Rotavirus nonstructural protein 1 suppresses virus-induced cellular apoptosis to facilitate viral growth by activating the cell survival pathways during early stages of infection

Abstract

Following virus infection, one of the cellular responses to limit the virus spread is induction of apoptosis. In the present study, we report role of rotavirus nonstructural protein 1 (NSP1) in regulating apoptosis by activating prosurvival pathways such as phosphatidylinositol 3-kinase (PI3K)/Akt and NF-kappaB (nuclear factor kappaB) during early hours of infections (2 to 8 hpi). The NSP1 mutant strain A5-16 induces weak and transient activation of Akt (protein kinase B) and p65 NF-kappaB compared to the isogenic wild-type strain A5-13 in MA104 or HT29 cells. The weak NF-kappaB promoter activity or Akt phosphorylation after A5-16 infection could be complemented in cells transfected with plasmid expressing NSP1 after infection with the rotavirus A5-16 strain. In cells either infected with A5-13 or transfected with pcD-NSP1, coimmunoprecipitation of NSP1 with phosphoinositide 3-kinase (PI3K) was observed, indicating that strong activation of PI3K/Akt could be due to its interaction with NSP1. In addition, after infection with same multiplicity of infection, A5-16 showed reduced number of viral particles compared to the A5-13 strain at the end of the replication cycle. A lower growth rate could be due to weak induction of PI3K/Akt and NF-kappaB, since the A5-13 strain also showed reduced growth in the presence of PI3K or NF-kappaB inhibitors. This effect was interferon independent; however, it was partly due to significantly higher caspase-3 activity, poly-ADP ribose polymerase (PARP) cleavage, and apoptosis during earlier stages of infection with the NSP1 mutant. Thus, our data suggest that NSP1 positively supports rotavirus growth by suppression of premature apoptosis for improved virus growth after infection.

FIG. 1.

(A) Schematic diagram of full-length (1,579 bases; wt A5-13) and deletion mutant (1,087 bases; mutant A5-16) gene segments 5 encoding NSP1 (59 kDa) of rotavirus. In A5-13, nucleotide positions 105 to 246 and 246 to 531 correspond to the deduced RING domain and cytoskeleton localization domain sequence, respectively, whereas position 981 to the rest of the gene implies an IRF3 binding domain. The NSP1 gene of A5-16 has a 500-nucleotide deletion from nucleotides 142 to 641, followed by an immediate stop codon at positions 183 to 185, indicating the lack of a functional RING domain and cytoskeleton localization domain. (B) Immunoblot analyses showing the expression of structural protein VP6 and nonstructural proteins NSP1 and NSP3 of both wt A5-13 and mutant A5-16 strains (MOI of 3, 12 hpi). Both A5-13- and A5-16-infected cells expressed VP6 and NSP3 proteins, but NSP1 expression was observed only in A5-13-infected cells. The blots were reprobed with β-actin antibody to confirm equal protein loading.

FIG. 2.

Rotavirus-induced PI3K/Akt activation is NSP1 dependent. (A) Western blot analysis showing phosphorylation of Akt (ser473) and its downstream substrate GSK3β at 2 to 12 hpi with A5-13 and A5-16 (MOI of 3) in MA104 cells. Equal protein loading was verified by reprobing the blots to analyze the basal expression of total Akt and GSK3β. The band intensities of phosphoproteins were normalized against that of total proteins. The data represent the means ± the standard deviations (SD) of three independent experiments. (B) Akt phosphorylation (ser473) level was compared by Western blotting in the 293T cell line after overexpression of full-length NSP1 (pcD-NSP1) and its truncated form (pcD-ΔNSP1) with respect to the vector-transfected control. Total Akt expression was monitored as a loading control. The NSP1 dependence of Akt activation was further verified by analyzing the pAkt level over the total Akt level in 293T cells overexpressed with full-length or truncated NSP1 protein (24 h), followed by infection with the NSP1 mutant virus A5-16 (MOI of 3) in a time-dependent manner. (C) Western immunoblot of Akt phosphorylation (ser473) from extracts of MA104 cells infected with wt A5-13 and NSP1 mutant A5-16 in a dose (MOIs of 5 and 1, respectively)- and time-dependent manner. Blots were reprobed to analyze the total Akt expression as a loading control. The VP6 protein level was also checked to verify the infection status of both viruses.

