An important host defense mechanism against virus infection, the type I IFNs, are rapidly induced to activate an antiviral state in infected and neighboring uninfected cells. Type I IFNs bind to the type I IFN receptor formed by IFNAR1 and IFNAR2 subunits [1,2] and trigger the activation of the canonical Jak/signal transducer and activator of transcription (STAT) signaling pathway and the trans-activation of antiviral genes, through a mechanism that depends to a great extent on STAT1, STAT2 and IRF9 [3]. These three transcription factors form the trimeric IFN-stimulated gene factor 3 (ISGF3) complex, which is relocated to the nucleus where it binds to the IFN-stimulated response elements (ISREs) in promoters of target genes to activate their transcription. Since type I IFNs stimulate p53 expression and virus infection activates p53 [4], this tumor suppressor has been recently included as a new component of the cellular antiviral defense mechanisms. Studies by us and others in recent years indicate that p53 is in fact, a key player in innate antiviral immunity by both enforcing the type I IFN response upon viral infection, and inducing apoptosis in infected cells [5-9]. Both actions coordinated by this tumor suppressor gene, help to thwart the replication of a wide variety of viruses both in vitro and in vivo [5-9]. In this review we try to provide a summary of the recent findings that involve p53 in antiviral immunity and that presumably, may help to explain not only why this protein is conserved in invertebrate organisms which do not undergo cancer-related diseases, but also, why this protein is so frequently targeted by viral proteins.

The potent activity of p53 as inducer of cellular apoptosis and cell cycle arrest demands tight control of its function. In normal cells the level of p53 protein is low due to its short half-life [10] regulated mainly by murine double minute 2 (Mdm2) protein. Mdm2 inhibits p53 transactivation and prompts p53 for proteasomal degradation by promoting its ubiquitylation [11,12]. In contrast, p53 is stabilized and activated when, in response to oncogene expression, another tumor suppressor gene, namely ARF, binds to Mdm2 and blocks its ability to inhibit p53 function [13]. Similarly, the nucleolar protein nucleophosmin (NPM) also interacts with Mdm2 and promotes p53 stability and function [14]. A recently discovered regulator of p53 activity is the type I IFN pathway. IFN-α/β stimulates transcription of the gene encoding p53, through ISRE sites present in its promoter, resulting in an early induction of the protein [4], which is later held in time by the action of ARF [15]. Importantly, type I IFN seems also to induce p53 protein stabilization through posttranslational mechanisms that involve STAT1-dependent transcriptional downregulation of Mdm2 and direct protein interaction with p53, at least upon DNA damage [16]. In addition, although acute IFN treatment does not cause the activation of p53 [4], chronic treatment of primary cells with IFN-β leads to p53 phosphorylation at Ser-15 and engagement of p53 downstream genes, indicative of p53 activation [17]; suggesting that the cellular response to IFN-β is time-dependent. Infection with some viral agents, such as vesicular stomatitis virus (VSV), Newcastle disease virus (NDV) or herpes simplex virus (HSV), probably through the production of interferon, boosts p53 expression, and induces the phosphorylation of p53 at Ser-15 by ataxia telangiectasia mutated (ATM) kinase [4]. Phosphorylation of p53 in other amino acid residues in response to IFN or virus infection cannot be excluded since p53 has been also identified as a substrate for the double-stranded RNA (dsRNA) activated protein kinase R (PKR), an IFN-inducible protein kinase which can induce the phosphorylation of human p53 at Ser-392 in vitro [18,19], an event that has been proposed to be related with the stabilization and activation of p53 [20-22]. In this regard, induction of p53 as a downstream event of tumor necrosis factor alpha (TNF-α)-induced upregulation of PKR has been reported, and impaired p53-mediated responses to doxorubicin in PKR KO cells have been described [17,18,23].

p53 induces the expression of a wide number of genes that ultimately impose a state of G1-phase cell cycle arrest or apoptosis. Previous studies indicate that the role of p53 in the control of virus infection depends at least to some extent on its ability to activate apoptosis through the transactivation of p53 classic target genes such as the members of the BH3-only family of proteins PUMA and Noxa [4,24]. Interestingly, a recent study also indicates that the type I IFN-inducible protein kinase PKR, is a novel p53 target gene, suggesting that PKR could also play a role in p53-dependent induction of apoptosis upon viral infection [25].

