Interaction between Hantavirus Nucleocapsid Protein (N) and RNA-Dependent RNA Polymerase (RdRp) Mutants Reveals the Requirement of an N-RdRp Interaction for Viral RNA Synthesis

Abstract

Viral ribonucleocapsids harboring the viral genomic RNA are used as the template for viral mRNA synthesis and replication of the viral genome by viral RNA-dependent RNA polymerase (RdRp). Here we show that hantavirus nucleocapsid protein (N protein) interacts with RdRp in virus-infected cells. We mapped the RdRp binding domain at the N terminus of N protein. Similarly, the N protein binding pocket is located at the C terminus of RdRp. We demonstrate that an N protein-RdRp interaction is required for RdRp function during the course of virus infection in the host cell.

TEXT

Hantaviruses are negative-strand emerging RNA viruses and members of the Bunyaviridae family. Humans get hantavirus infection by inhalation of aerosolized excreta from infected rodent hosts (1–4). Hantavirus infection causes hemorrhagic fever with renal syndrome (HFRS) and hantavirus cardiopulmonary syndrome (HCPS), with mortalities of 15% and 50%, respectively (5, 6). The spherical hantavirus particles (7) harbor three genomic RNA segments (S, M, and L) within a lipid bilayer (8), encoding viral nucleocapsid protein (N protein), viral RNA-dependent RNA polymerase (RdRp), and glycoprotein precursor (GPC), respectively. The GPC is posttranslationally cleaved at a highly conserved WAASA motif, generating two glycoproteins, Gn and Gc (9). N protein is multifunctional, primarily involved in encapsidation of the viral genome. However, recent studies have suggested that N protein plays diverse roles in the virus replication cycle. The viral ribonucleocapsids are used as the template by viral RdRp for the synthesis of viral mRNA and replication of the viral genome (10, 11). However, it is still a mystery how RdRp accesses the viral RNA (vRNA) genome, which is buried inside the compact nucleocapsid structure. Moreover, numerous reverse genetic systems have proven that RdRp from negative-strand RNA viruses requires assistance from N protein while performing its function inside the host cell (12–16). Based on these observations, we asked whether N protein directly interacts with RdRp and regulates its function.

