Middle east respiratory syndrome coronavirus 4a protein is a double-stranded RNA-binding protein that suppresses PACT-induced activation of RIG-I and MDA5 in the innate antiviral response

Abstract

Middle East respiratory syndrome coronavirus (MERS-CoV) is an emerging pathogen that causes severe disease in human. MERS-CoV is closely related to bat coronaviruses HKU4 and HKU5. Evasion of the innate antiviral response might contribute significantly to MERS-CoV pathogenesis, but the mechanism is poorly understood. In this study, we characterized MERS-CoV 4a protein as a novel immunosuppressive factor that antagonizes type I interferon production. MERS-CoV 4a protein contains a double-stranded RNA-binding domain capable of interacting with poly(I · C). Expression of MERS-CoV 4a protein suppressed the interferon production induced by poly(I · C) or Sendai virus. RNA binding of MERS-CoV 4a protein was required for IFN antagonism, a property shared by 4a protein of bat coronavirus HKU5 but not by the counterpart in bat coronavirus HKU4. MERS-CoV 4a protein interacted with PACT in an RNA-dependent manner but not with RIG-I or MDA5. It inhibited PACT-induced activation of RIG-I and MDA5 but did not affect the activity of downstream effectors such as RIG-I, MDA5, MAVS, TBK1, and IRF3. Taken together, our findings suggest a new mechanism through which MERS-CoV employs a viral double-stranded RNA-binding protein to circumvent the innate antiviral response by perturbing the function of cellular double-stranded RNA-binding protein PACT. PACT targeting might be a common strategy used by different viruses, including Ebola virus and herpes simplex virus 1, to counteract innate immunity.

Importance: Middle East respiratory syndrome coronavirus (MERS-CoV) is an emerging and highly lethal human pathogen. Why MERS-CoV causes severe disease in human is unclear, and one possibility is that MERS-CoV is particularly efficient in counteracting host immunity, including the sensing of virus invasion. It will therefore be critical to clarify how MERS-CoV cripples the host proteins that sense viruses and to compare MERS-CoV with its ancestral viruses in bats in the counteraction of virus sensing. This work not only provides a new understanding of the abilities of MERS-CoV and closely related bat viruses to subvert virus sensing but also might prove useful in revealing new strategies for the development of vaccines and antivirals.

Figures

FIG 1

MERS-CoV 4a protein is a dsRNA-binding protein. (A) Sequence alignment. The alignment was generated with the ClustalW program. Identical residues are boxed. Highly similar residues are underlined. The most conservative K63 and K67 residues mutated in the dsm mutant are identified with arrows. Predicted secondary structures are indicated. Classical dsRNA-binding domains are composed of an α-β-β-β-α architecture. dsRBD1 and dsRBD2, dsRNA-binding domains 1 and 2, respectively. (B) Poly(I·C) pulldown assay. Plasmids (1.5 μg each) expressing the indicated V5-tagged proteins were individually transfected into HEK293T cells for 30 h. Cell lysates were incubated with poly(C)- or poly(I·C)-coated agarose for pulldown assay. Lysates and proteins retained in the agarose beads were analyzed by Western blotting. dsm, 4a protein mutant in which the dsRNA-binding domain is disrupted by replacing K63 and K67 with A; pC, poly(C); pIC, poly(I·C).

FIG 2

MERS-CoV 4a inhibits type I IFN production induced by poly(I·C) or Sendai virus. (A and B) Poly(I·C)-induced IFN production. HEK293T cells grown in 12-well plates were transfected with escalating doses (200, 400, and 600 ng) of a 4a protein expression plasmid. Empty vector was added as appropriate to ensure that cells in each well received the same amount of plasmids. At 24 h posttransfection, cells were further transfected with 1 μg/ml of poly(I·C). Samples were collected after an additional 12 h. Similar results were also obtained from HeLa cells. pIC, poly(I·C). (C to F) Sendai virus-induced IFN production. HEK293T (C and D) and HeLa (E and F) cells grown in 12-well plates were transfected with 100 ng of pIFNβ-Luc, 5 ng of pTK-RLuc, and escalating doses (200, 400, and 600 ng) of 4a protein expression plasmid. At 24 h posttransfection, cells were infected with Sendai virus (100 hemagglutinating units/ml). Cells were harvested at 12 h postinfection. Relative expression of IFN-β mRNA was analyzed by quantitative RT-PCR and normalized to the level of GAPDH mRNA expression (A, C, and E). Results from the dual-luciferase assay are expressed as the fold activation calculated from the pIFNβ-Luc activity normalized to that of pTK-RLuc (B, D, and F). Expression of 4a protein in selected groups was verified by Western blotting (insets in panels A and C). SeV, Sendai virus; α-tub, α-tubulin. Data are means of triplicate groups in one transfection, and error bars indicate SDs. Two-tailed Student's t test was performed, and the differences between the selected groups were statistically significant with the following P values which were less than 0.001 (***): 0.00052 (bars 2 and 4 in panel A), 0.00062 (bars 2 and 4 in panel B), 0.00039 (bars 2 and 4 in panel C), 0.00027 (bars 2 and 4 in panel D), 0.00062 (bars 2 and 4 in panel E), and 0.00062 (bars 2 and 4 in panel F). Results are representative of those from three independent transfections.

