Phosphatase Cdc25A Negatively Regulates the Antiviral Immune Response by Inhibiting TBK1 Activity

The RLR-mediated antiviral immune response is critical for host defense against RNA virus infection. However, the detailed mechanism for balancing the RLR signaling pathway in host cells is not well understood. We found that the phosphatase Cdc25A negatively regulates the RNA virus-induced innate immune response. Our studies indicate that Cdc25A inhibits the RLR signaling pathway via its phosphatase activity. We demonstrated that Cdc25A reduces TBK1 activity and consequently restrains the activation of IFN-β transcription as well as the antiviral status of nearby cells. We showed that Cdc25A can also inhibit DNA virus-induced activation of IFN-β. Taken together, our findings uncover a novel function and mechanism for Cdc25A in regulating antiviral immune signaling. These findings reveal Cdc25A as an important negative regulator of antiviral immunity and demonstrate its role in maintaining host cell homeostasis following viral infection.

ABSTRACT

The phosphatase Cdc25A plays an important role in cell cycle regulation by dephosphorylating its substrates, such as cyclin-dependent kinases. In this study, we demonstrate that Cdc25A negatively regulates RIG-I-mediated antiviral signaling. We found that ectopic expression of Cdc25A in 293T cells inhibits the activation of beta interferon (IFN-β) induced by Sendai virus and poly(I·C), while knockdown of Cdc25A enhances the transcription of IFN-β stimulated by RNA virus infection. The inhibitory effect of Cdc25A on the antiviral immune response is mainly dependent on its phosphatase activity. Data from a luciferase assay indicated that Cdc25A can inhibit TBK1-mediated activation of IFN-β. Further analysis indicated that Cdc25A can interact with TBK1 and reduce the phosphorylation of TBK1 at S172, which in turn decreases the phosphorylation of its downstream substrate IRF3. Consistently, knockdown of Cdc25A upregulates the phosphorylation of both TBK1-S172 and IRF3 in Sendai virus-infected or TBK1-transfected 293T cells. In addition, we confirmed that Cdc25A can directly dephosphorylate TBK1-S172-p. These results demonstrate that Cdc25A inhibits the antiviral immune response by reducing the active form of TBK1. Using herpes simplex virus 1 (HSV-1) infection, an IFN-β reporter assay, and reverse transcription-quantitative PCR (RT-qPCR), we demonstrated that Cdc25A can also inhibit DNA virus-induced activation of IFN-β. Using a vesicular stomatitis virus (VSV) infection assay, we confirmed that Cdc25A can repress the RIG-I-like receptor (RLR)-mediated antiviral immune response and influence the antiviral status of cells. In conclusion, we demonstrate that Cdc25A negatively regulates the antiviral immune response by inhibiting TBK1 activity.

IMPORTANCE The RLR-mediated antiviral immune response is critical for host defense against RNA virus infection. However, the detailed mechanism for balancing the RLR signaling pathway in host cells is not well understood. We found that the phosphatase Cdc25A negatively regulates the RNA virus-induced innate immune response. Our studies indicate that Cdc25A inhibits the RLR signaling pathway via its phosphatase activity. We demonstrated that Cdc25A reduces TBK1 activity and consequently restrains the activation of IFN-β transcription as well as the antiviral status of nearby cells. We showed that Cdc25A can also inhibit DNA virus-induced activation of IFN-β. Taken together, our findings uncover a novel function and mechanism for Cdc25A in regulating antiviral immune signaling. These findings reveal Cdc25A as an important negative regulator of antiviral immunity and demonstrate its role in maintaining host cell homeostasis following viral infection.

INTRODUCTION

The innate immune response is the first line of host defense against viral infection. Molecular components released during pathogen infection are known as pathogen-associated molecular patterns (PAMPs) that can be recognized by host pattern recognition receptors (PRRs). Viral RNAs are typical PAMPs, which are mainly detected by two types of PRRs: RIG-I-like receptors (RLRs) and Toll-like receptors (TLRs) (1, 2). Viral RNA in the cytoplasm is predominantly recognized by the RLRs, which include RIG-I (retinoic acid-inducible gene I), MDA5 (melanoma differentiation-associated gene 5), and LGP2 (laboratory of genetics and physiology 2). RIG-I and MDA5 are two major RLRs that mediate the antiviral immune response, and their downstream signaling pathways largely overlap (3).

