) 575–583 581 with FADD and RIP1, leading to the activation of IRF3 and NF-κB. induces higher production of IFN-β and appears attenuated in Accordingly, TRADD−/− cells show a reduced production of type I comparison to the virus expressing the mitochondrial PB2 [110]. IFNs in response to RNA viruses [103]. Hepatitis B virus X (HBX) has been initially described as a pro- Although the links between MAVS and the downstream transacti- apoptotic protein [80,81] (see above) but a very recent study has vators mentioned above (Fig. 2) are not fully understood, the recent demonstrated that HBX interacts with MAVS and promotes the discovery of STING (also called MITA) [104,105], has added an degradation of MAVS through Lys(136) ubiquitin in MAVS protein, important new factor to this pathway. Two groups have reported that thus preventing the induction of IFN-β [111]. Analysis of clinical this protein interacts with both MAVS and RIG-I and is required for samples reveals that MAVS protein is downregulated in hepatocellu- signaling through these molecules [104,105]. Results by Ishikawa and lar carcinomas of HBV origin, which correlates with increased sen- Barber suggest that STING acts downstream of RIG-I and MAVS [104]. sitivities of primary murine hepatocytes isolated from HBX knock- Surprisingly, STING is found at the endoplasmic reticulum and co- in transgenic mice upon vesicular stomatitis virus (VSV) infections immunoprecipitates with SSR2 (also known as TRAPβ), a member of [111]. the translocon-associated protein (TRAP) complex required for In a recent study, we highlighted a link between mitochondrial protein translocation across the ER membrane following translation. dynamics and RLR [65]. Indeed, we observed that RLR activation The translocon associates with the exocyst that can associate with promotes elongation of the mitochondrial network. Mimicking this TBK1 leading to the hypothesis that RIG-I binds to translating viral elongation enhances signaling downstream from MAVS and favors the RNAs at the translocon and signals through STING to exocyst- binding of MAVS to STING. By contrast, enforced mitochondrial associated TBK1 [104]. Importantly, STING is not only required for fragmentation dampens signaling and reduces the association immune responses against certain RNA virus but STING seems to play between both proteins. Our finding that MAVS is associated with a a critical role in type I IFN induction by double-stranded B-form DNA pool of mitofusin 1, a protein of the mitochondrial fusion machinery, in the cytosol [104,105]. suggests that MAVS is capable of regulating mitochondrial dynamics to facilitate the mitochondria-ER association required for signal transduction. RLR activation actually induces degradation of one 6. Viral control of mitochondria-mediated anti-viral MAVS isoform with the consequent disruption of a MAVS–Mfn1– innate immunity STING complex located at sites of ER-mitochondrial tethering, freeing Mfn1 to orchestrate mitochondrial fusion. Fusion is required to During evolution, viruses have selected/acquired distinct strate- enhance further STING-MAVS interactions and the downstream gies to neutralize immunity in infected hosts [106]. In this review, signaling propagation. Interestingly, we observed that expression of only the viral proteins targeted to mitochondria to prevent mito- vMIA, a CMV anti-apoptotic protein that promotes mitochondrial chondria-mediated anti-viral innate immunity are detailed. fragmentation [60,61,64] (see above), hampers signaling downstream As mentioned above, MAVS is localized in the mitochondrial outer from MAVS [65] suggesting a possible new immune modulation membrane and must be localized at the mitochondria to exert its strategy of the CMV required for its replication and pathogenesis function [8]. One strategy that HCV employs