To answer this question, we cloned the gene encoding the Sin Nombre virus (SNV) RdRp in the pcDNA 3.1 (+) backbone containing either a green fluorescent protein (GFP) tag at the N terminus or a His tag at the C terminus. The expression construct was transfected into human umbilical vein endothelial cells (HUVECs) and HeLa and HEK293T cells for expression. An examination of the cell lysate by Western blot analysis did not show the expression of RdRp, which was due to lack of RdRp expression and not due to poor antibody reactivity. The gene was next cloned in pFastBac vector to examine the expression in Sf9 insect cells, using a baculovirus expression system. However, multiple trials revealed that RdRp was not expressed in insect cells as well (not shown). Hantavirus RdRp is a large protein of 250 kDa. To our knowledge, its structure is not known, although two functional domains have been predicted by in silico studies. Similar to other segmented negative-strand RNA viruses, the hantavirus RdRp initiates transcription by a unique cap-snatching mechanism (17–21). Based on in slico analysis, the regions around amino acids 1 to 250 and 751 to 1290 have been proposed to constitute the cap-snatching endonuclease domain and catalytic domain, respectively (22, 23). The regions from amino acids 251 to 750 and 1291 to 2153 have not been assigned any function (Fig. 1A) (22). However, the analysis by a domain prediction software that identifies conserved protein domains suggested the catalytic domain is slightly bigger and corresponds to the region from amino acids 562 to 1286. The upstream intervening sequence between the endonuclease domain and the catalytic domain constitutes the region from amino acids 238 to 562 (Fig. 1A). Since wild-type RdRp was not expressed, we chose to individually express the predicted domains of RdRp in mammalian cells and examine their interaction with N protein. We generated constructs expressing the N-terminal endonuclease domain (pol1–250), intervening region of unknown function (pol251–752), catalytic domain (pol751–1290), and the C-terminal region of unknown function (pol1291–2153) (Fig. 1A). Due to ambiguity in the domain prediction, two additional constructs expressing the intervening region (pol238–562) and catalytic domain (pol562–1286) were also generated (Fig. 1A). All of these RdRp fragments contained a C-terminal His tag. HEK293T cells were cotransfected with plasmids expressing the RdRp fragment of interest along with C-terminally myc-tagged N protein. An examination of whole-cell lysates (WCL) by Western blot analysis showed that all RdRp fragments were expressed in HEK293T cells, except the pol1–250 fragment. In addition, the pol1291–2153 fragment was truncated from the N terminus (Fig. 1B). The Western blot analysis of whole-cell lysates using anti-myc antibody revealed the similar expression of N protein in cotransfected HEK293T cells, except for the cells coexpressing N protein along with the pol1–250 fragment. Based on these observations, it is clear that the N-terminal endonuclease domain does not accumulate in cells, and it also inhibited the accumulation of N protein in cotransfected cells. To determine whether the pol1–250 fragment specifically inhibited N protein expression, we cotransfected HEK293T cells with plasmids expressing the pol1–250 fragment along with another plasmid expressing either GFP or the luciferase reporter. Interestingly, we observed that the endonuclease domain nonspecifically inhibited the expression of both reporters (not shown). An examination of whole-cell lysates (WCL) by Western blot analysis revealed that expression of the pol1–250 fragment also impacted the expression of endogenous proteins, such as GAPDH and tubulin (Fig. 1B). Recently, it has been suggested that the endonuclease domain of Andes virus RdRp cleaves the host cell transcripts and also its own mRNA and thereby regulates its own as well as host gene expression (24). It is possible that endonuclease activity maintains the lower steady-state levels of RdRp in virus-infected cells, which is required for efficient virus replication. As mentioned later in this article, we expressed and purified the pol1–250 fragment in Escherichia coli. The purified fragment cleaved both capped and uncapped mRNAs nonspecifically in vitro (not shown). Moreover, overexpression of the pol1–250 fragment showed cytotoxicity in HeLa cells (Fig. 1C). Based on these observations, it is likely that endonuclease activity of the pol1–250 fragment suppressed its own as well as N protein expression in HEK293T cells (Fig. 1B).

FIG 1.

Interaction of N with RdRp. (A) Pictorial representation of RdRp mutants used in this study. A conserved domain prediction software from NCBI (http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi) suggested that the catalytic domain is located in the region from amino acids 562 to 1286 and the upstream intervening region of unknown function corresponds to the region from amino acids 238 to 562. (B) HEK293T cells grown on 60-mm-diameter dishes were cotransfected with a plasmid expressing myc-tagged wild-type (w.t.) N protein along with another plasmid expressing the His-tagged RdRp fragment of interest (pol1–250, pol251–752, pol751–1290, pol1291–2153, pol238–562, or pol562–1286). Cells were lysed 48 h posttransfection, the resulting cell lysates were immunoprecipitated (IP) with anti-myc tag antibody, and immunoprecipitated material was examined by Western blot (WB) analysis using either anti-His tag antibody (panel i) or anti-myc tag antibody (panel ii). The light chain of anti-myc antibody is shown in panel iii. Cell lysates were also incubated with Ni-NTA beads, and the eluted material from washed beads was examined by Western blot analysis using anti-myc tag antibody (panel iv). Equal volumes of whole-cell lysates (WCL) were examined by Western blot analysis using either anti-His tag antibody (panel v) or anti-myc tag antibody (panel vi) or anti-GAPDH antibody (panel vii) or antitubulin antibody (panel viii). The band intensity for β-actin and tubulin was quantified and normalized to the last band from the left; the intensity is indicated at the bottom. Note that the plasmids used for transfection are shown in Table 1. (C) HeLa cells in a 96-well plate at 70% confluence were transfected with 2.5 μg of a plasmid expressing either the pol1–250 fragment or pol 238–562 fragment. Cell viability was measured 48 h posttransfection using the CellTox green cytotoxicity assay. The experiment was performed in duplicates. (D) Both pol1–250 and pol1291–2153 fragments were expressed in E. coli and purified using Ni-NTA beads. HEK293T cells were transfected with a plasmid (Table 1) expressing wild-type N protein. Cells were lysed 48 h posttransfection, and the resulting lysates were incubated with 2 μg of the purified RdRp fragment of interest at 4°C for 3 h. The lysates were immunoprecipitated with anti-myc tag antibody, and the immunoprecipitated material was analyzed by Western blotting using anti-His tag antibody to detect the RdRp fragments. (E) Huh7 cells in a 60-mm dish were infected with SNV at an MOI of 2. Twenty-four hours postinfection, cells were transfected with 2.5 μg of a plasmid (Table 1) expressing the Pol1291–2153 fragment. Cells were lysed, the resulting lysates were immunoprecipitated with polyclonal anti-N antibody, and the immunoprecipitated material was analyzed by Western blot analysis using anti-His antibody to detect the pol1291–2153 fragment (bottom). The cell lysates were also incubated with Ni-NTA beads, and the eluted material from washed beads was analyzed by Western blotting using anti-N antibody (top).