As cells are infected by a virus, the C-terminal domain (CTD) of RIG-I binds to the 5′ phosphate of viral RNA and the helicase domain of RIG-I binds to the backbone of viral RNA, leading to the release of its CARD domain. The CARD domain of RIG-I can interact with VISA (virus-induced signaling adapter; also called MAVS, IPS-1, or Cardif) and activate it (4–6). Activated VISA recruits downstream factors to form the signalosome complex, which induces the phosphorylation and activation of TBK1 (TANK-binding kinase 1) and IKKs (IκB kinases) (7, 8). Activated TBK1 and IKKε phosphorylate IRF3, which promotes the nuclear entry of IRF3 (9). On the other hand, IKKα/β phosphorylates IκB and stimulates the degradation of IκB, which facilitates the nuclear entry of NF-κB (10). Activated IRF3 and NF-κB can enhance the transcription of type I interferons and inflammatory cytokines.

Posttranslational modifications (PTMs) of proteins play critical roles in many processes, including antiviral signaling. TRIM25 mediates K63-linked ubiquitination of RIG-I and promotes its interaction with VISA and the downstream signaling (4, 11). RNF128 catalyzes the K63-linked polyubiquitination of TBK1 and enhances its activity (12). Ndfip1 facilitates Smurf1-mediated MAVS ubiquitination and degradation and inhibits RLR-mediated signaling (13). The SUMOylation of RIG-I by MAPL promotes RIG-I interaction with MAVS (14). The maintenance of the deacetylation status of TBK1 by HDAC9 enhances TBK1 kinase activity (15). In addition, the phosphorylation of proteins participates in many aspects of regulation of the antiviral immune response. The phosphatase PP1α/γ dephosphorylates MDA5 and RIG-I, which leads to the activation of MDA5 and RIG-I and the upregulation of downstream signaling (16). CK2 (casein kinase 2) inactivates RIG-I by phosphorylating it at Thr770 and Ser855/854 (17). CK1γ1 inhibits the RLR/TLR signaling pathway by phosphorylating p65 and promoting its degradation (18). DYRK2 phosphorylates TBK1 at Ser527, which promotes its ubiquitination and degradation (19). GSK3β enhances the dimerization and autophosphorylation of TBK1 upon viral infection (20). PTEN releases the inhibitory phosphorylation of IRF3, facilitating its import into the nucleus (21). These reports suggest that both kinases and phosphatases play important roles in the RLR-mediated immune response.

In this study, we demonstrate that the phosphatase Cdc25A inhibits the RNA virus-induced innate immune response. We found that Cdc25A inhibits RLR signaling and that this inhibitory activity is dependent on its phosphatase activity. Further study indicated that Cdc25A reduces the active phosphorylation of TBK1, which in turn decreases the transcription of beta interferon (IFN-β). Our findings reveal a novel function and mechanism for Cdc25A in regulating antiviral immune signaling.

RESULTS

Phosphatase Cdc25A inhibits the activation of IFN-β induced by RNA viruses.