FIG. 3.

NSP1 coimmunoprecipitates with the p85α subunit of PI3K. (A) MA104 cell lysates prepared at 6 hpi with A5-13 and A5-16 (MOI of 3) were subjected to immunoprecipitation (IP) with antibody directed against p85α subunit of PI3K (panel A, lanes 1 and 2) or rabbit serum control (panel A, lanes 3 and 4), respectively. The precipitates were detected by Western blotting with NSP1-specific antibody (upper panel). Blots were reprobed with PI3K (bottom panel) as an internal positive control. Input cell lysates were probed to check the specificity of NSP1 antibody. (B) After overexpression of full-length NSP1 (pcD-NSP1) and its truncated form (pcD-ΔNSP1) in MA104 cells for 24 h, immunoprecipitation was performed with p85α subunit of PI3K (panel B, lanes 3 and 4) or rabbit serum control (panel B, lanes 1 and 2). A Western blot showed proteins immunoprecipitated by preimmune serum or anti-PI3K-p85α probed with anti-His antibody. The blots were reprobed with PI3K as an internal positive control. An input cell lysate was used (panel B, bottom panel) to detect the presence of His-tagged NSP1 or its truncated form.

FIG. 4.

NSP1-dependent Akt activation during rotavirus infection was IFN independent. (A) qRT-PCR analysis of IRF3, IFN-β, and ISG56 transcripts from wt (A5-13) and NSP1 mutant (A5-16) strains infecting (MOI of 3) MA104 cells revealed a significant increase in IFN induction in mutant virus-infected cells in a time-dependent manner (2 to 8 h). In wt virus-infected cells, the IFN is induced at an early time point, followed by a decrease at 8 h. The data shown are means ± the SD (n = 3; P < 0.01). (B) Western blot analysis of Akt phosphorylation in Vero cells infected with either wt or NSP1 mutant virus (MOI of 10) at 6 to 12 h. The blot was reprobed to measure the total Akt as an endogenous control.

FIG. 5.

Sustained activation of prosurvival protein NF-κB during rotavirus infection is NSP1 dependent. (A) Nuclear translocation of NF-κB after rotavirus infection by immunofluorescence microscopy. MA104 cells either mock infected or infected with wt A5-13, SA11, NCDV, or mutant A5-16 (MOI of 3) strains were fixed at 6 hpi with paraformaldehyde and incubated with both viral rabbit polyclonal NSP5 and NF-κB specific mouse monoclonal antibody, followed by FITC-labeled anti-rabbit and RRX-labeled anti-mouse secondary antibodies. Nuclei were stained with DAPI (4′,6′-diamidino-2-phenylindole). Poor nuclear translocation of cellular NF-κB was observed upon infection with NSP1 mutant A5-16 compared to wt bovine A5-13 or simian SA11 strains. Bovine strain NCDV also showed relatively less nuclear localization compared to SA11 and A5-13. (B) Western blot analysis of NF-κB from nuclear lysates of MA104 cells either mock infected or infected with SA11, A5-13, NCDV, or A5-16 (MOI of 3) for 2 and 6 h. The blots were reprobed with anti-mouse PCNA antibody to analyze the expression of nuclear protein PCNA as an internal expression control. (C and D) Relative increases in NF-κB promoter activity measured by an NF-κB luciferase reporter assay in HEK293T cells show 3- to 3.5-fold-higher activities in A5-13-infected cells compared to A5-16 infection (MOI of 5). LY294002 (10 μM) and SN50 (5 μM) significantly inhibit NF-κB activation in A5-13-infected cells. TNF-α (10 nM) was used as a positive control for NF-κB activation. The relative fold change in NF-κB promoter activity in pcD-NSP1-transfected HEK293T cells showed a 2.5- to 3.2-fold increase over the vector control or the truncated NSP1 (pcD-ΔNSP1). NF-κB activation was restored in cells expressing wt NSP1, followed by A5-16 mutant virus infection. The data are presented as the fold change in luciferase units (mean ± the SD; n = 4) relative to mock infection or untransfected control and was normalized with the Renilla luciferase activity. *, Statistical significance (P < 0.01).