An intriguing unresolved question is whether TNF-related cytokines also contribute to p53-dependent apoptosis in response to viral infection. TNF-α is expressed at high levels upon infection with several viruses such as influenza A, human immunodeficiency virus (HIV) and Ebola virus (EBOV) [26-28]. Expression of TNF-α by virus-infected cells induces maturation of resident dendritic cells (DCs) through upregulation of the costimulatory molecules CD80 and CD86 [29]. Besides, TNF-α synergizes with type I IFNs and polarizes the adaptive immune response towards a Th1 type (cellular immunity) [30,31]. Moreover, recent reports indicate that TNFR-mediated apoptosis of virus-infected cells may be a key component at the interface between innate and adaptive immunity by allowing DCs to cross-present viral antigens to naïve T cells [32]. Specifically, the p53 direct target gene TRAIL-R2/DR5/KILLER has been previously shown to promote DC maturation and immune enhancement in vivo by activation of caspase 8-dependent apoptosis [33]. Viral antigens produced by apoptotic virus-infected cells are then engulfed by DCs which have the ability to present them to CD8⁺ T cells through major histocompatibility complex (MHC) class I [34]. This process may be especially important in triggering immunity against viruses that do not naturally infect DCs, and includes not only TRAIL-R2/DR5/KILLER but also other p53 direct target genes such as Fas/Apo1/CD95 [35,36]. Of note, activation of p53 downstream targets that specifically mediate its proapoptotic activity in response to virus infection may be also cell type specific. Thus, although VSV infection induced increased apoptosis in a p53 dose-dependent fashion in mouse embryonic fibroblasts (MEFs) [5], infection of HCT116 tumor cells retaining wt p53 with VSV, did not elicit an apoptotic response [37]. These findings also indicate, that virus-induced p53-dependent apoptosis may require the activity of p53 downstream target genes that could be impaired in cancer cell lines such as HCT116.

p53-dependent apoptosis has been demonstrated as a useful mechanism to control some virus infection, as shown for VSV [5,6,38], influenza A virus [7], herpes simplex virus (HSV) [8], or poliovirus [9]. In sharp contrast other viruses have evolved mechanisms to use p53 activity in its own benefit. Thus, p53 enhances the ability of human cytomegalovirus (HCMV) to replicate in fibroblasts [39], increases respiratory syncytial virus (RSV) [40] or adenovirus replication [41] and its absence has a detrimental effect in the growth of encephalomyocarditis virus (EMCV) and parainfluenza virus [38]. A possible explanation for this striking observation is that, while early apoptosis is probably detrimental for replication of some viruses, other viruses may benefit from apoptosis late in the replication cycle in order to improve transmission of newly-formed viral particles to other cells or hosts [42].

The complexity of the p53 response to viral infection is supported by studies addressing the role of p53 on dsRNA-induced apoptosis. Although p53 is often considered as a cell death inducer, there are also cell types and situations where the presence of p53 prevents cell death by inducing cell cycle arrest. Thus, the induction of G1 arrest mediated by p53 in response to dsRNA has been suggested as a putative mechanism to delay apoptosis and consequently increase EMCV replication [38]. Moreover, p53 functions as a transcription factor are also subverted by some viruses in their own benefit. Thus, p53 activates the transcription of the HCMV L44 protein, required for virus replication, and there are 21 exact matches for the p53 binding site within the virus genome, including sites that could influence the expression of proteins involved both in the structure and replication of HCMV [39]. p53 also cooperates with E1A to enhance transcription from the major late adenovirus promoter, triggering the adenovirus lytic cycle and increasing virus yield [41]. Finally, p53-dependent downregulation of IL-6 has been suggested to mediate the positive effect of p53 on RSV replication. However, the decrease in the p53 levels induced by RSV to delay apoptosis paradoxically enhances the host inflammation response that hampers virus replication [40].