Phosphorylation and dephosphorylation of key proteins play critical roles in RLR-mediated antiviral signaling pathways (22). We previously observed that Cdc25A inhibited Sendai virus (SeV)-induced activation of IFN-β in a screen for kinases and phosphatases involved in RLR-mediated signaling (18). To confirm the inhibitory effect of Cdc25A on the RLR pathway, we transfected 293T cells with a Cdc25A-expressing plasmid or a PP1γ-expressing plasmid (as a positive control) and infected the cells with SeV. We then detected the activity of the IFN-β reporter. The data showed that Cdc25A can inhibit activation of the IFN-β luciferase reporter induced by SeV, while PP1γ enhances activation of the IFN-β luciferase reporter, as previously reported (16) (Fig. 1A and B). Next, we examined whether Cdc25A can inhibit activation of the IFN-β luciferase reporter by other stimuli. As shown in Fig. 1C and D, ectopic expression of Cdc25A inhibited activation of the IFN-β luciferase reporter induced by influenza A virus RNA and poly(I·C). Consistently, we found that Cdc25A hindered the transcription of IFN-β mRNA induced by SeV (Fig. 1E). To examine whether knockdown of Cdc25A favors the activation of IFN-β, we generated Cdc25A knockdown 293T cells by use of two short hairpin RNAs (shRNAs) targeting Cdc25A (Fig. 1F). We transfected the cells with an IFN-β luciferase reporter and then infected them with SeV. As shown in Fig. 1G, the luciferase activity in Cdc25A knockdown cells was significantly higher than that in control cells. Consistently, the level of IFN-β mRNA in Cdc25A knockdown cells was higher than that in control cells (Fig. 1H). We also performed a similar experiment by using small interfering RNA (siRNA) to transiently knock down Cdc25A and observed that knockdown of Cdc25A enhanced the transcription of IFN-β induced by SeV (Fig. 1I and J). Taken together, these results indicate that Cdc25A negatively regulates the activation of IFN-β induced by RNA virus infection.

FIG 1.

Cdc25A negatively regulates IFN-β production. (A and B) HEK293T cells were cotransfected with pCMV myc-Cdc25A or pCMV myc-PP1γ and the IFN-β luciferase reporter plasmid for 24 h and then infected with SeV for 8 h. The cell lysates were harvested for immunoblotting with the indicated antibodies (A) and for luciferase assay (B). (C and D) HEK293T cells were transfected with myc-Cdc25A and the IFN-β luciferase reporter plasmid and then transfected with influenza A virus RNA (C) and poly(I·C) (D). The cell lysates were collected for luciferase assay. (E) HEK293T cells were transfected with pCMV myc-Cdc25A or empty vector (as a control) and then infected with SeV. Total RNA was extracted and subjected to RT-qPCR to quantify IFN-β mRNA. (F to H) 293T cells were infected with lentiviruses carrying sh-Cdc25A#1 and sh-Cdc25A#2 to generate stable cell lines. (F) Cell lysates were collected and subjected to immunoblotting to verify the efficiency of Cdc25A knockdown. The cells were transfected with the IFN-β luciferase reporter plasmid and then infected with SeV. (G) The cell lysates were harvested for luciferase assay. (H) Total RNA was extracted for RT-qPCR to quantify IFN-β mRNA. (I and J) HEK293T cells were transfected with si-Cdc25A#1, si-Cdc25A#2, and control siRNA and then infected with SeV. (I) The cell lysates were collected for immunoblotting with the indicated antibodies. (J) The IFN-β mRNA levels were quantified by RT-qPCR. The results shown in graphs are means and standard deviations (SD) for three independent biological replicates per group. *, P < 0.05; **, P < 0.01; ***, P < 0.001. The data are representative of three independent experiments.

Cdc25A negatively regulates IFN-β transcription mainly via its phosphatase activity.

To determine whether Cdc25A inhibits IFN-β transcription via its phosphatase activity, we transfected 293T cells with plasmids expressing Cdc25A or its phosphatase activity-deficient mutant Cdc25A (C431S) (23, 24) and infected the cells with SeV. The immunoblotting data indicated that the levels of ectopically expressed Cdc25A and Cdc25A (C431S) were comparable (Fig. 2A). Reverse transcription-quantitative PCR (RT-qPCR) showed that Cdc25A attenuated the transcription of IFN-β mRNA, while Cdc25A (C431S) did not (Fig. 2B). We then transfected 293T cells with Cdc25A- and Cdc25A (C431S)-expressing plasmids and the IFN-β luciferase reporter and infected the cells with SeV. The luciferase assay indicated that Cdc25A could inhibit the SeV-induced activation of IFN-β, while Cdc25A (C431S) could not (Fig. 2C). Next, we performed a Cdc25A knockdown and rescue experiment. We transfected Cdc25A knockdown 293T cells with pCMV myc-Cdc25A or pCMV myc-Cdc25A (C431S) and infected the cells with SeV (Fig. 2D). RT-qPCR indicated that wild-type (WT) Cdc25A, but not Cdc25A (C431S), could reverse the effect of sh-Cdc25A on IFN-β transcription (Fig. 2E). We transfected Cdc25A knockdown 293T cells with pCMV myc-Cdc25A or pCMV myc-Cdc25A (C431S) together with the IFN-β luciferase reporter and infected the cells with SeV. Consistently, the luciferase assay demonstrated that WT Cdc25A, but not Cdc25A (C431S), could overturn the effect of sh-Cdc25A on activation of the IFN-β promoter (Fig. 2F). These results demonstrate that Cdc25A inhibits IFN-β transcription mainly via its phosphatase activity.

