Transcription and replication mechanisms of Bunyaviridae and Arenaviridae L proteins

Abstract

Bunyaviridae and Arenaviridae virus families include an important number of highly pathogenic viruses for humans. They are enveloped viruses with negative stranded RNA genomes divided into three (bunyaviruses) or two (arenaviruses) segments. Each genome segment is coated by the viral nucleoproteins (NPs) and the polymerase (L protein) to form a functional ribonucleoprotein (RNP) complex. The viral RNP provides the necessary context on which the L protein carries out the biosynthetic processes of RNA replication and gene transcription. Decades of research have provided a good understanding of the molecular processes underlying RNA synthesis, both RNA replication and gene transcription, for these two families of viruses. In this review we will provide a global view of the common features, as well as differences, of the molecular biology of Bunyaviridae and Arenaviridae. We will also describe structures of protein and protein-RNA complexes so far determined for these viral families, mainly focusing on the L protein, and discuss their implications for understanding the mechanisms of viral RNA replication and gene transcription within the architecture of viral RNPs, also taking into account the cellular context in which these processes occur. Finally, we will discuss the implications of these structural findings for the development of antiviral drugs to treat human diseases caused by members of the Bunyaviridae and Arenaviridae families.

Keywords: Antivirals; Arenavirus; Bunyavirus; L protein; Replication; Transcription.

Fig. 1

Genome structure of Bunyaviridae and Arenaviridae. The genome segment for the five genus of the Bunyaviridae and Arenaviridae families are shown. The size of each genome segment is indicated by the upper 1Kb rule. The regions containing the ORFs are indicated by green arrows for those proteins whose mRNA is transcribed from the vRNA and orange for those transcribed from the cRNA. The red triangles indicate the location of the protease cleavage sites within the indicated polyproteins and the red spots indicate the location of transcription termination signals. The conserved 3′ (red) and 5′ (blue) termini vRNA sequences are indicated on the right, and show the Watson-Crick base pairing complementary. Conserved non complementary bases are written in black.

Fig. 2

Polymerase active site motifs of Bunyaviridae and Arenaviridae L proteins. Structure of the core of the polymerase domain (PDB code: 5amr) of LACV with the five motifs: G (cyan), F (orange), A (green),B (blue),C (red),D (purple), E (pink). Surrounding the structure are presented the weblogos of each motif, on the top are the Bunyaviridae motifs (numbering based on LACV) and in the bottom the Arenaviridae motifs (numbering based on LASV). Despite similarities in the catalytic site and surrounding areas, the overall broad divergence of the L protein sequences prevents to derive a reasonable structural model of Arenaviridae polymerase based on the crystal structured determined for LACV polymerase.

Fig. 3

LACV L protein crystal structure. Three views of the structure of the L protein first 1750 aa residues are shown in complex with viral RNAs represented in cartoons with the RdRpol domain in wheat, the core N-terminal extension in pink (within the vRNA binding lobe highlighted in violet), the core C-terminal extension in yellow and the hanging EN in green. A more detailed description of the structure and domains within the core C-terminal and N-terminal extensions is found in Gerlach et al. (2015). The vRNAs are coloured in red (3′) and blue (5′). The inner cavities are represented by green spheres as described in Gerlach et al. (2015). A, view of the polymerase clearly showing the four tunnels for the template exit and entry, the NTP entry and the product exit, converging in a central cavity where the RNA synthesis occurs. Core C-terminal and the RdRpol appear delimiting the cavity and tunnels. The cap-snatching EN is hanging from the core attached through a flexible link. B, rotating 90° the structure to the right is shown the NTP entrance close to the 5′ RNA binding site, the RdRpol and core N-terminal extensions appear delimiting the cavities and the NTP entry site. In violet is shown the vRNA binding lobe. C, rotating again the structure 90° towards the top is shown the vRNA binding lobe which is binding the 3′ RNA in one side, with a clamp insertion trapping the RNA, and 5′ RNA in the other, with an arch insertion partially covering the RNA.