to establish chronic in vivo. It appears therefore that some viruses may also subvert innate infection is to use the viral Serine protease NS3/4A to cleave some immunity by expressing proteins that induce mitochondrial morpho- unknown cellular targets involved in innate immunity. Recent studies logical changes to impede signaling downstream of MAVS. Interest- have shown that one of the targets of NS3/4A is MAVS. NS3/4A cleaves ingly, the viral pro-apoptotic factors HBx, viroporin 2B, OrfC and E1E4 MAVS at Cys-508, resulting in its inactivation as a consequence of the described above also promote mitochondrial aggregation [80,82–84] dislocation of the N-terminal fragment of MAVS from the mitochon- and in our hands, induction of mitochondrial aggregation leads to an dria. Remarkably, a point mutation of MAVS at Cys-508 renders MAVS inhibition of MAVS-mediated signaling as well (unpublished resistant to cleavage by NS3/4A, thus maintaining the ability of MAVS observations). to induce interferons in HCV replicon cells. NS3/4A binds to and colocalizes with MAVS in the mitochondrial membrane, and it can cleave MAVS directly in vitro [7,107]. Identical observations were 7. Concluding remarks made with GB virus B (GBV-B) [108], phylogenetically the closest related virus to HCV which causes generally acute and occasionally The relation between mitochondria in innate immune responses chronic hepatitis in small primates. Similarly, hepatitis A virus (HAV), has been studied, until recently, mostly through the angle of host a hepatotropic picornavirus, ablates type 1 IFN responses by targeting cell death/survival. This is indeed a crucial point, as the fate of the 3ABC precursor of its 3C(pro) cysteine protease to mitochondria infected cells critically impacts on the outcome of an infection at the where it colocalizes with and cleaves MAVS, thereby disrupting level of a tissue or the whole organism. Moreover, because activation of IRF3. The 3ABC cleavage of MAVS requires both the apoptosis and necrosis of infected cells result in dramatically protease activity of 3C(pro) and a transmembrane domain in 3A that different inflammatory responses in the tissue microenvironment, directs 3ABC to mitochondria. Lacking this domain, mature 3C(pro) understanding how viruses regulate cell death pathways is of protease is incapable of MAVS proteolysis [109]. crucial importance. The fact that so many cases of viral manipula- The PB2 subunit of the influenza virus RNA polymerase is one of tion of mitochondrial cell death pathways have been reported the virulence determinant of influenza viruses. Very recently, it was reinforces the concept that the control of these cell death cascades reported that in addition to its nuclear localization, PB2 accumulated is essential for the physiopathology of a number of viruses. Yet, the in the mitochondria where it interacts and inhibits MAVS [110]. The pathways linking detection of viral infection by PRRs to mitochon- authors also observed that the PB2 proteins of seasonal human dria-mediated cell death remain poorly characterized. An interest- influenza viruses localize to the mitochondria, unlike the PB2 proteins ing challenge for the years will be to characterize in detail how of avian influenza viruses. This difference in localization is explained PRRs link viral detection to mitochondria-dependent cell death. by a single amino acid polymorphism in the PB2 mitochondrial Finally, mitochondria appear to control much more than cell death/ targeting sequence. While the generation of two recombinant survival following infection. This organelle seems to centralize a human influenza viruses encoding either mitochondrial or non- number of critical innate immune responses, as illustrated by the mitochondrial PB2 proteins shows that the mitochondrial localization identification of MAVS as a mitochondrial protein, and understand- of PB2 has no impact on viral growth in cell culture, in an animal ing the reasons for such an interplay will undoubtedly represent an model, the virus expressing the non-mitochondrial PB2 proteins interesting field of investigation for the coming years. 582 C. Castanier, D. Arnoult / Biochimica et Biophysica Acta 1813 (2011) 575–583