FIG 2.

Cdc25A restrains IFN-β transcription primarily via its phosphatase activity. (A and B) HEK293T cells were transfected with pCMV myc-Cdc25A or pCMV myc-Cdc25A (C431S) for 24 h and then infected with SeV for 8 h. (A) The cell lysates were collected for immunoblotting with the indicated antibodies. (B) Total RNA was extracted for RT-qPCR. (C) HEK293T cells were transfected with pCMV myc-Cdc25A and pCMV myc-Cdc25A (C431S) together with an IFN-β luciferase reporter for 24 h and then infected with SeV for 8 h. The cell lysates were harvested for luciferase assay. (D and E) Cdc25A knockdown 293T cells were transfected with pCMV myc-Cdc25A or pCMV myc-Cdc25A (C431S) and then infected with SeV. (D) Cdc25A levels were detected by immunoblotting with a Cdc25A antibody. (E) The IFN-β mRNA levels were determined by RT-qPCR. (F) Cdc25A knockdown 293T cells were transfected with pCMV myc-Cdc25A or pCMV myc-Cdc25A (C431S) and an IFN-β luciferase reporter and then infected with SeV. The cell lysates were harvested for luciferase assay. The data shown in graphs are means and SD for three independent biological replicates per group. *, P < 0.05; ***, P < 0.001; NS, not significant. The data are representative of three independent experiments.

Cdc25A attenuates IFN-β production induced by RLR-mediated signaling.

Next, we determined the level at which Cdc25A elicits its inhibitory effect on RLR signaling. We transfected 293T cells with the IFN-β luciferase reporter and the RIG-I, VISA, TBK1, IRF3 (WT and 5D), or p65 expression plasmid together with pCMV myc-Cdc25A. The luciferase assay showed that Cdc25A inhibited RIG-I-, VISA-, and TBK1-mediated IFN-β luciferase activation but not IRF3-, IRF3 (5D)-, or p65-induced IFN-β luciferase activation (Fig. 3A and B), implying that Cdc25A restrains RLR signaling at the TBK1 level. Therefore, we examined whether Cdc25A can interact with TBK1 by use of a coimmunoprecipitation (co-IP) assay. The data showed that myc-Cdc25A interacted with FLAG-TBK1 (Fig. 3C). In addition, we confirmed that endogenous Cdc25A and TBK1 can interact with each other by use of a co-IP assay (Fig. 3D). These data suggest that Cdc25A inhibits RLR-mediated activation of IFN-β at the TBK1 level.

FIG 3.

Cdc25A attenuates IFN-β production induced by RLR-mediated signaling. (A and B) HEK293T cells were transfected with the indicated plasmids, an IFN-β luciferase reporter, and pCMV myc-Cdc25A or control vector for 24 h. The cell lysates were harvested for immunoblotting (A) and luciferase assay (B). (C) HEK293T cells were transfected with pCMV FLAG-TBK1 and/or pCMV myc-Cdc25A or with empty vector (as a control). The cell lysates were collected and immunoprecipitated (IP) with anti-FLAG beads and then immunoblotted with myc, FLAG, and β-actin antibodies. (D) HEK293T cells were infected with SeV for 8 h or left uninfected, and cell lysates were harvested for immunoprecipitation with an anti-Cdc25A antibody followed by immunoblotting with TBK1, Cdc25A, and β-actin antibodies. The results shown in panel B are means and SD for three independent biological replicates per group. *, P < 0.05; **, P < 0.01; ***, P < 0.001; NS, not significant. Data in panels A and B are representative of three independent experiments, and those in panels C and D are representative of two independent experiments.

