Influenza virus polymerase: Functions on host range, inhibition of cellular response to infection and pathogenicity

Highlights

• Influenza virus structure, function and host factor association.

• Determinants of host range and pathogenicity identified in the influenza A virus polymerase.

• Recently discovered proteins expressed from influenza A virus polymerase genes.

• Inhibition of cellular response to influenza A virus infection.

• Cellular transcription blockage induced by influenza A virus.

Abstract

The viral polymerase is an essential complex for the influenza virus life cycle as it performs the viral RNA transcription and replication processes. To that end, the polymerase carries out a wide array of functions and associates to a large number of cellular proteins. Due to its importance, recent studies have found numerous mutations in all three polymerase protein subunits contributing to virus host range and pathogenicity. In this review, we will point out viral polymerase polymorphisms that have been associated with virus adaptation to mammalian hosts, increased viral polymerase activity and virulence. Furthermore, we will summarize the current knowledge regarding the new set of proteins expressed from the viral polymerase genes and their contribution to infection. In addition, the mechanisms used by the virus to counteract the cellular immune response in which the viral polymerase complex or its subunits are involved will be highlighted. Finally, the degradative process induced by the viral polymerase on the cellular transcription machinery and its repercussions on virus pathogenicity will be of particular interest.

Introduction

The influenza A virus (IAV) is an important respiratory human pathogen causing yearly recurrent seasonal epidemics with an average global burden of >600 million cases (www.who.int). In rare instances IAV can also spread from its natural zoonotic reservoirs (aquatic birds) to cross species barriers and transmit to humans where it can evolve into strains that cause diseases ranging from mild to severe, with occasional widespread distribution known as pandemic. In the past century, the most devastating pandemic took place in 1918–1920 (also known as the “1918 flu” or “Spanish flu”), infecting hundreds of millions and killing between 20 and 50 million people worldwide (Johnson and Mueller, 2002, Taubenberger and Morens, 2006). Two additional pandemic events originating in Asia took place during the second half of the 20th century, the so-called Asian H2N2 pandemic of 1957 (1 million deaths) and the Hong Kong H3N2 pandemic in 1968 (1 million deaths). Although of relatively low virulence, the recent H1N1 influenza 2009 pandemic of swine origin emerged unexpectedly in Mexico to spread around the world in just a few months. IAV naturally infects wild aquatic birds making up an extremely heterogeneous population which includes many possible combinations between the two surfaces glycoproteins, the hemagglutinin (HA) and the neuraminidase (NA); a total of 18 different HA and 11 NA have been recognized. At present there are concerns that avian influenza strains, including the highly pathogenic influenza viruses H5N1 and the novel H7N9 subtypes not yet capable of spreading from one human to another, could adapt and become more easily transmissible among humans.

The ability of influenza A viruses to infect a variety of hosts is based on their genetic diversity due to two main reasons: (i) their RNA polymerase is error-prone, and (ii) they contain a segmented genome, which allows for exchange of RNA segments between genotypically diverse influenza viruses. These features lead to the rapid generation of novel strains and subtypes and thus contribute to the constant threat that newly emerging and re-emerging influenza viruses pose to the human population.

IAV, a member of the Orthomyxoviridae family, possesses a negative-sense single-stranded RNA genome (vRNA) divided into eight segments. vRNAs are protected in a structure known as viral ribonucleoprotein (vRNP), in which the RNA strand is wrapped by the nucleoprotein (NP), and a single viral polymerase complex interacts with the complementary 3′ and 5′ genomic end sequences. Recent cryo-EM reconstructions of vRNPs obtained from different sources show a double-helical stem structure in which the NP proteins and the protected viral RNA form two anti-parallel strands (Arranz et al., 2012, Moeller et al., 2012). Both strands are connected by a short loop at one end of the particle and interact with the viral polymerase at the other end. The incoming parental vRNPs are released into the cytoplasm of the infected host cell and then quickly transported into the nucleus. IAV is a rare RNA-genome virus in that expression and amplification of viral genomes take place inside the infected cell nucleus and it therefore heavily depends on host nucleocytoplasmic trafficking and nuclear functions. Within the host nucleus, cellular insoluble fractions, such as nuclear matrix and chromatin structures, have been shown to encompass part of the viral polymerase transcription and replication activity and act as a platform for the release of progeny vRNPs outside the nucleus (Chase et al., 2011, Garcia-Robles et al., 2005, Lopez-Turiso et al., 1990, Takizawa et al., 2006).