FIG. 6.

(A) Attenuated growth of the NSP1 mutant virus is due to the poor induction of PI3K/Akt and NF-κB activity. Virus growth assays were done by infecting the MA104 and HT29 cells (MOI of 3) either with wt A5-13 or the mutant A5-16 or A5-13 strain in the presence of PI3K (LY294002; 10 μM) or NF-κB (SN50; 5 μM) inhibitors. Virus titers, as determined by plaque assay (24 hpi), showed slower growth rates of A5-16 compared to the A5-13 strain. A significant decrease in virus titers of A5-13 was observed in the presence of PI3K/Akt and NF-κB inhibitors. The data represent the means of three experiments (n = 3, P < 0.05). (B) Enhanced viral gene expression in the presence Akt and NF-κB agonists (EGF [10 μM] and TNF-α [10 nM]). VP6 transcripts were quantitated by qRT-PCR after A5-16 infection in either Akt or NF-κB agonist-treated MA104 cells or untreated MA104 cells. The data represent means ± the SD (n = 3, P < 0.01). (C) The percentage of cytotoxicity was measured by MTS assay in MA104 and HT29 cells in serum-free conditions, after treatment with LY294002 (10 μM), SN50 (5 μM), or DMSO (control) in a time-dependent manner. In both cell lines, 4.5 to 7% and 11.5 to 17% cytotoxicities were observed with either inhibitor at 24 and 48 h of treatment, respectively.

FIG. 7.

Rotavirus NSP1 inhibits virus-induced premature apoptosis. MA104 cells were infected with either wt A5-13 or mutant A5-16 (MOI of 3). Cell lysates were prepared at increasing time intervals, and Western blot analysis was performed to detect the expression of antiapoptotic or apoptotic proteins: XIAP (A) and caspase-3 and PARP (B). β-Actin was used as an endogenous control for equal protein loading. (C) Caspase-3 activity was measured in MA104 cells infected with wt A5-13 or mutant A5-16 strains, using the fluorogenic substrate DEVD-AFC, revealed higher caspase-3 activities during the initial stages of infection in A5-16-infected cells compared to A5-13-infected cells. Caspase-3 inhibitor DEVD-fmk was used as an internal control. (D) Caspase-3 enzyme activity as measured in MA104 cells infected with A5-13 (MOI of 3) in the presence of PI3K inhibitor (LY294002, 10 μM), NF-κB inhibitor SN50 (5 μM), or DMSO as a control. In the presence of PI3K or NF-κB inhibitors, increased caspase-3 activity was observed as early as 8 hpi. (E) DNA fragmentation was measured by TUNEL assay by flow cytometry analysis. MA104 or HT29 cells were infected with either wt A5-13 or mutant A5-16 (MOI of 3) or mock infected for 8, 12, and 18 h or 8, 16, and 24 h, respectively, and incubated with terminal deoxynucleotidyltransferase and FITC-conjugated anti-BrdU monoclonal antibodies according to standard protocols (BD Pharmingen). In A5-16-infected MA104 or HT29 cells, increased TUNEL positivity was observed during earlier stages of infection compared to A5-13-infected or mock-infected controls. The data shown represent means ± the SD (n = 4).

FIG. 8.

Putative model for mechanism of rotavirus NSP1-mediated proviral function. The model is based on past (dashed arrows) and present (solid arrows) results, as summarized in the Discussion. In brief, stimulation of rotavirus growth may result from either subversion of IFN signaling by NSP1 protein, as well as the suppression of premature apoptosis by the NSP1-dependent PI3K→Akt→ NF-κB pathway. Both pathways may act independently of each other and may also cross talk at one or more steps.