Acknowledgements [32] M. Altmann, W. Hammerschmidt, Epstein–Barr virus provides a new paradigm: a requirement for the immediate inhibition of apoptosis, PLoS Biol. 3 (2005) e404. [33] Q. Huang, A.M. Petros, H.W. Virgin, S.W. Fesik, E.T. Olejniczak, Solution structure We apologize to our colleagues for the works that are not cited of the BHRF1 protein from Epstein–Barr virus, a homolog of human Bcl-2, J. Mol. because of the lack of space. Research in our laboratory is supported Biol. 332 (2003) 1123–1130. [34] W.L. Marshall, C. Yim, E. Gustafson, T. Graf, D.R. Sage, K. Hanify, L. Williams, J. by grants from the Agence Nationale de La Recherche sur le SIDA Fingeroth, R.W. Finberg, Epstein–Barr virus encodes a novel homolog of the bcl-2 (ANRS), Fondation pour La Recherche Médicale (FRM) and La Ligue oncogene that inhibits apoptosis and associates with Bax and Bak, J. Virol. 73 (1999) contre le Cancer (équipe labellisée) and from Université Paris Sud. 5181–5185. [35] S. Henderson, D. Huen, M. Rowe, C. Dawson, G. Johnson, A. Rickinson, Epstein–Barr virus-coded BHRF1 protein, a viral homologue of Bcl-2, protects human B cells from References programmed cell death, Proc. Natl Acad. Sci. USA 90 (1993) 8479–8483. [36] G. Cabras, G. Decaussin, Y. Zeng, D. Djennaoui, H. Melouli, P. Broully, A.M. [1] S. Akira, S. Uematsu, O. Takeuchi, Pathogen recognition and innate immunity, Bouguermouh, T. Ooka, Epstein–Barr virus encoded BALF1 gene is transcribed in Cell 124 (2006) 783–801. Burkitt's lymphoma cell lines and in nasopharyngeal carcinoma's biopsies, J. Clin. [2] T. Kawai, S. Akira, The roles of TLRs, RLRs and NLRs in pathogen recognition, Int. Virol. 34 (2005) 26–34. Immunol. 21 (2009) 317–337. [37] T. Hickish, D. Robertson, P. Clarke, M. Hill, F. di Stefano, C. Clarke, D. Cunningham, [3] V. Hornung, E. Latz, Intracellular DNA recognition, Nat. Rev. Immunol. 10 (2010) Ultrastructural localization of BHRF1: an Epstein–Barr virus gene product which 123–130. has homology with bcl-2, Cancer Res. 54 (1994) 2808–2811. [4] O. Takeuchi, S. Akira, Pattern recognition receptors and inflammation, Cell [38] A.L. Desbien, J.W. Kappler, P. Marrack, The Epstein–Barr virus Bcl-2 homolog, 140 (2010) 805–820. BHRF1, blocks apoptosis by binding to a limited amount of Bim, Proc. Natl Acad. [5] D. Arnoult, L. Carneiro, I. Tattoli, S.E. Girardin, The role of mitochondria in cellular Sci. USA 106 (2009) 5663–5668. defense against microbial infection, Semin. Immunol. 21 (2009) 223–232. [39] A.M. Flanagan, A. Letai, BH3 domains define selective inhibitory interactions [6] T. Kawai, K. Takahashi, S. Sato, C. Coban, H. Kumar, H. Kato, K.J. Ishii, O. Takeuchi, with BHRF-1 and KSHV BCL-2, Cell Death Differ. 15 (2008) 580–588. S. Akira, IPS-1, an adaptor triggering RIG-I- and Mda5-mediated type I interferon [40] C.W. Dawson, J. Dawson, R. Jones, K. Ward, L.S. Young, Functional differences induction, Nat. Immunol. 6 (2005) 981–988. between BHRF1, the Epstein–Barr virus-encoded Bcl-2 homologue, and Bcl-2 in [7] E. Meylan, J. Curran, K. Hofmann, D. Moradpour, M. Binder, R. Bartenschlager, J. human epithelial cells, J. Virol. 