INTRODUCTION
RNA interference (RNAi), also called RNA silencing, is a conserved eukaryotic pathway involved in suppression of gene expression via sequence-specific interactions that are mediated by small-interfering RNAs (siRNAs), which are 21–23 nt in length and are derived from exogenous double-stranded RNA (dsRNA) or endogenous microRNA (miRNA) (
Ruiz et al., 1998;
Zamore et al., 2000;
Bernstein et al., 2001;
Huang et al., 2007). In addition to its diverse biological functions, RNAi has recently emerged as an important antiviral mechanism in host cells (
Coleman, 2007;
van Rij, 2008). It was first discovered in plants that RNAi provides an antiviral mechanism by which viral dsRNAs are cleaved by the ribonuclease Dicer to generate viral siRNAs. These siRNAs are then incorporated into the RNA-induced silencing complex (RISC) and direct RISC activity to messenger RNAs (mRNAs) in a sequence-specific manner (
Ding and Voinnet, 2007;
Ding, 2010;
Olivieri et al., 2010;
Reyes et al., 2011). Complementary sequences in the viral mRNA can be targeted by the viral siRNAs for endonucleic cleavage or translational inhibition, which represses viral replication.
Whether RNAi has natural antiviral activity in mammals is a hotly debated issue. Two main points have been put forward by those who do not think that RNAi has natural antiviral activity in mammals. The first is that virus-derived siRNAs cannot be detected by small-RNA cloning in tissue-culture models for the hepatitis C virus (HCV) and yellow fever virus (
Pfeffer et al., 2005). Although a single specific viral-derived siRNA corresponding to the structured Rev responsive element has been detected in human immunodeficiency virus-1 (HIV-1) infected cells (
Bennasser et al., 2005), this observation could not be reproduced by another laboratory (
Lin and Cullen, 2007). The second reason is that many viruses, such as influenza A virus and vaccinia virus, replicated to similar, or even lower, viral titers in Dicer
−/− macrophages (
Otsuka et al., 2007). On the other hand, although whether viruses encode miRNAs during the infection process has been controversial, increasing evidence suggests that miRNAs which are encoded by host cells can regulate the replication of the virus. Song et al. reported that host miR-323, miR-491 and miR-654 inhibit the replication of the H1N1 influenza virus in MDCK cells by binding and degrading PB1 (a subunit of influenza virus RNA polymerase) (
Song et al., 2010). Otsuka et al. have reported that host miR-24 and miR-93 can target viral large protein and phosphoprotein genes, and that a lack of miR-24 and miR-93 is responsible for increased vesicular stomatitis virus (VSV) replication (
Otsuka et al., 2007). These data suggest that RNAi is probably part of the innate immune system in mammals.
Since RNAi is an important antiviral pathway in host cells in plants, as a counter-defensive strategy, many viruses encode silencing suppressors, a kind of protein which can suppress RNA silencing at different stages of the process (
Moissiard and Voinnet, 2004). The first hint that viruses encode silencing suppressors came from seminal experiments aimed at understanding the phenomenon of synergism (
Vance, 1991). In 1997, the potyviral helper component proteinase (Hc-Pro) was first identified as a silencing suppressor. With further research, an increasing number of silencing suppressors have been identified in plant viruses and also in mammalian viruses, which indicates silencing suppressors as a host counter defensive mechanism widely adopted by different viruses. In this review, we will discuss a few examples of silencing suppressors and the molecular basis of their function.
SILENCING SUPPRESSORS ENCODED BY PLANT VIRUSES
In plants, RNA silencing forms the basis of an elaborate immune system that is activated by, and targeted against, viruses. As a counter-defensive strategy, viruses have evolved suppressor proteins that inhibit various stages of the silencing process (Table 1). These suppressors are diverse in sequence and structure.
HC-Pro
The first silencing suppressor discovered is potyvirus-encoded helper component-proteinase (HC-Pro) which enhances the replication of many unrelated viruses (
Kasschau et al., 1997;
Pruss et al., 1997).