To determine whether N protein interacts with any of the RdRp fragments, the cell lysates were immunoprecipitated with anti-myc antibody, and immunoprecipitated material was examined by Western blot analysis using anti-His tag antibody. It is evident from Fig. 1B that only the pol1291–2153 fragment copurified with N protein. Interestingly, the cleavage product of this fragment also copurified with N protein. Based on the size of the cleavage product, it is likely that the N protein binding domain is located at the C-terminal 400 amino acids of RdRp, although this needs further verification. To confirm these observations, cell lysates were incubated with Ni-nitrilotriacetic acid (NTA) beads, and the eluted material from washed beads was examined by Western blot analysis using anti-myc tag antibody. This experiment further verified the interaction between the pol1291–2153 fragment and N protein (Fig. 1B). Since the pol1–250 fragment was not expressed in transfected cells, we expressed and purified both pol1–250 and pol1291–2153 fragments in E. coli using Ni-NTA beads. Cell lysates from HEK293T cells expressing C-terminally myc-tagged N protein were incubated with either the purified pol1–250 or pol1291–2153 fragment. The mixture was immunoprecipitated with anti-myc antibody, and immunoprecipitated material was examined by Western blot analysis using anti-His tag antibody. It is evident from Fig. 1D that unlike the pol1291–2153 fragment, the pol1–250 fragment did not bind to N protein. Also, the purified pol1291–2153 fragment from E. coli did not contain the truncation product, suggesting that truncation either specifically occurs in mammalian cells or the truncation product is unstable in E. coli or was lost during the purification process.

To reexamine the interaction between N protein and the C-terminal uncharacterized fragment of RdRp in hantavirus-infected cells, we infected Huh7 cells with SNV at a multiplicity of infection (MOI) of 2.0, followed by transfection with a plasmid expressing the C-terminally His-tagged pol1291–2153 fragment. Cell lysates were incubated with Ni-NTA beads. An examination of the eluted material from Ni-NTA beads using Western blot analysis confirmed the interaction between N protein and the pol1291–2153 fragment (Fig. 1D, top panel). To further confirm this interaction in virus-infected cells, the cell lysates were immunoprecipitated with anti-N protein antibody, and immunoprecipitated material was examined by Western blot analysis using anti-His tag antibody. It is evident from Fig. 1E (bottom panel) that N protein interacts with the pol1291–2153 fragment in virus-infected cells.