72 (1998) 9016–9024. Tschopp, Cardif is an adaptor protein in the RIG-I antiviral pathway and is [41] T. Williams, D. Sale, S.A. Hazlewood, BHRF1 is highly conserved in primate virus targeted by hepatitis C virus, Nature 437 (2005) 1167–1172. analogues of Epstein–Barr virus, Intervirology 44 (2001) 55–58. [8] R.B. Seth, L. Sun, C.K. Ea, Z.J. Chen, Identification and characterization of MAVS, a [42] F. Khanim, C. Dawson, C.A. Meseda, J. Dawson, M. Mackett, L.S. Young, BHRF1, a mitochondrial antiviral signaling protein that activates NF-kappaB and IRF 3, Cell viral homologue of the Bcl-2 oncogene, is conserved at both the sequence and 122 (2005) 669–682. functional level in different Epstein–Barr virus isolates, J. Gen. Virol. 78 (Pt 11) [9] L.G. Xu, Y.Y. Wang, K.J. Han, L.Y. Li, Z. Zhai, H.B. Shu, VISA is an adapter protein (1997) 2987–2999. required for virus-triggered IFN-beta signaling, Mol. Cell 19 (2005) 727–740. [43] C.A. Meseda, J.R. Arrand, M. Mackett, Herpesvirus papio encodes a functional [10] N.N. Danial, S.J. Korsmeyer, Cell death: critical control points, Cell 116 (2004) homologue of the Epstein–Barr virus apoptosis suppressor, BHRF1, J. Gen. Virol. 205–219. 81 (2000) 1801–1805. [11] S.J. Riedl, Y. Shi, Molecular mechanisms of caspase regulation during apoptosis, [44] M. Howell, T. Williams, S.A. Hazlewood, Herpesvirus pan encodes a functional Nat. Rev. Mol. Cell Biol. 5 (2004) 897–907. homologue of BHRF1, the Epstein–Barr virus v-Bcl-2, BMC Microbiol. 5 (2005) 6. [12] X. Wang, The expanding role of mitochondria in apoptosis, Genes Dev. 15 (2001) [45] E.H. Cheng, J. Nicholas, D.S. Bellows, G.S. Hayward, H.G. Guo, M.S. Reitz, J.M. 2922–2933. Hardwick, A Bcl-2 homolog encoded by Kaposi sarcoma-associated virus, human [13] S. Galban, C.S. Duckett, XIAP as a ubiquitin ligase in cellular signaling, Cell Death herpesvirus 8, inhibits apoptosis but does not heterodimerize with Bax or Bak, Differ. 17 (2010) 54–60. Proc. Natl Acad. Sci. USA 94 (1997) 690–694. [14] S.N. Willis, J.M. Adams, Life in the balance: how BH3-only proteins induce [46] V.E. Nava, E.H. Cheng, M. Veliuona, S. Zou, R.J. Clem, M.L. Mayer, J.M. Hardwick, apoptosis, Curr. Opin. Cell Biol. 17 (2005) 617–625. Herpesvirus saimiri encodes a functional homolog of the human bcl-2 oncogene, [15] J.E. Chipuk, T. Moldoveanu, F. Llambi, M.J. Parsons, D.R. Green, The BCL-2 family J. Virol. 71 (1997) 4118–4122. reunion, Mol. Cell 37 (2010) 299–310. [47] G.H. Wang, T.L. Garvey, J.I. Cohen, The murine gammaherpesvirus-68M11 [16] D.R. Green, G. Kroemer, The pathophysiology of mitochondrial cell death, protein inhibits Fas- and TNF-induced apoptosis, J. Gen. Virol. 80 (Pt 10) (1999) Science 305 (2004) 626–629. 2737–2740. [17] N. Zamzami, G. Kroemer, The mitochondrion in apoptosis: how Pandora's box [48] Q. Huang, A.M. Petros, H.W. Virgin, S.W. Fesik, E.T. Olejniczak, Solution structure opens, Nat. Rev. Mol. Cell Biol. 2 (2001) 67–71. of a Bcl-2 homolog from Kaposi sarcoma virus, Proc. Natl Acad. Sci. USA 99 [18] J.C. Martinou, D.R. Green, Breaking the mitochondrial barrier, Nat. Rev. Mol. Cell (2002) 3428–3433. Biol. 2 (2001) 63–67. [49] J. Loh, Q. Huang, A.M. Petros, D. Nettesheim, L.F. van Dyk, L. Labrada, S.H. Speck, B. [19] A. Degterev, J. Yuan, Expansion and evolution of cell death programmes, Nat. Levine, E.T. Olejniczak, H.W.T. Virgin, A surface groove essential for viral Bcl-2 Rev. Mol. Cell Biol. 9 (2008) 378–390. function during chronic infection in vivo, PLoS Pathog. 1 (2005) e10. [20] G. Kroemer, B. Levine, Autophagic cell death: the story of a misnomer, Nat. Rev. [50] S. Gangappa, L.F. van Dyk, T.J. Jewett, S.H. Speck, H.W.T. Virgin, Identification of Mol. Cell Biol. 9 (2008) 1004–1010. the in vivo role of a viral bcl-2, J. Exp. Med. 195 (2002) 931–940. [21] A. Cuconati, E. White, Viral homologs of BCL-2: role of apoptosis in the regulation [51] S. Sinha, C.L. Colbert, N. Becker, Y. Wei, B. Levine, Molecular basis of the of virus infection, Genes Dev. 16 (2002) 2465–2478. regulation of Beclin 1-dependent autophagy by the gamma-herpesvirus 68 Bcl-2 [22] L. Galluzzi, C. Brenner, E. Morselli, Z. Touat, G. Kroemer, Viral control of homolog M11, Autophagy 4 (2008) 989–997. mitochondrial apoptosis, PLoS Pathog. 