FIG 3

Comparative analysis of 4a proteins. (A) poly(I·C)-induced IFN production. Escalating doses (200, 400, and 600 ng) of a 4a protein plasmid were transfected into HEK293T cells grown in 12-well plates. Empty vectors were used to adjust the total amounts of plasmids in the transfection so that the cells in each well received the same dose of plasmids. Expression of 4a proteins in selected groups was verified by Western blotting (top). α-tub, α-tubulin; dsm, K63A/K67A mutant of MERS-CoV 4a protein. (B) Sendai virus-induced IFN production. Data presented are means from triplicate groups in one transfection, and error bars indicate SDs. Statistical analysis was performed on selected groups by two-tailed Student's t test. The P value (***) for bars 2 and 7 in panel A was 0.00066, and the one for bars 2 and 7 in panel B was 0.00033; both were less than 0.001. Results are representative of those from three independent transfections.

FIG 4

MERS-CoV 4a protein does not affect IFN production induced by RIG-I (A), MDA5 (B), MAVS (C), TBK1 (D), or IRF3 (E). Escalating doses (400 and 600 ng) of a 4a protein plasmid plus a fixed dose (50 ng) of activator plasmid was transfected into HEK293T cells grown in 12-well plates. Cells in the control group received empty vector alone. An expression plasmid (400 ng) for SARS-CoV M protein (SCV-M) was used as a positive control. Data are means of triplicate groups in one transfection, and error bars indicate SDs. A two-tailed Student's t test was performed, and no statistically significant difference was found between the tested groups. The P values were as follows: 0.34 (bars 2 and 5 in panel A), 0.28 (bars 2 and 5 in panel B), 0.32 (bars 2 and 5 in panel C), 0.42 (bars 2 and 5 in panel D), and 0.59 (bars 2 and 4 in panel E). n.s., not significant. Results are representative of those from three independent transfections.

FIG 5

MERS-CoV 4a protein inhibits PACT-induced activation of RIG-I and MDA5. (A and B) Influence of RIG-I on PACT activation. HEK293T cells grown in 12-well plates were transfected with pIFNβ-Luc or the pIRF3-Luc reporter as well as expression plasmids for RIG-I, PACT, and 4a proteins from MERS-CoV, bCoV-HKU4, and bCoV-HKU5. dsm is the K63A/K67A mutant of MERS-CoV 4a protein defective in dsRNA binding. HSV-1 Us11 was included as a control. Escalating doses (400 and 600 ng) of viral protein were used. The dual-luciferase assay was carried out at 30 h posttransfection. (C and D) Influence of MDA5 on PACT activation. Data are presented as fold activation and means ± SDs derived from triplicate groups in one transfection. A two-tailed Student's t test was performed for selected groups. The differences between bars 4 and 7 in panel A (P = 0.00032), between bars 4 and 6 in panel B (P = 0.00052), between bars 4 and 7 in panel C (P = 0.00041), as well as between 4 and 6 in panel D (P = 0.00032) were statistically significant (***, P < 0.001). The differences between bars 4 and 9 in panel A (P = 0.34), between bars 4 and 8 in panel B (0.22), between bars 4 and 9 in panel C (P = 0.38), as well as between bars 4 and 8 in panel D (P = 0.35) were statistically not significant (n.s.). Results are representative of those from three independent transfections. vec, vector.

FIG 6

RNA-dependent association of MERS-CoV 4a protein with PACT. (A) Coimmunoprecipitation. HEK293T cells were transfected with plasmids (1.5 μg each) expressing the indicated proteins. Immunoprecipitation (IP) was carried out at 48 h posttransfection with 0.5 μg of mouse anti-FLAG (α-FLAG) or anti-V5 (α-V5) and 20 μl of recombinant protein A-Sepharose fast-flow beads (GE Healthcare Life Sciences). Input cell lysates and precipitates were probed by Western blotting. (B) The interaction between MERS-CoV 4a and PACT is mediated through RNA. Coimmunoprecipitation was carried out in the presence of 1 or 5 U of RNase A. (C) Digestion of dsRNA by RNase A. dsRNA of about 1 kb, the sequence of which was derived from segment 7 (S7) of the influenza A virus WSN strain, was in vitro transcribed and annealed. It was then incubated with 1 U RNase A for 15 min. Agarose gel electrophoresis was performed to check for RNA integrity. (D) MERS-CoV 4a protein does not interact with RIG-I or MDA5.