Takaoka and colleagues [4] identified p53 as a type I IFN transcriptional target and demonstrated the existence of a crosstalk between p53 and the IFN pathway both in tumor suppression and antiviral defense [4]. An increasing body of work indicates that in fact, p53 crosstalk takes place both at the level of IFN production and IFN signaling. These findings indicate that p53, through direct transcriptional upregulation of several target genes, influences the type I IFN pathway that further suggests an important role for p53 in innate immunity. Here, we have summarized a list of genes that are transcriptional targets of both type I IFN and p53 and the possible implications of this crosstalk in antiviral immunity.

Upon virus infection, IFN and the tumor suppressor p53 are employed by host cells as components of their antiviral defense mechanisms. Thus, viruses need to tightly oppose these antiviral surveillance mechanisms of the host. Viruses have evolved elaborate mechanisms to subvert p53-mediated host innate immune responses. Either p53 itself or cellular factors involved in its downstream activities are inactivated by various viral proteins either by releasing cells from cell-cycle checkpoints or by protecting cells from p53-dependent apoptosis [78-80]. The number of viruses that have been shown to interfere with p53 activity has increased during the last years. Simian virus 40 (SV40) large T antigen [78,81]; Epstein Barr virus (EBV) BZLF1 [82] and EBV EBNA3C [83]; Kaposi’s sarcoma associated herpesvirus (KSHV) LANA, LANA2 and ORF K8 [84-86]; the Human herpesvirus 6 (HHV6) ORF1 [87], all interact with p53 and inhibit its activity. Adenovirus E1B-55K and E4-ORF6 proteins, human papillomavirus (HPV) E6, EBV EBNA-5, KSHV vIRF4 and the vaccinia virus (VV) B1R kinase, all induce the degradation of p53 [88-93]. Other proteins such as Adenovirus E1A or HPV E7 stabilize p53 but inhibit its transcriptional activity [94-98]. KSHV vIRF1 interacts with the ATM kinase to block its activity, thereby reducing p53 phosphorylation at the serine 15 residue and increasing p53 ubiquitylation [99,100]. EBNA-5 and polyoma virus target and inhibit critical ARF-mediated signals for p53 activation [101,102]. Other example such as the hepatitis B virus (HBV) X protein has been shown to interact with p53 and inhibits its functional activity in multiple ways [103].

As mentioned earlier, the number of viruses that has been demonstrated to control apoptosis induced by a p53-dependent mechanism has increased in the latest years, however not all the viruses induce p53 downregulation or block apoptosis induced by p53-dependent pathways. Some studies have demonstrated how some viruses induce an increase in p53 levels or activation of p53-regulated pathways. Thus, p53 activation by EBV LMP1, cytomegalovirus (CMV) IE84 or IE72, RSV fusion protein F or HIV-1 [104-107] and both upregulation and activation of p53 by KSHV viral cyclin [108] have been reported.

p53 is a tumor suppressor gene which is found mutated or absent in more than 50% of human cancers [109]. In fact, due to the ability of p53 to respond in many ways to DNA damaging insults, this protein has been called ‘the guardian of the genome’ [110]. However, p53 is also present and functional in organisms whose life span precludes them from undergoing cancer-related diseases such as flies and worms [111,112]. Moreover, a recent body of work indicates that besides ‘classic’ p53 inducers such as DNA damage and oncogene expression, its function can be also promoted by hypoxia, nutrient deprivation and viral infection [4,113,114]. These findings, and the recently identified functions of p53 in both virus-induced apoptosis and upregulation of the type I IFN response, are summarized in Figure 1, and suggest, that p53 may act as a virus induced cellular stress sensor. The role of p53 in antiviral immunity may help to explain to some extent why this protein is so highly conserved in evolution.