The IAV genome encodes for 10 major proteins, although alternative protein products have been characterized from several genome segments (see below). The complete coding capacity of the IAV genome is far from known and processes like splicing, protein truncations, and the use of alternative initiation codons or overlapping frames are known to increase the diversity of proteins generated during infection. The widespread synthesis of small viral non-coding RNAs with largely unknown roles in the outcome of infection further complicates the understanding of IAV full expression capacity (Perez et al., 2010, Perez et al., 2012). The three largest viral genome segments encode for the polymerase heterotrimeric complex, responsible for the RNA-dependent RNA polymerase activity of the virus (Fig. 1). The three polymerase subunits, named PA, PB1 and PB2, together with the nucleoprotein, form the minimum set of viral proteins required for viral RNA transcription and replication (reviewed in (Fodor, 2013, Resa-Infante et al., 2011)). Viral transcription depends on primers of host origin obtained through a cap-snatching process targeting newly synthesized host pre-mRNAs (Krug et al., 1979). The viral positive-sense mRNA also includes a 3′ polyA tail generated by repetitive polymerization of a polyU track on the genomic vRNP, thus creating a transcript that is structurally undistinguished from host mRNAs. The vRNP undergoing transcription is processed by a cis-acting polymerase (Jorba et al., 2009). The requirement of newly synthesized cellular pre-mRNAs as a source of cap-olignonucleotides for viral transcription initiation involves a functional association between the viral and the cellular transcription machineries, which has significant consequences for viral outcome (see below).

A switch from viral transcription to replication is required during a successful infectious cycle. The action of viral short virion RNAs (svRNAs) and the availability of cellular nucleotides and newly synthesized viral polymerase and NP, have been involved in this process (Jorba et al., 2009, Perez et al., 2010, Vreede and Brownlee, 2007, Vreede et al., 2004). The viral replication step is initiated early during infection with the synthesis of the cRNA, a complete unpolyadenylated positive-polarity copy of the vRNA (Hay et al., 1977). cRNAs within the infected host cell form cRNPs structures, and serve as templates for the synthesis of new vRNPs which are ready for either further rounds of viral gene expression or their transportation outside the nucleus and subsequent encapsidation. As opposed to what happens with the viral mRNA, the production of cRNAs is only started after viral protein synthesis has begun and seems to require a soluble trans-acting polymerase different from the one resident in the RNP (Hay et al., 1977, Jorba et al., 2009).

Basic protein 1, PB1, is the core of the complex, the most conserved of the polymerase subunits, and contains the enzymatic motifs needed for RNA polymerization activity (Kobayashi et al., 1996). The PB2 subunit has a key role in viral transcription due to its recognition and binding of host 5′ mRNA cap structures generated by the cellular transcription machinery (Blaas et al., 1982, Braam et al., 1983). Acidic protein PA has an endonucleolytic activity needed for the viral cap-snatching process (Dias et al., 2009, Yuan et al., 2009). In recent years some parts of the three proteins have been structurally characterized including important functional domains, such as the cap-binding site on PB2 (Guilligay et al., 2008), the carboxy-terminal part of PB2 that contains the nuclear localization sequence (NLS) domain and important host range determinants (Tarendeau et al., 2007, Tarendeau et al., 2008), and the endonuclease domain at the N-terminus of PA (Dias et al., 2009, Yuan et al., 2009). Additionally, crystal structures of the interactions between the carboxy-terminal part of PA and the first 15 amino-terminal amino acids of PB1 (He et al., 2008), and between the last 72 amino acids of carboxy-terminal PB1 and the first 35 amino-terminal amino acids of protein PB2 (Sugiyama et al., 2009), have been obtained. Furthermore the cryoEM reconstructions of the vRNP, as well as other structural analyses of the viral polymerase out of the context of the RNP, illustrate this heterotrimer complex as a compact structure with some indications of flexibility (Area et al., 2004, Arranz et al., 2012, Coloma et al., 2009, Moeller et al., 2012, Resa-Infante et al., 2010, Torreira et al., 2007). In agreement with that, the crystal structures of the complete RNA polymerase complexes from influenza A and B viruses show that the three subunits have multiple interactions with each other and all of them participate in the binding of the two strands of the RNA promoter (Pflug et al., 2014, Reich et al., 2014).

Overall, the viral polymerase is a major component of the virus as it provides the necessary machinery to translate its genome and replicate itself. In addition to its basic functions on viral genome expression, the viral polymerase has a key role in transmission between hosts and pathogenesis. For instance, a viral polymerase with higher activity usually correlates with an increased ability of transmission and adaptability, as well as higher pathogenicity (Naffakh et al., 2008).

