Cellular RNA binding proteins NS1-BP and hnRNP K regulate influenza A virus RNA splicing

Abstract

Influenza A virus is a major human pathogen with a genome comprised of eight single-strand, negative-sense, RNA segments. Two viral RNA segments, NS1 and M, undergo alternative splicing and yield several proteins including NS1, NS2, M1 and M2 proteins. However, the mechanisms or players involved in splicing of these viral RNA segments have not been fully studied. Here, by investigating the interacting partners and function of the cellular protein NS1-binding protein (NS1-BP), we revealed novel players in the splicing of the M1 segment. Using a proteomics approach, we identified a complex of RNA binding proteins containing NS1-BP and heterogeneous nuclear ribonucleoproteins (hnRNPs), among which are hnRNPs involved in host pre-mRNA splicing. We found that low levels of NS1-BP specifically impaired proper alternative splicing of the viral M1 mRNA segment to yield the M2 mRNA without affecting splicing of mRNA3, M4, or the NS mRNA segments. Further biochemical analysis by formaldehyde and UV cross-linking demonstrated that NS1-BP did not interact directly with viral M1 mRNA but its interacting partners, hnRNPs A1, K, L, and M, directly bound M1 mRNA. Among these hnRNPs, we identified hnRNP K as a major mediator of M1 mRNA splicing. The M1 mRNA segment generates the matrix protein M1 and the M2 ion channel, which are essential proteins involved in viral trafficking, release into the cytoplasm, and budding. Thus, reduction of NS1-BP and/or hnRNP K levels altered M2/M1 mRNA and protein ratios, decreasing M2 levels and inhibiting virus replication. Thus, NS1-BP-hnRNPK complex is a key mediator of influenza A virus gene expression.

Figure 1. NS1-BP interacts with RNA-binding proteins, RNA polymerase II, and influenza A virus polymerase.

(A) Top panel, HeLa cell lysates were immunoprecipitated with control IgG or NS1-BP antibodies. Interacting proteins were resolved by SDS-PAGE and identified by mass spectrometry. Bottom panel, A549 cells were treated with non-targeting or NS1-BP siRNAs (1, 2, and 3) for 48 h. Cell extracts were subjected to immunoblot analysis, which shows NS1-BP knockdown. β-actin served as loading control. (B) Immunoprecipitation was performed with control IgG or anti-NS1-BP antibodies, in the absence or presence of RNase A. Western blots were then performed with the depicted antibodies, selected based on the proteins identified in A. (C) A549 cells were mock-infected or infected with influenza virus at MOI 5 for 5 h. Cells were lysed and subjected to size exclusion chromatography. The fractions were concentrated by TCA precipitation and analyzed by western blot with the indicated antibodies. (D) A549 cells were mock-infected or infected with influenza virus (A/WSN/33) at MOI 2 for 16 h. Cell lysates were immunoprecipitated with control IgG or anti-NS1-BP antibodies. Western blots were performed with the depicted antibodies.

Figure 2. NS1-BP promotes splicing of influenza A virus M1 mRNA segment to yield proper levels of M2 mRNA and protein.

(A) A549 cells were transfected with non-targeting or NS1-BP siRNAs for 48 h and infected with A/WSN/33 at MOI 2 for the indicated time points. Cells lysates were subjected to western blot analysis using antibodies against influenza virus proteins. α-tubulin was used as loading control. Each protein band was quantified by ImageJ and normalized to α-tubulin levels. (B) M1 mRNA and its alternatively spliced products are depicted and the arrowheads show primer positions for detection of various mRNAs in the following experiments. (C, D, and E) Cells were transfected and infected as in A and total RNA was isolated and analyzed by real-time RT-PCR with primers specific to M1, M2, mRNA₃, and M4 mRNAs. Ratios of M2 mRNA to M1 (C) or to mRNA₃ (D), or to M4 (E) over time are presented. (F and G) RNA samples from C were analyzed with primers specific to NS2 and NS1 viral mRNAs. (G) Ratios of NS2 mRNA to NS1 mRNA over time are shown. (H) NS1-BP mRNA levels were measured by real time RT-PCR in samples from C. Error Bars represent mean ± SD (n = 3).

