One of the two key domains of the Lassa Virus Nucleoprotein is the C terminal Domain. This domain characterised by residues 341-569 is believed to contain a 3-5 exonuclease function. The immunosuppressive properties of the Lassa virus is believed to stem from this exonuclease function of the NP.
One key feature of the LV’s pathogenicity is its ability to suppress the innate immune system of its target. It is suggested that this C domain exonuclease is responsible for inhibiting the nuclear translocation of IFN regulatory factor 3 (IRF-3) and subsequently preventing the expression of IFNa/b. These cytokines are crucial in initiating an antiviral response by the infected cells and as a result of their repression viral replication can proceed unimpeded. The exact mechanism by which this is carried out is yet to be determined. Xiaoxuan et al postulate that the LASV NP C domain is responsible for specifically degrading the viral PAMP (Pathogen-Associated Molecular Patterns) RNA ligands. This may prevent recognition of viral RNA by PRRs (Pattern Recognition Receptors) such as TLR-3 and this may lead to the virus remaining undetected.
One key feature of the LV’s pathogenicity is its ability to suppress the innate immune system of its target. It is suggested that this C domain exonuclease is responsible for inhibiting the nuclear translocation of IFN regulatory factor 3 (IRF-3) and subsequently preventing the expression of IFNa/b. These cytokines are crucial in initiating an antiviral response by the infected cells and as a result of their repression viral replication can proceed unimpeded. The exact mechanism by which this is carried out is yet to be determined. Xiaoxuan et al postulate that the LASV NP C domain is responsible for specifically degrading the viral PAMP (Pathogen-Associated Molecular Patterns) RNA ligands. This may prevent recognition of viral RNA by PRRs (Pattern Recognition Receptors) such as TLR-3 and this may lead to the virus remaining undetected.
The crystal structure of the C terminal domain of the NP has been resolved to 1.5Å (Hastie et al 2010). Both Zinc and manganese ions were found to be present within the structure. (The conditions required for crystallisation are given in Methods.) A depiction of the LASV NP D340 domain is seen below:
 |
Figure 1. The C domain of LASV NP (LASV NP D340). The Mn2+ ion is shown as a magenta sphere. This is coordinated in the putative exonuclease catalytic site. A Zn2+ is also found (cyan sphere) deeper within the protein. |
The next image reveals the structure comprised of 5 stranded b sheets with an anti-parallel strand and six a helices linked by a series of loops. These a helices are associated in three sets: (a2), (a5 + a6) and (a1 + a3 + a 4) and these surround the b sheet forming an a/b/a sandwich structure:
 |
Figure 2. The cartoon structure of LASV NP D340. Once again showing both Zinc and Manganese ions |
Despite the apparent lack of a archetypal zinc finger motif a Zn atom is identified in the structure and is found to be coordinated by residues E399, C506, H509 and C529 as shown below:
 |
Figure 3. An image depicting the residues (blue) involved in the coordination of the Zinc atom within LASV NP D340. The image also shows the proximity of the Zn ion to the active site (red residues) and associated Manganese suggesting the importance of the Zinc ion to the local structure at the exonuclease active site. |
The coordination of the second ion (Mn2+) is depicted below. Residues D389, E391 and D533 are key to the positioning of the Mn ion within the active site:
 |
| Figure 4. An image showing the coordination of the Mn ion by residues (red) E391, D389 and D533. The zinc ion is also visible behind the site of the Mn ion. The two darker red residues show the other two residues of the putative catalytic site not involved in Mn coordination. |
 |
| Figure 5. The C domain catalytic active site. This image focuses in on 5 residues of the C domain involved in the riboexonuclease activity. |
Having elucidated the structure of the LASV NP bioinformatics tools were used to determine possible roles of the protein. BLAST analysis yielded no significant similarity to other known protein sequences outside of the arenavirus family. However DALI and HHpred searches based upon 3D structure and sequence alignment respectively yielded unexpected results. The C terminal portion of the protein was found to share a similar structure to the family of DEDD exonucleases. More accurately the C domain of the NP was found to resemble a DEDDh exonuclease. This subfamily of exonucleases are known to contain a catalytic site consisting of the conserved sequence Asp-Glu-Asp-Asp-His. An rmsd of ~3Å was found between LASV NP and these other known exonucleases highlighting the striking similarity of the NP to other exonucleases. Perhaps most significantly it was determined that residues D389, E391, D533 and H528 almost exactly aligned with D1, E2, D4 and H that form 4 out of the 5 catalytic residues in other DEDDh enzyme. These include the e subunit of E.coli DNA polymerase III and the TREX 1 exonuclease, the major 3’—5’ exonuclease in humans. Based on these structural comparisons alone it is evident that the role of this portion of the nucleoprotein was likely to involve a 3’—5’ exonuclease. Studies by (Yan, N et al 2010) have shown that TREX 1 exonuclease plays a role in suppressing the innate immune response to HIV infection giving further evidence to the belief that the C terminal domain of Lassa virus NP may be the important structure in immune suppression.