Tobacco etch virus silencing suppressor is comprised of the 5′-proximal region of the tobacco etch potyviral genomic RNA encoding P1, HC-Pro and a small part of P3, and is termed P1/HC-Pro. P1/HC-Pro can suppress virus-induced posttranscriptional gene silencing and further assessment of the function of P1/HC-Pro showed that expression of the HC-Pro coding sequence alone is sufficient to suppress virus-induced gene silencing, and that the HC-Pro protein product is required for suppression (
Anandalakshmi et al., 1998;
Brigneti et al., 1998;
Kasschau and Carrington, 1998). As HC-Pro is a pathogenicity determinant and is involved in other viral functions such as polyprotein processing, it is speculated that silencing suppressors play other roles in addition to suppressing RNAi.
P19
Another well-characterized suppressor is the tombusvirus-encoded P19 protein. P19, which is highly conserved and expressed in many different tombusvirus species, was first identified as a silencing suppressor in 1999 (
Voinnet et al., 1999). P19 suppresses RNAi in plants and blocks RNAi by binding siRNA via its dsRNA-binding domain. High dosages of P19 are essential for viral pathogenicity. Since siRNAs are universally involved in RNA-silencing, these findings predict that P19 should be effective in a broad range of organisms. Dunoyer et al. reported that P19 can bind siRNA and suppress RNAi in HeLa cells, suggesting that this factor may be useful for dissecting the RNA silencing pathway in animals (
Dunoyer et al., 2004).
DO MAMMALIAN VIRUSES ENCODE SILENCING SUPPRESSORS?
It has been clearly demonstrated that plant viruses encode silencing suppressors during infection. However, whether mammalian viruses encode silencing suppressors has been controversial. Kok and Jin reported that the influenza A virus non-structural (NS1) protein does not suppress RNA interference in mammalian cells (
Kok and Jin, 2006). On the other hand, Haasnoot et al. have reported that NS1 can act as a silencing suppressor in Vero cells which lack the interferon (IFN) pathway. Their results suggest that certain cellular factors involved in IFN pathway may effect the silencing suppressor activity of NS1 in mammalian cells (
Haasnoot et al., 2007). In addition, Bennasser et al. reported that HIV-1 evades elicited RNAi through a suppressor of RNA silencing (SRS) function encoded in its Tat protein; Tat suppresses an otherwise effective RNAi restriction through functional abrogation of the cell’s Dicer activity (
Bennasser et al., 2005). However, Lin reported that neither HIV-1 nor human T-cell leukemia virus type 1 (HTLV-1) expresses significant levels of either small interfering RNAs or microRNAs in persistently infected T cells (
Lin and Cullen, 2007). The author also demonstrates that the retroviral nuclear transcription factors HIV-1 Tat and HTLV-1 Tax, as well as the Tas transactivator encoded by the primate foamy virus, fail to inhibit RNA interference in human cells.
Although whether silencing suppressors are encoded by mammalian viruses has been controversial, increasing evidence shows that mammalian viruses should encode silencing suppressors to counter the defense system of host cells. Many miRNAs have been found to play very important functions in inhibiting the replication of viruses. We would expect that, in line with typical interactions between viruses and their hosts, silencing suppressors should be encoded by mammalian viruses to counter host responses.
But why has this debate lasted such a long time? It may be because, first, in mammalian cells, the main antiviral mechanism involves both the IFN pathway and the 2′–5′ oligoadenylate cyclase (2′–5′ OAS)/RNAseL/PKR pathway. Whether miRNA is an antiviral pathway is still unclear; second, that NS1 only acts as a silencing suppressor in IFN-lacking cells suggests that the function of silencing suppressors may be determined by certain cellular factors; third, since the defense system of mammalian cells is very intricate, when mammalian viruses infect host cells, several proteins may work together as a silencing suppression complex during infection.
SILENCING SUPPRESSORS ENCODED BY MAMMALIAN VIRUSES
At present, only a few silencing suppressors encoded by mammalian viruses have been identified. These silencing suppressors have many common characteristics. First, silencing suppressors have IFN/PKR antagonistic properties which are related to the pathogenicity of the virus. Second, most silencing suppressors have dsRNA binding domains and can bind to dsRNA or siRNA. Finally, silencing suppressors are multi-functional proteins during the infection process.
Tat protein of HIV-1
HIV-1 Tat is a multi-functional protein (
El Kharroubi et al., 1998;
Isel and Karn, 1999;
Liu et al., 2002;
Epie et al., 2005). In infected cells, Tat not only increases initiation of HIV-1 transcription but can also help to maintain the transcription process (
Herrmann and Rice, 1995;
Yang et al., 1997;
Kiernan et al., 1999).