In this review, we highlight those mutations in IAV polymerase that have been associated with higher virus adaptability to a new host, increased viral polymerase activity and pathogenicity. Additionally, some of the host cellular mechanisms that have been linked to these processes are discussed. Special attention is paid to the consequences derived from the functional association between viral and cellular transcription machineries as well as the detrimental effect caused by the degradative process induced by the viral polymerase upon the cellular transcription apparatus and its implications for virus pathogenicity.

While most of the basic aspects of virus replication and transcription have been known for decades, the study of host-associated factors and the cellular pathways involved during these processes represent an emerging research field. Numerous proteomic approaches and yeast two-hybrid analysis have vastly extended the number of interacting partners for the viral RNPs and polymerase components (Bradel-Tretheway et al., 2011, Jorba et al., 2008, Mayer et al., 2007, Munier et al., 2013, Shapira et al., 2009, Tafforeau et al., 2011). Further, several RNAi-based screenings have uncovered numerous host proteins involved in the IAV replication cycle (Brass et al., 2009, Hao et al., 2008, Karlas et al., 2010, Konig et al., 2010, Shapira et al., 2009). All these studies describe a complex and dynamic interrelationship between the virus and its host, including associations between the viral polymerase and several cell structures and components, such as the mitochondria, the nuclear pore and nucleocytoplasmic transport machinery, signaling pathways, protein translation components, RNA binding and splicing related proteins, and cellular RNA polymerase subunits and accessory factors. Although most of the RNAi-based screenings use common H1N1 or H3N2 laboratory virus strains infecting a particular mammalian cell line, other studies have been carried out pinpointing polymerase-associated host factors that show virus-strain and host-cell specificity functions (Bortz et al., 2011, Gabriel et al., 2011, Resa-Infante et al., 2008). The different association of viral proteins to host cell factors depending on the virus strain and type and origin of the infected cell, combined with host response analysis through transcriptome and proteome approaches, contribute to a better understanding of the complex mechanisms involved in virus host adaptability and pathogenicity.

Few viral polymerase-associated host proteins have been mechanistically described in some detail. Polymerase interacting partners known to facilitate virus RNA expression include: the minichromosome maintenance (MCM) complex, a cellular DNA replication fork helicase that interacts with the PA subunit of the viral polymerase and promotes the elongation of nascent cRNA in vitro (Kawaguchi and Nagata, 2007); the heat shock protein (Hsp) 90 and the nuclear import RanBP5 and Importin-α proteins which play important roles in the assembly and nuclear import of the viral polymerase (Deng et al., 2006, Momose et al., 2002, Resa-Infante et al., 2008); and the RNA polymerase II (RNAPII) and transcription related proteins CDK9, SFPQ/PSF, CLE/C14orf166, RED, and DDX17 thought to play important roles in several steps of the viral RNA synthesis (Bortz et al., 2011, Engelhardt et al., 2005, Fournier et al., 2014, Huarte et al., 2001, Landeras-Bueno et al., 2011, Perez-Gonzalez et al., 2006, Rodriguez et al., 2011, Zhang et al., 2010). On the contrary, several cellular proteins that interact with the viral polymerase and have repressive roles in its activity have also been characterized. Among these are the protein chaperone Hsp70, the dsRNA-binding protein NF90 and the retinoblastoma-like protein RBL2, all known to interact with components of the viral RNPs and block virus transcription and replication activities (Kakugawa et al., 2009, Li et al., 2011, Wang et al., 2009); HAX1, a protein with antiapoptotic activity that interacts with the PA subunit in the cytoplasm and blocks its entry into the nucleus (Hsu et al., 2013); and CHD6, a cellular transcription related chromatin remodeling protein that acts as a suppressor of the viral polymerase transcription and replication activities (Alfonso et al., 2013). The interaction between IAV proteins and specific host proteins can also involve a complex dynamic interplay as shown for cellular DDX21, an RNA helicase implicated in innate immunity that binds to PB1 and inhibits polymerase assembly at early stages of infection (Chen et al., 2014, Zhang et al., 2011). However, at later times the viral non-structural protein 1 (NS1) overcomes this inhibition by binding to DDX21 and displacing PB1. Thus, it seems that by using these sequential interactions IAV not only overcomes a cellular restriction, but also temporally regulates viral gene expression (Chen et al., 2014). Collectively, there are a large number of host factors that cooperate to modulate infection severity and outcome, and the viral polymerase plays a crucial role as a major contributor to viral-host cell interactions.