Fig. 4

Structural alignment of the L proteins cap-snatching endonucleases. The crystal structures of cap-snatching ENs belonging to mammarenavirus, orthobunyavirus and hantavirus. LCMV (PDB code: 3jsb, light brown); LASV (PDB code: 5j1n, pink); PICV (PDB code: 4i1t, cyan); LACV (PDB code: 2xi5, salmon); Hantaan virus (PDB code: 5ize, light green). Structures are represented in cartoon and residues of the catalytic site of LASV and Hantaan virus are shown in sticks with their respective manganese ions (coloured magenta and red respectively). The three conserved regions (His/Asp; LOOP; PD(ED)K motif) are shown on the side with Clustal color code scheme. Conserved residues are shown in bar diagrams under the alignment. For the His/Asp motif top letters and bottom red stars locate the key residues of the motif. For the PD(ED)K motif top letters locate the key residues of the motif. Side caption show a zoomed view of the catalytic site.

Fig. 5

Structural organization of L proteins. Schematic superposition of L proteins belonging to each genus of the Bunyaviridae and Arenaviridae families. The structural organization of the LACV L protein (top) and influenza heterotrimeric polymerase (bottom) derived from their respective crystal structures is represented like in Gerlach et al., 2015 and Pflug et al., 2014. The polymerases are aligned with respect to the conserved polymerase motifs shown in Fig. 3 and the conserved PD. (E/D)XK motif. The residue numbers for the polymerases correspond to LACV (orthobunyavirus), RVFV (phlebovirus), Hantaan (hantavirus), Tomato spotted wilt virus (tospovirus), CCHFV (nairovirus), Lassa (mammarenavirus) and Golden gate virus (reptarenavirus). The length (amino acid residues) for each polymerase is indicated in their upper right side. The distances between conserved motifs and characterised domains are indicated for LACV (Gerlach et al., 2015) and the tospo- and nairovirus indicating the length of the insertions within the dashed lines stretches. Since the EN domains have not been isolated yet for tospo- and nairovirus the distances between the ENs and polymerase motifs or OTU domain are estimated values taking into account the conserved PD. (E/D)XK motif and the approximate size of EN domains (200aa) and are written in italics. The scheme illustrates the overall architecture of L proteins and the predicted positions of the more relevant insertions for the larger tospo- and nairovirus polymerases.

Fig. 6

NMR structure of Lassa virus Z protein. Z protein (PDB code: 2M1S) in ribbon surrounded by weblogos corresponding to motifs conserved across Arenaviridae family members. N-terminus and C-terminus are labelled N and C respectively. The protein has a central globular RING domain extended in both directions by disordered regions. The central RING domain has three conserved motifs (coloured in red, orange and green). The N-terminal region embed a conserved G motif (yellow) critical for myristoylation and membrane attachment, the C-terminal region contains proline rich late domain motifs (blue), which are variable in length and composition. Alignments of Z mamma- and reptarenavirus amino acid sequences was done by T-coffee including 37 Z protein sequences.

Fig. 7

Schematic representation of differences between transcription and replication processes carried out by the L proteins. From the preinitiation state the L protein starts transcription (right side) or replication (left side). Transcription initiation occurs by cap-snatching where RNA synthesis is primed by a capped cellular mRNA shortened to 12–18 or 4–5 nucleotides (for bunyaviruses and arenaviruses, respectively) by the EN (not shown in the scheme). In Influenza, the cap-binding domain is the responsible for directing the capped RNA towards the active site, the presence of a cap-binding domain in L proteins is still non proven. Likewise, it remains to be determined whether cellular factors are also necessary for transcription activation. The requirement of active translation for viral transcription elongation strongly suggest the coupling between the viral L protein and the host cell translational machinery in bunyaviruses. During transcription the polymerase stops at the transcription termination signal, because of its closed cage structure the polymerase cannot detach from the RNA template and would need to slide to the end of the genome to start another replicative or transcriptional round. The replication starts de novo (left side) and can need a helper polymerase that would interact with the nascent 5′ strand and coordinate the RNA synthesis with the RNP assembly. The L protein does not stop when reading the transcription termination signal and can continue the replication until the end of the genome to synthetize the complete cRNA.