To directly visualize the interaction between N protein and the pol1291–2153 fragment, we fused GFP and mCherry reporters at the N terminus of N protein and C terminus of the pol1291–2153 fragment, respectively. Similarly, mCherry was also fused at the C terminus of the pol238–562 fragment and used as a negative control. HeLa cells were cotransfected with plasmids expressing either the GFP-N fusion protein along with another plasmid expressing either the pol1291–2153 or pol238–562 fragment fused to mCherry. Cells were examined under a confocal microscope. As shown in Fig. 2A, both RdRp fragments and N protein showed the perinuclear punctate morphology. The green and red puncta representing N protein and the pol1291–2153 fragment strongly colocalized with each other (Fig. 2A). In comparison, we did not observe a noticeable colocalization between N protein and the pol238–562 fragment (Fig. 2A). These results are consistent with the specific binding of N protein to the pol1291–2153 fragment. To visualize the interaction between N protein and the pol1291–2153 fragment in virus-infected cells, we infected Vero E6 cells with SNV at an MOI of 2.0. Thirty-six hours postinfection, cells were transfected with a plasmid expressing the pol1291–2153 fragment fused to mCherry, as mentioned above. N protein was visualized in virus-infected cells by confocal microscopy using a fluorescein isothiocyanate (FITC)-conjugated antibody. As shown in Fig. 2B, N protein showed a rod-shaped perinuclear morphology, presumably representing the virion RNPs. Again, a noticeable colocalization between the pol1291–2153 fragment and N protein was observed. The expression of RdRp fragments pol238–562 or pol1291–2153 does not seem to influence the N protein morphology in virus-infected cells. These results clearly demonstrate that N protein interacts with the C-terminal uncharacterized fragment of RdRp.

FIG 2.

Confocal imaging of N and RdRp fragments. (A) HeLa cells were grown on a coverslip up to 70% confluence in a 35-mm-diameter dish. Cells were cotransfected with a plasmid expressing GFP-N fusion protein along with another plasmid expressing either the pol238–562 or pol1293–2153 fragment fused with mCherry. Thirty-six hours posttransfection, cells were fixed with 3.7% paraformaldehyde (PFA) and visualized with a Nikon Eclipse C1si confocal microscope. Due to the large image size, all images in each row were stacked, and the same region was excised for presentation in this figure. DAPI, 4′,6-diamidino-2-phenylindole. (B) We next examined the colocalization of N protein with pol238–562 and pol1291–2153 fragments in virus-infected cells. Vero E6 cells were grown on a coverslip as mentioned above and infected with SNV at the MOI of 2. Thirty-six hours postinfection, cells were transfected with 2.5 μg of plasmid expressing either the pol238–562 or pol1293–2153 fragment fused to mCherry. Eighteen hours posttransfection, cells were fixed and examined under confocal microscope. N protein was visualized using anti-N primary antibody and a secondary antibody conjugated with FITC. In the control experiment, rat IgG was used instead of anti-N primary antibody, and the secondary antibody was conjugated with FITC, which did not generate any signal (not shown). This demonstrates the specificity of anti-N antibody in this assay. The bar in the right bottom panel is 10 μm.

To identify the RdRp binding domain in the N protein, we used a panel of both the C- and N-terminal deletion mutants of the N protein (Fig. 3A). All of these mutants contained a C-terminal His tag. We also modified the pol1291–2153 expression construct (Fig. 1A) by replacing the C-terminal His tag with a FLAG tag (Table 1). HEK293T cells were cotransfected with a plasmid expressing the FLAG-tagged pol1291–2153 fragment along with another plasmid expressing the N mutant of interest. An examination of whole-cell lysates (WCL) by Western blot analysis using anti-N protein antibody revealed that expression levels of few N mutants were weaker than those of other ones (Fig. 3B), although differential recognition of N protein mutants by anti-SNV N antibody cannot be ruled out. However, the levels of expression of the FLAG-tagged pol1291–2153 fragment in whole-cell lysates were relatively similar in all samples. Cell lysates were immunoprecipitated by anti-FLAG tag antibody, and immunoprecipitated material was examined by Western blot analysis using anti-N protein antibody. As shown in Fig. 3B, that deletion of up to 252 amino acids from the C terminus (N1–175 mutant) did not abrogate the interaction between N protein and pol1291–2153 fragment. In comparison, the deletion of just 50 amino acids from the N terminus of N protein abrogated the interaction. To further confirm these results, we incubated the cell lysates with Ni-NTA beads, and the bound material eluted from washed beads was examined by Western blot analysis using anti-FLAG antibody. It is evident that unlike the N-terminal deletion mutants, all of the C-terminal deletion mutants bound to FLAG-tagged pol1291–2153 fragment, consistent with similar observations from the immunoprecipitation experiment. Taken together, these results suggest that the first 50 amino acids at the N terminus of N protein constitute the RdRp binding domain.