In addition to its function in HIV-1 transcription, Tat may contribute to HIV-1 pathogenesis by regulating signal transduction in endothelial cells, functioning as a secreted growth factor for Kaposisarcoma and endothelial cells (
Cantaluppi et al., 2001;
Rusnati and Presta, 2002), and inducing apoptosis in T-cells by binding to microtubules and delaying tubulin depolymerization (
Nath et al., 1996;
Chen et al., 2002). In 2005, Bennasser el al. reported that HIV encodes viral siRNA precursors in its genome and that natural HIV-1 infection provokes nucleic acid-based immunity in human cells. To combat this cellular defense, HIV has evolved a suppressor of RNA silencing function in its Tat protein. Their results show that Tat plays its silencing suppressor function by inhibiting Dicer activity and that its silencing suppressor activity is distinct from its transcriptional function (
Bennasser et al., 2005).
VP35 of the Ebola virus
VP35, an ebolavirus protein, plays a key role in the interaction of the virus with its target cells (
Hartman et al., 2004). VP35 is required for viral replication and transcription and antagonizes the host cell type I IFN system during viral infection (
Mühlberger et al., 1999;
Basler et al., 2000). Haasnoot et al. reported that VP35 is a silencing suppressor in mammalian cells and that its RNA silencing suppressor (RSS) activity is functionally equivalent to that of the HIV-1 Tat protein. The VP35 dsRNA-binding domain is required for this silencing suppressor activity (
Haasnoot et al., 2007).
The non-structural protein of influenza A virus
Since the non-structural (NS1) protein of influenza A virus, like the silencing suppressors encoded by mammalian viruses described above, possesses IFN or protein kinase R (PKR) antagonistic properties, we speculated that it is a silencing suppressor during viral infection. The NS1 protein of influenza A viruses is a non-essential virulence factor that has multiple accessory functions during viral infection. It has been demonstrated that the NS1 protein has silencing suppressor activity in insect and plant cells (
Bucher et al., 2004;
Delgadillo et al., 2004;
Li et al., 2004). However, the silencing suppressor activity of NS1 in mammalian cells was not observed until 2007; results showed that NS1 protein is able to suppress RNAi in Vero (IFN
−) cells but not in HEK293T (IFN
+) cells (
Haasnoot et al., 2007). Haasnoot et al. suggest that NS1 is able to suppress short hairpin RNA (shRNA)-induced RNAi in mammalian cells, and that the cell type-dependent silencing suppressor activity of NS1 is related to the expression level of certain cellular RNAi co-factors.
Vaccinia virus E3L (E3L)
Vaccinia virus, a member of the
Orthopoxviridae, contains a large double-stranded DNA genome carrying approximately 190 genes (
Langland et al., 2006). The vaccinia virus E3L gene is an essential virulence gene that confers IFN resistance in cell culture of a broad range of hosts (
Beattie et al., 1996;
Langland and Jacobs, 2002). The E3L gene encodes a 190-amino-acid protein with a highly conserved, carboxyl-terminal, double-stranded RNA (dsRNA)-binding domain (
Chang and Jacobs, 1993;
Saunders and Barber, 2003). The dsRNA-binding domain has been shown to be required for both IFN resistance and broad-host-range phenotypes associated with the vaccinia virus (
Chang et al., 1995). In addition to the above-mentioned function of E3L, Li et al. reported that E3L can block induced RNAi against reporter gene constructs. Moreover, E3L is able to replace the RNAi suppression function of the b2 protein encoded by the Flock house virus and to support virus replication in insect cells (
Li et al., 2004). The mechanism of the RNAi suppression function of E3L is still not clear.
Hepatitis C virus core
The core protein is the first synthesized protein during the early phases of HCV infection. It can influence cellular immunity, cell growth, apoptosis, cell transformation, and eventually tumor development (hepatocellular carcinoma and possibly B cell lymphoma) (
Basu et al., 2001). Wang et al. reported that the HCV core protein is able to counteract the RNA-silencing response through a direct interaction with Dicer. This ability of the core protein to counter host defenses may contribute to the establishment of persistent HCV infections (
Wang et al., 2006).
