SARS coronavirus 3b accessory protein modulates transcriptional activity of RUNX1b

Abstract

Background: The causative agent of severe acute respiratory syndrome, SARS coronavirus (SARS-CoV) genome encodes several unique group specific accessory proteins with unknown functions. Among them, accessory protein 3b (also known as ORF4) was lately identified as one of the viral interferon antagonist. Recently our lab uncovered a new role for 3b in upregulation of AP-1 transcriptional activity and its downstream genes. Thus, we believe that 3b might play an important role in SARS-CoV pathogenesis and therefore is of considerable interest. The current study aims at identifying novel host cellular interactors of the 3b protein.

Methodology/principal findings: In this study, using yeast two-hybrid and co-immunoprecipitation techniques, we have identified a host transcription factor RUNX1b (Runt related transcription factor, isoform b) as a novel interacting partner for SARS-CoV 3b protein. Chromatin immunoprecipitaion (ChIP) and reporter gene assays in 3b expressing jurkat cells showed recruitment of 3b on the RUNX1 binding element that led to an increase in RUNX1b transactivation potential on the IL2 promoter. Kinase assay and pharmacological inhibitor treatment implied that 3b also affect RUNX1b transcriptional activity by regulating its ERK dependent phosphorylation levels. Additionally, mRNA levels of MIP-1α, a RUNX1b target gene upregulated in SARS-CoV infected monocyte-derived dendritic cells, were found to be elevated in 3b expressing U937 monocyte cells.

Conclusions/significance: These results unveil a novel interaction of SARS-CoV 3b with the host factor, RUNX1b, and speculate its physiological relevance in upregulating cytokines and chemokine levels in state of SARS virus infection.

Figure 1. SARS-CoV 3b interacts with RUNX1b.

A. A schematic representation of full-length 3b, pHybLexA/Zeo-3b bait plasmid, full-length RUNX1b and pYesTrp2-RUNX1b (51–421 aa) prey plasmid. The 3b protein has a nucleolar localization signal (NoLS) at the C-terminus. The RUNX1b protein has a runt homology domain (RHD, 50–177 aa), an activation domain (AD, 291–171 aa), and an inhibitory domain (ID, 346–411 aa). B. The 3b-RUNX1b interaction was assessed on a selective growth media (supplemented with 3-AT) and by filter lift β-galactosidase activity assay in a yeast two-hybrid experiment. pHybLexA/Zeo, pYesTrp2, pHybLexA/Zeo-3b, pYesTrp2-RUNX1b, pHybLexA/Zeo-Fos and pYesTrp2-Jun constructs were co-transformed in L40 in combinations tabulated above. pHybLexA/Zeo-Fos and pYesTrp2-Jun were used as positive control. C, D. In vitro analysis of 3b and RUNX1b interaction. C. In vitro translated S^35-labelled 3b and RUNX1b lysates (input) were subjected to co-immunoprecipitation alone or together, using α–RUNX1 antibody. D. Total cell lysates of Huh7 cells expressing indicated proteins were immunoprecipitated with α-Flag antibody and western blotted with α-myc antibody to probe 3b protein (panel 1). Lysates were probed for the expression of RUNX1b and 3b with α-Flag and α-myc antibodies, respectively.

Figure 2. 3b partially co-localizes with RUNX1b in the nucleus.

Cellular distribution of 3b and RUNX1b proteins were visualized by subjecting Flag-RUNX1b and HA-3b transfected HEK293 cells to immunofluorescence assay. 3b was visualized using primary α–HA and alexa-488 conjugated secondary antibody. RUNX1b was visualized using primary α–RUNX1 and alexa-594 conjugated secondary antibody. Nuclei were visualized by DAPI (4′6-diamidino-2-phenylindole) staining. Arrows indicate extra nucleolar nucleus area of partial co-localization.

Figure 3. 3b recruitment on RUNX1 binding elements on the IL2 promoter.

Chromatin immunoprecipitation assays were performed with jurkat cells transfected with vector alone or HA-3b using α-RUNX1 and α-HA antibodies. PCR amplifications were performed using IL2 promoter primers and 3′ distal IL2 promoter primers. Results are representative of three independent experiments.

Figure 4. 3b increases RUNX1b transactivation potential on the mouse IL2 promoter.

A. HEK293 cells were transfected with WT IL2-Luc plasmid alone or with Flag-RUNX1b, Flag-CBFβ and indicated amounts of Myc-3b. Relative luciferase activities were calculated 48 h post transfection. B. Jurkat cells were transfected with WT IL2-Luc and indicated amounts of Myc-3b. Relative luciferase activities were calculated 48 h post-transfection. 3b expression was probed using α-myc antibody C. Jurkat cells were transfected with WT IL2-Luc or mutant IL2-Luc plasmid in the presence or absence of Myc-3b. Results in each panel are represented as mean±S.D. of triplicate cultures. Bar values represent fold increase in luciferase activity. *, p<0.005.

Figure 5. 3b expression increases phosphorylated RUNX1b levels through ERK activation.

A. HEK293 cells were transfected with vector, 3b and RUNX1b expression plasmids. ERK immunoprecipitated from vector and 3b lysates were subjected to kinase assay with RUNX1b beads. Phosphorylated RUNX1b was visualized by autoradiography. Input levels of immunoprecipitated ERK and phospho ERK levels in lysates were probed by western blotting. Graph depicts fold increase in the levels of phosphorylated RUNX1b procured after three independent experiments. Bar represents mean±SD of values obtained by densitometry. #, p<0.05. B. Jurkat cells were transfected with WT IL2-Luc in the presence or absence of Myc-3b and treated with DMSO or U0126. Relative luciferase activity was measured and is shown as the mean±SD of three independent experiments performed in triplicates. *, p<0.005. Phospho ERK levels in lysates were probed by western blotting.

Figure 6. 3b and RUNX1b cooperatively increase MIP-1α mRNA levels.

Relative mRNA levels of MIP-1α to actin were estimated in U937 cells expressing indicated proteins, using quantitative RT-PCR. Histogram is the result of three independent experiments. Bar values represent fold increase in the mRNA levels. Asterisk *, p<0.005.