FIG 3.

The RdRp binding domain is located at the N terminus of N. (A) Pictorial representation of N deletion mutants used in this work. The deleted regions are not shown. (B) HEK293T cells grown on 60-mm-diameter dishes were cotransfected with a plasmid (Table 1) expressing FLAG-tagged pol1291–2153 fragment along with another plasmid expressing the His-tagged wild type (w.t.) or N mutant (N51–428, N101–428, N151–428, N176–428 N231–428, N1–402, N1–346, N1–237, or N1–175). Forty-eight hours posttransfection, cells were lysed, the resulting cell lysates were immunoprecipitated (IP) with anti-FLAG antibody, and the immunoprecipitated material was examined by Western blotting (WB) using either anti-SNV N antibody to detect N mutants (panel i) or anti-FLAG antibody to detect pol1291–2153 fragment (panel ii). The heavy chain of anti-FLAG antibody is shown in panel iii. The cell lysates were also incubated with Ni-NTA slurry (Qiagen), and the eluted material from washed beads was examined by Western blotting using anti-FLAG antibody to detect the pol1291–2153 mutant (panel iv). Equal volumes of whole-cell lysates (WCL) were also examined by Western blot analysis using either anti-SNV N antibody (panel v) or anti-FLAG tag antibody (panel vi). Note that the plasmids used for transfection are shown in Table 1.

TABLE 1.

Constructs used in this study^a

Plasmid	Sequence of primer		Backbone
	Forward	Reverse
pN51-428-His	CATGCCATGGTGTCTGCATTGGAGACCAAACTCG	TGGTGGTGCTCGAGTTTAAGTGGTTCTTGGTTAGAGATTTCC	pTrix1.1
pN101-428-His	CATGCCATGGTCCTTGATGTAAATTCCATTGACT	TGGTGGTGCTCGAGTTTAAGTGGTTCTTGGTTAGAGATTTCC	pTrix1.1
pN151-428-His	CATGCCATGGAAAATAAGGGAACAAGAATCCGATT	TGGTGGTGCTCGAGTTTAAGTGGTTCTTGGTTAGAGATTTCC	pTrix1.1
pN176-428-His	CATGCCATGGGACATCTATATGTTTCTATGCCAAC	TGGTGGTGCTCGAGTTTAAGTGGTTCTTGGTTAGAGATTTCC	pTrix1.1
pN231-428-His	CATGCCATGGATTGGATGGAAAGGATTGATGACT	TGGTGGTGCTCGAGTTTAAGTGGTTCTTGGTTAGAGATTTCC	pTrix1.1
pN1-175-His	CTAGCCATGGGCACCCTCAAAGAAGTGCAAG	TATAATCTCGAGTGGCTTACGTATTCCATTAACT	pTrix1.1
pN1-237-His	CTAGCCATGGGCACCCTCAAAGAAGTGCAAG	TATAATCTCGAGATCATCCTTGAATCGGATTCTT	pTrix1.1
pN1-346-His	CTAGCCATGGGCACCCTCAAAGAAGTGCAAG	TATAATCTCGAGAGATTTTGATGCCATTATGGTG	pTrix1.1
pN1-402-His	CTAGCCATGGGCACCCTCAAAGAAGTGCAAG	TATAATCTCGAGATCCATATCATCTCCAAGATGG	pTrix1.1
pPol1–250	ATAATATCCATGGAGAAGTACCGCGAGATCC	ATAATATCTCGAGCCAGTGCTTGCAGTACTGGATCAGG	pTrix1.1
pPol251–752	ATAATATCCATGGTGACCGAGGATCACGATTTCGTGTTC	ATAATATCTCGAGCGGGCCCACTCCACGGTC	pTrix1.1
pPol751–1290	ATAATATCCATGGCCCGCAAGTTCGAGGCCAAG	ATAATATCTCGAGGCCTTCACGCAGCGGCTCTG	pTrix1.1
pPol1291–2153	ATAATATCCATGGCCTTCTACAGCTACAAGCACACCC	ATAATATCTCGAGGTAGAAGCTGCTCACGGGATC	pTrix1.1
pPol238–562	ATATATCCATGGCGCCCGAGATCACCAACCTGATCCAGT	ATATATCTCGAGGATGCTCATGACCTTGCTGAAGCAC	pTrix1.1
pPol562–1286	ATATATCCATGGCGATCGATCTGAACCGCCTGCTGGCCC	ATATATCTCGAGATCGGCCATGCCGATGCCGGCGGTG	pTrix1.1
pCDNA-FLAG	AATAATACTCGAGGATTACAAGGAGTACGATGACAAGGATTACAAGGATGACGATGACAAGGATTACAAGGAT	ATAATTTCTAGATTACTTGTCATCGTCATCCTTGTAATCCTTGTCATCGTCATCCTTGTAATCCTTGTATCGTC	pcDNA3.1(+)
pPol1291–2153-FLAG	ATAATATGCTAGCGCCTTCTACAGCTACAAGCACACCC	ATAATATCTCGAGGTAGAAGCTGCTCACGGGATC	pCDNA-FLAG
pTriEx-mCherry	TATAATCTCGAGGTGAGCAAGGGCGAGGAGGATAACAT	TAATATGCATGCTTACTTGTACAGCTCGTCCATGCCG	pTriEx1.1