Primate foamy virus type 1 (PFV-1) Tas
Lecellier et al. reported that cellular microRNA effectively restricts the accumulation of PFV-1 in human cells (
Lecellier et al., 2005). To counter the defense system of the host cell, PFV-1 encodes a viral transactivator, Tas, which acts as a silencing suppressor during infection. Like several plant viral suppressors, Tas can promote the nonspecific over-accumulation of cellular miRNAs (
Kasschau et al., 2003;
Dunoyer et al., 2004). Tas also shows silencing suppression activity in a heterologous system.
THE MOLECULAR BASIS OF SILENCING SUPPRESSION
At present, research on the action mechanism of silencing suppression is just beginning. Here we review the modes of action of these silencing suppressors (Fig. 2). (1) Silencing suppressor can inhibit the key components of RNA-silencing pathways, as highlighted by the function of the tombusviral P19 protein.
In vitro, P19 binds PTGS (posttranscriptional gene silencing)-generated, 21–25 nt dsRNAs and 21 nt synthetic dsRNA with 2-nt 3′ overhanging end(s), while it barely interacts with single stranded (ss) RNAs, long dsRNAs or blunt-ended 21-nt dsRNAs. Silhavy et al. propose that P19 mediates silencing suppression by sequestering PTGS-generated 21–25 nt dsRNAs (
Silhavy et al., 2002). P19 also binds to classes of small RNAs
in vivo and P19-mediated sequestration of siRNA in virus-infected cells blocks the spread of the mobile, systemic signal of RNA silencing (
Lakatos et al., 2004). The mechanism by which P19 selectively recognizes the siRNA silencing signal was unknown until 2003 when Vargason et al. presented the crystal structure of the
Carnation Italian Ringspot Virus (CIRV) P19 protein in complex with a 21 nt siRNA. The results showed that the CIRV P19 protein acts as a molecular caliper to specifically select siRNAs based on the length of the duplex region of the RNA. In addition to binding to the siRNA to inhibit the RNAi pathway, some silencing suppressors such as HIV-1 Tat and hepatitis C virus core protein can suppress the RNAi pathway by inhibiting Dicer activity. (2) A second silencing-suppression strategy involves the recruitment of endogenous negative regulators of RNA silencing. For example, yeast two-hybrid system results show that HC-Pro can interact with calmodulin-related protein (termed rgs-CaM), another cellular suppressor of PTGS (
Anandalakshmi et al., 2000). (3) Some silencing suppressors function by interacting with molecules from host cell. For example, geminivirus transcriptional-activator proteins (TrAPs, also known as AL2, AC2, or C2 protein) have been identified as silencing suppressors. It has been reported that TrAPs from
tomato golden mosaic virus (TGMV) bind zinc and that zinc is necessary for their optimal interaction with single-stranded DNA. The nuclear localization and zinc- and DNA-binding activities of TrAPs are all required for their suppressor function, indicating that they are involved in the function of TrAPs at the host-DNA level (
Hartitz et al., 1999;
van et al., 2002;
Van Wezel et al., 2003). VP35 can bind to dsRNA and this dsRNA binding capacity is required for its silencing suppressor activity. These results suggest that VP35-mediated RNAi suppression and VP35-dsRNA-dependent IFN antagonistic properties are linked. Haasnoot et al. proposed that VP35 is able to sequester virus-specific siRNA or dsRNA precursors of siRNA, resulting in suppression of an antiviral RNAi response that acts upstream of the 2′–5′ OAS/RNAseL/PKR and IFN pathway (
Haasnoot et al., 2007).
CONCLUSION AND FUTURE DIRECTIONS
Our understanding of silencing suppressor has considerably improved over the past few years. This research has also promoted the emergence of new and important concepts, such as that RNAi is a part of the native immunity system and that many viruses can encode negative regulators of RNA silencing and achieve successful replication in host cells. Mechanisms of silencing suppressors encoded by plant viruses are becoming clearer; however, the mechanisms of silencing suppressors encoded by mammalian viruses still require further research. The major challenge for the future will be to fully determine the extent to which RNA silencing is integrated within the interlinked and numerous layers of innate host defenses, such as the IFN pathway, and to understand the impact of silencing suppression on these pathways. In any case, it is anticipated that immune systems that are based on small RNAs either of pathogenic or cellular origin, are likely to be widespread.
Higher Education Press and Springer-Verlag Berlin Heidelberg 2011