Figure 3. NS1-BP interacts with hnRNPs that directly bind M1 mRNA.

(A) The entire M1 RNA was radiolabeled uniformly at either C (left) or U (right) residues, incubated with nuclear extract under splicing conditions, crosslinked with 254 nm light and digested with RNase. The reaction was then either resolved by SDS-PAGE (10% Total) or incubated in separate reactions with the antibodies indicated (IP: anti-). Immunoprecipitated proteins were resolved by SDS-PAGE. Position of molecular weight markers are also shown. Vertical dotted line shows that the samples were loaded in the same gel but not next to each other. Figure S4 shows immunoblot analysis of nuclear extract with antibodies against NS1-BP and hnRNP U, which did not directly bind M1 mRNA but are present in the extract. (B) A549 cells were infected with A/WSN/33 at MOI 5 for 5 h and were then cross-linked with 0.3% formaldehyde. Cells were lysed with RIPA buffer, and the lysates were immunoprecipitated with control IgG or NS1-BP antibodies. After a series of washes, the cross-link was reversed, and associated RNAs were isolated. Real-time quantitative RT-PCR was used to measure M1 mRNA. Housekeeping genes (GAPDH, β-actin, and α-tubulin) were included to test the specificity of the assay. Data was normalized to the viral M segment. The graph displays the mean ± SEM (n = 3). Comparison between cross-linked samples shows significant (p<0.05) differences between association of NS1-BP with M1 mRNA and the lack of association of NS1-BP with the 3 host mRNAs tested.

Figure 4. hnRNP K regulates M1 mRNA splicing.

(A) Cell extracts from A549 cells transfected with non-targeting siRNAs or siRNAs that target various indicated hnRNPs for 48 h were subjected to immunoblot analysis with antibodies specific to each hnRNP. (B and C) A549 cells were transfected as in B and then infected with A/WSN/33 at MOI 2 for the indicated time points. In C, NS1-BP and hnRNP K were simultaneously knocked down. Total RNA was isolated at each depicted time point and analyzed by real-time RT-PCR with primers specific to M1 and M2 mRNAs. M2/M1 mRNA ratios are shown. Error bars represent mean ± SD (n = 3). *p<0.05, **p<0.01, ***p<0.001.

Figure 5. NS1-BP is required for optimal influenza A virus replication.

(A) and (B) A549 cells were transfected with non-targeting or NS1-BP siRNAs for 48 h and infected with A/WSN/33 at MOI 0.001. Supernatants were harvested at 24 h and 36 h post-infection, and virus titer was determined by plaque assay. (C) and (D) After removal of supernatants from plaque assays in A and B, the remaining cells were used to measure ATP levels, at 24 h and 36 h post-infection, to determine cell viability. (E) Human bronchial epithelial cells (HBEC) were transfected and infected as in A except that different NS1-BP siRNAs were used. Supernatants were subjected to plaque assay. (F) Cell extracts from A549 cells transfected with siRNAs as in A and HBEC transfected with siRNAs as in E were subjected to immunoblot analysis with anti-NS1-BP antibodies. * denotes a cross-reacting band that serves as loading control. *p<0.05, **p<0.01.

Figure 6. hnRNP K is required for optimal M2 protein production and influenza A virus replication.

A549 cells were transfected with non-targeting or hnRNP K siRNAs for 48 h prior to infection. siRNA transfected cells were infected with A/WSN/33 at (A) MOI 2 or (B) MOI 0.001. Cells were harvested at the indicated hours post-infection, and viral protein accumulation was assessed by immunoblot analysis. Each protein band in (A) and (B) was quantified by ImageJ and normalized to α-tubulin levels. (C) Control or hnRNP K siRNA transfected cells were infected with A/WSN/33 at MOI 0.001. At 36 hours post-infection, cell supernatants were collected and subjected to viral titer analysis (n = 3) or (D) ATP level analysis to determine cell viability (n = 3, representative experiment). Error bars denote mean + SEM. *p<0.05.