4 (2008) e1000018. [52] D.S. Bellows, B.N. Chau, P. Lee, Y. Lazebnik, W.H. Burns, J.M. Hardwick, [23] E. White, R. Cipriani, P. Sabbatini, A. Denton, Adenovirus E1B 19-kilodalton protein Antiapoptotic herpesvirus Bcl-2 homologs escape caspase-mediated conversion overcomes the cytotoxicity of E1A proteins, J. Virol. 65 (1991) 2968–2978. to proapoptotic proteins, J. Virol. 74 (2000) 5024–5031. [24] S.K. Chiou, C.C. Tseng, L. Rao, E. White, Functional complementation of the [53] H.W. Wang, T.V. Sharp, A. Koumi, G. Koentges, C. Boshoff, Characterization of an adenovirus E1B 19-kilodalton protein with Bcl-2 in the inhibition of apoptosis in anti-apoptotic glycoprotein encoded by Kaposi's sarcoma-associated herpesvi- infected cells, J. Virol. 68 (1994) 6553–6566. rus which resembles a spliced variant of human survivin, Embo J. 21 (2002) [25] L. Rao, M. Debbas, P. Sabbatini, D. Hockenbery, S. Korsmeyer, E. White, The 2602–2615. adenovirus E1A proteins induce apoptosis, which is inhibited by the E1B 19-kDa [54] L. Banadyga, J. Gerig, T. Stewart, M. Barry, Fowlpox virus encodes a Bcl-2 and Bcl-2 proteins, Proc. Natl Acad. Sci. USA 89 (1992) 7742–7746. homologue that protects cells from apoptotic death through interaction with the [26] E. White, P. Sabbatini, M. Debbas, W.S. Wold, D.I. Kusher, L.R. Gooding, The 19- proapoptotic protein Bak, J. Virol. 81 (2007) 11032–11045. kilodalton adenovirus E1B transforming protein inhibits programmed cell death [55] D. Westphal, E.C. Ledgerwood, M.H. Hibma, S.B. Fleming, E.M. Whelan, A.A. and prevents cytolysis by tumor necrosis factor alpha, Mol. Cell. Biol. 12 (1992) Mercer, A novel Bcl-2-like inhibitor of apoptosis is encoded by the parapoxvirus 2570–2580. ORF virus, J. Virol. 81 (2007) 7178–7188. [27] E. White, Regulation of the cell cycle and apoptosis by the oncogenes of [56] L. Banadyga, K. Veugelers, S. Campbell, M. Barry, The fowlpox virus BCL-2 adenovirus, Oncogene 20 (2001) 7836–7846. homologue, FPV039, interacts with activated Bax and a discrete subset of BH3- [28] A. Cuconati, K. Degenhardt, R. Sundararajan, A. Anschel, E. White, Bak and Bax only proteins to inhibit apoptosis, J. Virol. 83 (2009) 7085–7098. function to limit adenovirus replication through apoptosis induction, J. Virol. [57] D. Westphal, E.C. Ledgerwood, J.D. Tyndall, M.H. Hibma, N. Ueda, S.B. Fleming, 76 (2002) 4547–4558. A.A. Mercer, The orf virus inhibitor of apoptosis functions in a Bcl-2-like [29] J. Han, D. Modha, E. White, Interaction of E1B 19K with Bax is required to block manner, binding and neutralizing a set of BH3-only proteins and active Bax, Bax-induced loss of mitochondrial membrane potential and apoptosis, Onco- Apoptosis 14 (2009) 1317–1330. gene 17 (1998) 2993–3005. [58] C.L. Afonso, J.G. Neilan, G.F. Kutish, D.L. Rock, An African swine fever virus Bc1-2 [30] J. Han, P. Sabbatini, D. Perez, L. Rao, D. Modha, E. White, The E1B 19K protein homolog, 5-HL, suppresses apoptotic cell death, J. Virol. 70 (1996) 4858–4863. blocks apoptosis by interacting with and inhibiting the p53-inducible and death- [59] A. Brun, C. Rivas, M. Esteban, J.M. Escribano, C. Alonso, African swine fever virus promoting Bax protein, Genes Dev. 10 (1996) 461–477. gene A179L, a viral homologue of bcl-2, protects cells from programmed cell [31] M. Yasuda, P. Theodorakis, T. Subramanian, G. Chinnadurai, Adenovirus E1B- death, Virology 225 (1996) 227–230. 19K/BCL-2 interacting protein BNIP3 contains a BH3 domain and a mitochon- [60] V.S. Goldmacher, L.M. Bartle, A. Skaletskaya, C.A. Dionne, N.L. Kedersha, C.A. drial targeting sequence, J. Biol. Chem. 273 (1998) 12415–12421. Vater, J.W. Han, R.J. Lutz, S. Watanabe, E.D. Cahir McFarland, E.D. Kieff, E.S. C. Castanier, D. Arnoult / Biochimica et Biophysica Acta 1813 (2011) 575–583 583