Cdc25A inhibits activation of TBK1 via its phosphatase activity.

Because the status of phosphorylation at TBK1 S172 represents its kinase activity (20, 25), we examined whether Cdc25A affects the level of TBK1 S172 phosphorylation. We transfected HEK293T cells with pCMV FLAG-TBK1 and pCMV myc-Cdc25A or pCMV myc-Cdc25A (C431S) and then infected the cells with SeV. The cell lysates were harvested for immunoblotting. The data showed that the intensity of phosphorylation at TBK1 S172 in Cdc25A-expressing cells was greatly reduced compared to that in control and Cdc25A (C431S) mutant-expressing cells. Consistently, the level of phosphorylation of the TBK1 downstream substrate IRF3 was also lower in Cdc25A-expressing cells than in control and Cdc25A (C431S) mutant-expressing cells (Fig. 4A). We then analyzed the effect of Cdc25A knockdown on the phosphorylation of TBK1 S172. As shown in Fig. 4B, the level of phosphorylation of TBK1 S172 was higher in sh-Cdc25A-treated cells than in control cells. Furthermore, we performed a Cdc25A rescue experiment. The data showed that ectopically expressed wild-type Cdc25A could counteract the effect of sh-Cdc25A on TBK1 phosphorylation, while the Cdc25A (C431S) mutant could not (Fig. 4C).

FIG 4.

Cdc25A inhibits activation of TBK1 primarily via its phosphatase activity. (A) 293T cells were transfected with pCMV FLAG-TBK1 and pCMV myc-Cdc25A or pCMV myc-Cdc25A (C431S) together with an IFN-β luciferase reporter and then infected with SeV or left uninfected. The cell lysates were harvested and subjected to immunoblotting with the indicated antibodies. (B) 293T-sh-Cdc25A cells and 293T-Ctrl cells were stimulated with SeV for 8 h. The cell lysates were collected for immunoblotting with the indicated antibodies. (C) 293T-sh-Cdc25A cells were transfected with pCMV myc-Cdc25A (WT or C431S) or control vector. The transfected cells and 293T-Ctrl cells were then infected with SeV or left uninfected. The cell lysates were collected for immunoblot analysis. Data in panels A and C are representative of two independent experiments, and those in panel B are representative of five independent experiments. The numbers below the bands indicate the mean ratios of the values for independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001; NS, not significant.

Next, we examined whether Cdc25A inhibits TBK1-mediated activation of IFN-β via its phosphatase activity. We transfected 293T cells with pCMV FLAG-TBK1 and pCMV myc-Cdc25A or pCMV myc-Cdc25A (C431S) together with the IFN-β luciferase reporter. The luciferase assay indicated that overexpression of Cdc25A greatly inhibited TBK1-mediated activation of the IFN-β promoter, while Cdc25A (C431S) only slightly reduced the luciferase activity induced by TBK1 (Fig. 5A). Consistently, we observed that Cdc25A, but not Cdc25A (C431S), inhibited the phosphorylation of TBK1 (Fig. 5B). To test whether Cdc25A can directly reduce the phosphorylation of TBK1 S172, we transfected 293T cells with pCMV FLAG-TBK1. FLAG-TBK1 was immunoprecipitated with anti-FLAG beads, incubated with GST-Cdc25A or glutathione S-transferase (GST) (as a control), and then subjected to immunoblotting. Simultaneously, the phosphorylation of Cdc2 Y15 was detected as a positive control (23). As shown in Fig. 5C, Cdc25A caused a reduction of TBK1 S172 phosphorylation. In addition, we examined TBK1 S172 phosphorylation in Cdc25A knockdown cells and found that knockdown of Cdc25A enhanced TBK1 S172 phosphorylation (Fig. 5D). Taken together, these results suggest that Cdc25A inhibits activation of TBK1 via its phosphatase activity.

FIG 5.