^aTo generate His-tagged N mutants, PCR product was generated from pSNVN using the appropriate primers listed above. The PCR product was cloned into pTriEX 1.1 between the NcoI and XhoI sites. To generate His-tagged RdRp mutants, the region of interest was PCR amplified using the appropriate primers listed above, and the PCR product was cloned into the pTriEX 1.1 backbone between the NcoI and XhoI sites. pCNDA-FLAG was modified from pCDNA3.1(+) by annealing the listed primers and inserting the resulting DNA into pCDNA3.1(+) between the XhoI and XbaI sites. pPol1291–2153-FLAG was generated from pPol1291–2153 using the appropriate primers listed above, and the PCR product was cloned into pCDNA-FLAG between NheI and XhoI. To generate pTriEx mCherry, the PCR product was generated from pmCherry using the appropriate primers listed above. The PCR product was cloned into the pTriEX 1.1 backbone between the XhoI and SphI sites. The pPol238–562-mCherry and pPol1291–2153-mCherry constructs were generated by excising the open reading frame (ORF) from pPol238–562 and pPol1291–2153 using the NcoI and XhoI restriction enzymes. The excised ORF was cloned into pTriEx-mCherry between the same restriction sites.

To determine whether an N-RdRp interaction plays a role in hantavirus replication, we transfected Huh7 cells with a plasmid expressing either the pol1291–2153 or pol751–1290 fragment, followed by infection with SNV at an MOI of 0.5 16 h posttransfection. Cells were transfected again 36 h postinfection to boost the expression of RdRp fragments. Virus replication was monitored over time by quantitative estimation of viral S-segment RNA using real-time PCR analysis, as previously reported (25). It is evident that expression of the pol1291–2153 fragment significantly inhibited the SNV replication in cells (Fig. 4A). In comparison, the coexpression of the pol751–1290 fragment, which does not bind to N, had no impact upon virus replication. To determine whether the pol1291–2153 fragment selectively inhibits SNV replication, we examined the effect of this fragment upon the replication of vesicular stomatitis virus (VSV) (Fig. 4B) , another negative-strand RNA virus, in the cell culture model. HeLa cells grown in 24-well plates were transfected with a plasmid expressing the pol1291–2153 fragment, followed by infection with VSV at 10³ PFU 16 h posttransfection. An examination of virus replication over time revealed that coexpression of the pol1291–2153 fragment had no impact upon the replication of VSV, confirming the selectivity inhibition of SNV N by the pol1291–2153 fragment. Similar results were obtained with hepatitis C virus (HCV) (data not shown).