Mocarski, T. Chittenden, A cytomegalovirus-encoded mitochondria-localized Susin, D. Cointe, Z.H. Xie, J.C. Reed, B.P. Roques, G. Kroemer, The HIV-1 viral inhibitor of apoptosis structurally unrelated to Bcl-2, Proc. Natl Acad. Sci. USA 96 protein R induces apoptosis via a direct effect on the mitochondrial permeability (1999) 12536–12541. transition pore, J. Exp. Med. 191 (2000) 33–46. [61] D. Arnoult, L.M. Bartle, A. Skaletskaya, D. Poncet, N. Zamzami, P.U. Park, J. Sharpe, [87] M. Silic-Benussi, I. Cavallari, T. Zorzan, E. Rossi, H. Hiraragi, A. Rosato, K. Horie, D. R.J. Youle, V.S. Goldmacher, Cytomegalovirus cell death suppressor vMIA blocks Saggioro, M.D. Lairmore, L. Willems, L. Chieco-Bianchi, D.M. D'Agostino, V. Bax- but not Bak-mediated apoptosis by binding and sequestering Bax at Ciminale, Suppression of tumor growth and cell proliferation by p13II, a mitochondria, Proc. Natl Acad. Sci. USA 101 (2004) 7988–7993. mitochondrial protein of human T cell leukemia virus type 1, Proc. Natl Acad. Sci. [62] A.L. Pauleau, N. Larochette, F. Giordanetto, S.R. Scholz, D. Poncet, N. Zamzami, V.S. USA 101 (2004) 6629–6634. Goldmacher, G. Kroemer, Structure–function analysis of the interaction between [88] A.N. Chanturiya, G. Basanez, U. Schubert, P. Henklein, J.W. Yewdell, J. Zimmerberg, Bax and the cytomegalovirus-encoded protein vMIA, Oncogene 26 (2007) PB1-F2, an influenza A virus-encoded proapoptotic mitochondrial protein, 7067–7080. creates variably sized pores in planar lipid membranes, J. Virol. 78 (2004) [63] R. Sharon-Friling, J. Goodhouse, A.M. Colberg-Poley, T. Shenk, Human cytomeg- 6304–6312. alovirus pUL37x1 induces the release of endoplasmic reticulum calcium stores, [89] J.S. Gibbs, D. Malide, F. Hornung, J.R. Bennink, J.W. Yewdell, The influenza A virus Proc. Natl Acad. Sci. USA 103 (2006) 19117–19122. PB1-F2 protein targets the inner mitochondrial membrane via a predicted basic [64] A.L. McCormick, V.L. Smith, D. Chow, E.S. Mocarski, Disruption of mitochondrial amphipathic helix that disrupts mitochondrial function, J. Virol. 77 (2003) networks by the human cytomegalovirus UL37 gene product viral mitochon- 7214–7224. drion-localized inhibitor of apoptosis, J. Virol. 77 (2003) 631–641. [90] D. Zamarin, A. Garcia-Sastre, X. Xiao, R. Wang, P. Palese, Influenza virus PB1-F2 [65] C. Castanier, D. Garcin, A. Vazquez, D. Arnoult, Mitochondrial dynamics regulate protein induces cell death through mitochondrial ANT3 and VDAC1, PLoS the RIG-I-like receptor antiviral pathway, Embo Rep. 11 (2010) 133–138. Pathog. 1 (2005) e4. [66] D. Arnoult, A. Skaletskaya, J. Estaquier, C. Dufour, V.S. Goldmacher, The murine [91] J. Liu, T. Wei, J. Kwang, Avian encephalomyelitis virus induces apoptosis via cytomegalovirus cell death suppressor m38.5 binds Bax and blocks Bax- major structural protein VP3, Virology 300 (2002) 39–49. mediated mitochondrial outer membrane permeabilization, Apoptosis 13 [92] J. Liu, T. Wei, J. Kwang, Avian encephalomyelitis virus nonstructural protein (2008) 1100–1110. 2C induces apoptosis by activating cytochrome c/caspase-9 pathway, Virology [67] M. Cam, W. Handke, M. Picard-Maureau, W. Brune, Cytomegaloviruses inhibit 318 (2004) 169–182. Bak- and Bax-mediated apoptosis with two separate viral proteins, Cell Death [93] Y. Nomura-Takigawa, M. Nagano-Fujii, L. Deng, S. Kitazawa, S. Ishido, K. Sada, H. Differ. 17 (2010) 655–665. Hotta, Non-structural protein 4A of hepatitis C virus accumulates on mitochon- [68] S.T. Wasilenko, T.L. Stewart, A.F. Meyers, M. Barry, Vaccinia virus encodes a dria and renders the cells prone to undergoing mitochondria-mediated previously uncharacterized mitochondrial-associated inhibitor of apoptosis, apoptosis, J. Gen. Virol. 