The structure of the putative catalytic site is shown below. (n.b. Only one Mn2+ was found to present in the active site. This partially occupied active site is consistent with other known RNase enzymes such as RNase D in which only one of the two catalytically functional metal ions are present in the absence of the ribonucleotide substrate.)
The catalytic site is a negatively charged cavity comprised of residues D389, E391, D466, D533 and H528. Comparisons between the LASV NP D340 and other DEDD exonucleases show the similarity between the active sites of the enzymes. The example given below is eukaryote TREX1. This protein has a well characterized 3’—5’ exonuclease activity and is a member of the DEDDh exonuclease family:
 |
Figure 6. The exonuclease catalytic site of eukaryote TREX1. Note the same amino acid residues are conserved as in the LASV NP D340. (Asp18/Glu20/Asp130/Asp200/His195) (n.b. the Mn ion was not co crystallised with this structure. PDB ID 3MXJ) |
Experimental evidence for function of LASV NP D340 as a 3’—5’ exonuclease and its role in immune suppression:
Structural analysis strongly suggests that LASV NP D340 is an exonuclease and has a role in the suppression of the innate immune system. To substantiate these beliefs both activity assays and site directed mutagenesis studies were carried out. (Hastie et al 2010) (Qi et al 2010)
An in vitro functional analysis was carried out to determine the exact specificity and directionality of the LASV NP D340. A range of nucleic acid including ssDNA, dsDNA, ssRNA and dsRNA were incubated with purified LASV NP D340 and results were viewed using denaturing PAGE and autoradiography (see Methods for more details). It was found that LASV NP D340 preferentially hydrolysed 18bp dsRNA oligonucleotides. No hydrolysis of ssRNA, ssDNA or dsDNA was observed suggesting specificity of LASV NP D340 for dsRNA. A time course experiment involving dsRNA with a labeled 5’ end showed that the directionality of the enzyme was 3’—5 as the 5’ label remained throughout the course of the experiment. Experiments have also proved that LASV NP D340 can digest both blunt ended and overhang dsRNA.
Site directed mutagenesis studies were carried out to determine residues key to the structure and function of the LASV NP D340. Alanine mutations at any of the DEDDh residues were all found to inhibit exonuclease activity (e.g. E391 --> E391A). Site directed mutagenesis on any of the residues found to coordinate the Zn atom were also found to prevent exonuclease activity (e.g. H509 --> H509A). Incubation with EDTA was also found to block all exonuclease activity suggesting that the Zn and Mn atoms are crucial to function. Xiaoxuan et al suggest that the Zn may be required ‘to stabilise the structure of the C domain and/or contribute to the substrate binding and specificity of the exonuclease activity’.
Results from an activity assay carried out by Qi et al show the effects of a few alanine mutations of the key catalytic residues. The PAGE electrophoresis results show that alanine mutations at E391, D466 and D389 prevent all exonuclease activity on ssRNA oligomers. The absence of bands in lanes "30 min" & "60 min" suggest that the LASV NP (trimer) exonuclease has cleaved the RNA into sequences too short to be seen on the gel. The sample in lane 3 (control 2) was prepared in the presence of EDTA. The presence of all the bands shows that EDTA prevents exonuclease activity by chelating the crucial metal ions.
 |
| Figure 7. Results from an PAGE electrophoresis experiment showing the activity of the LASV NP exonuclease activity. Image courtesy of Xiaoxuan et al 2010. Cap binding and Immune Evasion revealed by Lassa nucleoprotein strucutre. Supplementary Information Nature Vol 468 pg 12 Dec 2010 |
Evidence for the role of LASV NP D340 in immune suppression was also found using site directed mutagenesis (Hastie et al 2010). Mutations that inhibit exonuclease activity were also found to prevent the suppression of IRF-3 nuclear translocation using a luciferase construct under the control of an IRF-3 dependant promoter. This corroborates with the theory that the exonuclease activity of LASV NP D340 is responsible for immune suppression. The rest of the LASV NP was not found to be required for either the exonuclease or immuno suppressive properties of the NP.