FIG 4.

Overexpression of the pol1291–2153 mutant specifically inhibits hantavirus replication in cells. (A) Huh7 cells were transfected with 2.5 μg of plasmid expressing either the pol1291–2153 or pol751–1290 mutant. Sixteen hours posttransfection, cells were infected with SNV at an MOI of 0.5. Thirty-six hours postinfection, cells were again transfected to boost the expression of RdRp fragments. Cells were harvested at 0, 12, 24, 36, 48, and 72 h postinfection. Total RNA was extracted from half of the cells collected at each time point with the RNeasy minikit and converted to cDNA using a random primer. S-segment RNA levels were quantified by real-time PCR using β-actin as an internal control, as previously reported (4, 14). Fold changes in S-segment RNA levels related to the zero hour time point are shown. Fold changes calculated from three independent experiments were averaged and used to calculate the standard deviation, shown as error bars. The remaining half of the cells were lysed with 1× Laemmli sample buffer containing 2% SDS and analyzed by Western blotting using the monoclonal anti-His antibody to monitor the expression of RdRp mutants. (B) HeLa cells were transfected with 2.5 μg of either empty vector or a plasmid expressing the Pol1291–2153 fragment. Cells were infected with vesicular stomatitis virus 16 h posttransfection (10³ PFU/well in a 24-well plate). Cells were harvested at 0, 6, 12, 18, and 24 h postinfection. Half of the cells were used for RNA extraction to quantify the VSV genomic RNA levels using real-time PCR. Fold changes in vRNA levels were calculated as mentioned in panel A. The remaining half of the cells were lysed for Western blot analysis to check the expression of tubulin and the pol1291–2153 fragment (bottom), as described in panel A.

Due to low expression levels, hantavirus RdRp has never been shown by Western blot analysis in virus-infected cells. We speculate that overexpression of the pol1291–2153 fragment in virus-infected cells (Fig. 4A) outcompeted wild-type RdRp for binding to N protein, which resulted in the inhibition of wild-type RdRp function. This clearly demonstrates that the N-RdRp interaction plays a critical role in the function of RdRp, although the mechanism for such a role is still unclear. This is consistent with requirement of N protein expression for RdRp function in numerous Bunyaviridae reverse genetic systems (13, 26). We propose that an N-RdRp interaction might play a role in cap snatching. Recently, multiple studies have identified the functional manganese-dependent endonuclease domain at the N terminus of the Bunyaviridae RdRp that shares a type II endonuclease α/β architecture similar to that of the N-terminal endonuclease domain of the influenza virus PA subunit (23, 27–31). These studies have led to the proposition that the RdRp endonuclease domain functions in cap snatching and is highly conserved among the Arenaviridae, Orthomyxoviridae, and Bunyaviridae families. Interestingly, the purified endonuclease domain used in these studies (23, 27) nonspecifically cleaved the RNA irrespective of the 5′ cap, raising the question of how capped primers of the appropriate length and specificity are generated by such a nonspecific cap-snatching endonuclease. The cap-binding activity of hantavirus RdRp has not been reported yet, again raising the question of how RdRp is loaded onto the mRNA 5′ cap during cap snatching. We previously reported that N protein binds to the mRNA 5′ cap. It is likely that simultaneous binding of N protein to both the RdRp and the mRNA 5′ cap may recruit RdRp at the mRNA 5′ cap for the specific cleavage of capped host cell mRNA to generate RNA primers of the appropriate length and specificity. It is possible that overexpressed pol1291–2153 fragment (Fig. 4A) competitively inhibited the binding of wild-type RdRp with N protein in virus-infected cells, which culminated in the inhibition of the cap-snatching process. It is equally possible that RdRp is recruited to the nucleocapsid templates by the N-RdRp interaction during transcription and replication of the viral genome. The strong colocalization of the overexpressed pol1291–2153 fragment with N protein in virus-infected cells suggests that the pol1291–2153 fragment might have competitively blocked the recruitment of wild-type RdRp on the nucleocapsid templates for transcription initiation.