87 (2006) 1935–1945. Proc. Natl Acad. Sci. USA 100 (2003) 14345–14350. [94] M. Yoneyama, M. Kikuchi, T. Natsukawa, N. Shinobu, T. Imaizumi, M. [69] A. Postigo, J.R. Cross, J. Downward, M. Way, Interaction of F1L with the BH3 Miyagishi, K. Taira, S. Akira, T. Fujita, The RNA helicase RIG-I has an essential domain of Bak is responsible for inhibiting vaccinia-induced apoptosis, Cell function in double-stranded RNA-induced innate antiviral responses, Nat. Death Differ. 13 (2006) 1651–1662. Immunol. 5 (2004) 730–737. [70] J.M. Taylor, D. Quilty, L. Banadyga, M. Barry, The vaccinia virus protein F1L [95] T. Satoh, H. Kato, Y. Kumagai, M. Yoneyama, S. Sato, K. Matsushita, T. interacts with Bim and inhibits activation of the pro-apoptotic protein Bax, Tsujimura, T. Fujita, S. Akira, O. Takeuchi, LGP2 is a positive regulator of RIG-I- J. Biol. Chem. 281 (2006) 39728–39739. and MDA5-mediated antiviral responses, Proc. Natl Acad. Sci. USA 107 (2010) [71] H. Everett, M. Barry, S.F. Lee, X. Sun, K. Graham, J. Stone, R.C. Bleackley, G. 1512–1517. McFadden, M11L: a novel mitochondria-localized protein of myxoma virus that [96] S. Myong, S. Cui, P.V. Cornish, A. Kirchhofer, M.U. Gack, J.U. Jung, K.P. Hopfner, T. blocks apoptosis of infected leukocytes, J. Exp. Med. 191 (2000) 1487–1498. Ha, Cytosolic viral sensor RIG-I is a 5′-triphosphate-dependent translocase on [72] G. Wang, J.W. Barrett, S.H. Nazarian, H. Everett, X. Gao, C. Bleackley, K. Colwill, double-stranded RNA, Science 323 (2009) 1070–1074. M.F. Moran, G. McFadden, Myxoma virus M11L prevents apoptosis through [97] S.K. Saha, E.M. Pietras, J.Q. He, J.R. Kang, S.Y. Liu, G. Oganesyan, A. Shahangian, B. constitutive interaction with Bak, J. Virol. 78 (2004) 7097–7111. Zarnegar, T.L. Shiba, Y. Wang, G. Cheng, Regulation of antiviral responses by a [73] J. Su, G. Wang, J.W. Barrett, T.S. Irvine, X. Gao, G. McFadden, Myxoma virus M11L direct and specific interaction between TRAF3 and Cardif, Embo J. 25 (2006) blocks apoptosis through inhibition of conformational activation of Bax at the 3257–3263. mitochondria, J. Virol. 80 (2006) 1140–1151. [98] G. Oganesyan, S.K. Saha, B. Guo, J.Q. He, A. Shahangian, B. Zarnegar, A. Perry, G. [74] M. Kvansakul, M.F. van Delft, E.F. Lee, J.M. Gulbis, W.D. Fairlie, D.C. Huang, P.M. Cheng, Critical role of TRAF3 in the Toll-like receptor-dependent and -independent Colman, A structural viral mimic of prosurvival Bcl-2: a pivotal role for antiviral response, Nature 439 (2006) 208–211. sequestering proapoptotic Bax and Bak, Mol. Cell 25 (2007) 933–942. [99] T. Zhao, L. Yang, Q. Sun, M. Arguello, D.W. Ballard, J. Hiscott, R. Lin, The NEMO [75] A.E. Douglas, K.D. Corbett, J.M. Berger, G. McFadden, T.M. Handel, Structure of adaptor bridges the nuclear factor-kappaB and interferon regulatory factor M11L: A myxoma virus structural homolog of the apoptosis inhibitor, Bcl-2, signaling pathways, Nat. Immunol. 8 (2007) 592–600. Protein Sci. 16 (2007) 695–703. [100] J. Inoue, J. Gohda, T. Akiyama, Characteristics and biological functions of TRAF6, [76] N. Bartlett, J.A. Symons, D.C. Tscharke, G.L. Smith, The vaccinia virus N1L protein Adv. Exp. Med. Biol. 597 (2007) 72–79. is an intracellular homodimer that promotes virulence, J. Gen. Virol. 83 (2002) [101] K. Takahashi, T. Kawai, H. Kumar, S. Sato, S. Yonehara, S. Akira, Roles of caspase- 1965–1976. 8 and caspase-10 in innate immune responses to double-stranded RNA, J. Immunol. [77] S. Cooray, M.W. Bahar, N.G. Abrescia, C.E. McVey, N.W. Bartlett, R.A. Chen, D.I. 176 (2006) 4520–4524. Stuart, J.M. Grimes, G.L. Smith, Functional and structural studies of the vaccinia [102] S. Balachandran, E. Thomas, G.N. Barber, A FADD-dependent innate immune virus virulence factor N1 reveal a Bcl-2-like anti-apoptotic protein, J. Gen. Virol. mechanism in mammalian cells, Nature 432 (2004) 401–405. 88 (2007) 1656–1666. [103] M.C. Michallet, E. Meylan, M.A. Ermolaeva, J. Vazquez, M. Rebsamen, J. Curran, H. [78] M. Aoyagi, D. Zhai, C. Jin, A.E. Aleshin, B. Stec, J.C. Reed, R.C. Liddington, Vaccinia Poeck, M. Bscheider, G. Hartmann, M. Konig, U. Kalinke, M. Pasparakis, J. virus N1L protein resembles a B cell lymphoma-2 (Bcl-2) family protein, Protein Tschopp, TRADD protein is an essential component of the RIG-like helicase Sci. 16 (2007) 118–124. antiviral pathway, Immunity 28 (2008) 651–661. [79] C.M. Kim, K. Koike, I. Saito, T. Miyamura, G. Jay, HBx gene of hepatitis B virus [104] H. Ishikawa, G.N. Barber, STING is an endoplasmic reticulum adaptor that induces liver cancer in transgenic mice, Nature 351 (1991) 317–320. facilitates innate immune signalling, Nature 455 (2008) 674–678. [80] Z. Rahmani, K.W. Huh, R. Lasher, A. Siddiqui, Hepatitis B virus X protein [105] B. Zhong, Y. Yang, S. Li, Y.Y. Wang, Y. Li, F. Diao, C. Lei, X. He, L. Zhang, P. Tien, H.B. colocalizes to mitochondria with a human voltage-dependent anion channel, Shu, The Adaptor Protein MITA Links Virus-Sensing Receptors to IRF3 HVDAC3, and alters its transmembrane potential, J. Virol. 74 (2000) 2840–2846. Transcription Factor Activation, Immunity 29 (2008) 538–550. [81] F. Su, R.J. Schneider, Hepatitis B virus HBx protein sensitizes cells to apoptotic [106] B.B. Finlay, G. McFadden, Anti-immunology: evasion of the host immune system killing by tumor necrosis factor alpha, Proc. Natl Acad. Sci. USA 94 (1997) by bacterial and viral pathogens, Cell 124 (2006) 767–782. 8744–8749. [107] X.D. Li, L. Sun, R.B. Seth, G. Pineda, Z.J. Chen, Hepatitis C virus protease NS3/4A [82] V. Madan, A. Castello, L. Carrasco, Viroporins from RNA viruses induce caspase- cleaves mitochondrial antiviral signaling protein off the mitochondria to evade dependent apoptosis, Cell. Microbiol. 10 (2008) 437–451. innate immunity, Proc. Natl Acad. Sci. USA 102 (2005) 17717–17722. [83] W.A. Nudson, J. Rovnak, M. Buechner, S.L. Quackenbush, Walleye dermal sarcoma [108] Z. Chen, Y. Benureau, R. Rijnbrand, J. Yi, T. Wang, L. Warter, R.E. Lanford, S.A. virus Orf C is targeted to the mitochondria, J. Gen. Virol. 84 (2003) 375–381. Weinman, S.M. Lemon, A. Martin, K. Li, GB virus B disrupts RIG-I signaling by NS3/4A- [84] K. Raj, S. Berguerand, S. Southern, J. Doorbar, P. Beard, E1 empty set E4 protein of mediated cleavage of the adaptor protein MAVS, J. Virol. 81 (2007) 964–976. human papillomavirus type 16 associates with mitochondria, J. Virol. 78 (2004) [109] Y. Yang, Y. Liang, L. Qu, Z. Chen, M. Yi, K. Li, S.M. Lemon, Disruption of innate 7199–7207. immunity due to mitochondrial targeting of a picornaviral protease precursor, [85] E. Jacotot, K.F. Ferri, C. El Hamel, C. Brenner, S. Druillennec, J. Hoebeke, P. Rustin, Proc. Natl Acad. Sci. USA 104 (2007) 7253–7258. D. Metivier, C. Lenoir, M. Geuskens, H.L. Vieira, M. Loeffler, A.S. Belzacq, J.P. [110] K.M. Graef, F.T. Vreede, Y.F. Lau, A.W. McCall, S.M. Carr, K. Subbarao, E. Fodor, The PB2 Briand, N. Zamzami, L. Edelman, Z.H. Xie, J.C. Reed, B.P. Roques, G. Kroemer, subunit of the influenza virus RNA polymerase affects virulence by interacting with Control of mitochondrial membrane permeabilization by adenine nucleotide MAVS and inhibiting IFN-{beta} expression, J. Virol. 84 (2010) 8433–8445. translocator interacting with HIV-1 viral protein rR and Bcl-2, J. Exp. Med. 193 [111] C. Wei, C. Ni, T. Song, Y. Liu, X. Yang, Z. Zheng, Y. Jia, Y. Yuan, K. Guan, Y. Xu, X. (2001) 509–519. Cheng, Y. Zhang, Y. Wang, C. Wen, Q. Wu, W. Shi, H. Zhong, The hepatitis B virus [86] E. Jacotot, L. Ravagnan, M. Loeffler, K.F. Ferri, H.L. Vieira, N. Zamzami, P. X protein disrupts innate immunity by downregulating mitochondrial antiviral Costantini, S. Druillennec, J. Hoebeke, J.P. Briand, T. Irinopoulou, E. Daugas, S.A. signaling protein, J. Immunol. 185 (2010) 1158–1168.