Cdc25A reduces phosphorylation of TBK1 S172. (A) 293T cells were transfected with pCMV FLAG-TBK1 and pCMV myc-Cdc25A or pCMV myc-Cdc25A (C431S) together with an IFN-β luciferase reporter and then infected with SeV. The cell lysates were harvested for luciferase assay. (B) 293T cells were transfected with pCMV FLAG-TBK1 and pCMV myc-Cdc25A or pCMV myc-Cdc25A (C431S). The cell lysates were collected for immunoblotting with the indicated antibodies. (C) FLAG-TBK1 and myc-Cdc2 were immunoprecipitated from 293T cells that were transfected with pCMV FLAG-TBK1 or pCMV myc-Cdc2, and GST-Cdc25A or GST was purified from E. coli. An in vitro phosphatase assay was performed, and the reaction products were immunoblotted with the indicated antibodies. (D) 293T-Ctrl and 293T-sh-Cdc25A cells were transfected with pCMV FLAG-TBK1. The cell lysates were harvested for immunoblot analysis. The results shown are means and SD for three independent biological replicates per group. *, P < 0.05; **, P < 0.01; ***, P < 0.001; NS, not significant. The data in panels A, B, and D are representative of three independent experiments, and those in panel C are representative of two independent experiments. The numbers below the bands indicate the mean ratios of the values from independent experiments.

Cdc25A reduces IFN-β production induced by DNA viruses.

In addition to RNA viruses, DNA viruses can also activate TBK1 and induce IFN-β production (26, 27). To investigate whether Cdc25A can negatively regulate IFN-β production induced by DNA viruses, we ectopically expressed Cdc25A in HEK293T cells and infected the cells with herpes simplex virus 1 (HSV-1). We found that Cdc25A could inhibit the activation of the IFN-β luciferase reporter (Fig. 6A) as well as the transcription of IFN-β mRNA (Fig. 6B) induced by HSV-1. These data indicate that Cdc25A can inhibit DNA virus-induced activation of IFN-β.

FIG 6.

Cdc25A inhibits IFN-β production induced by DNA viruses. (A and B) HEK293T cells were transfected with pCMV myc-Cdc25A and an IFN-β luciferase reporter plasmid and then infected with HSV-1 (MOI = 0.02). (A) The cell lysates were collected for luciferase assay. (B) Total RNA was extracted for RT-qPCR to quantify IFN-β mRNA. The results shown are means and SD for three independent biological replicates per group. *, P < 0.05; ***, P < 0.001.

Cdc25A negatively regulates the antiviral immune response.

To further investigate the role of Cdc25A in the antiviral immune response, we employed a vesicular stomatitis virus (VSV) infection assay as shown in Fig. 7A. Briefly, 293T cells either overexpressing Cdc25A or Cdc25A (C431S) or with knockdown of Cdc25A were transfected with poly(I·C). The supernatants were collected and applied to A549 cells. As shown in Fig. 7B, the mRNA levels for interferon-stimulated genes (ISGs) (MX1, ISG15, ISG20, IFIT1, and IFITM1) in A549 cells treated with supernatant from Cdc25A-expressing cells were significantly lower than those in A549 cells treated with supernatant from control or Cdc25A (C431S)-expressing cells. The levels of these ISG mRNAs in A549 cells treated with supernatant from sh-Cdc25A-transfected cells were clearly higher than those in A549 cells treated with supernatant from control cells (Fig. 7C). These results suggest that Cdc25A inhibits type I interferon expression in poly(I·C)-transfected 293T cells. Next, we infected the supernatant-treated A549 cells with green fluorescent protein (GFP)-modified VSV (VSV-GFP) and examined VSV M gene expression by RT-qPCR. The data showed that the amount of VSV M mRNA was higher in A549 cells treated with supernatant from Cdc25A-expressing cells than in A549 cells treated with supernatant from control or Cdc25A (C431S)-expressing cells (Fig. 7D). The level of VSV M mRNA in A549 cells treated with supernatant from sh-Cdc25A-transfected cells was clearly lower than that in A549 cells treated with supernatant from control cells (Fig. 7E). Fluorescence-activated cell sorter (FACS) analysis indicated that the ratio of VSV-GFP-positive cells for A549 cells treated with supernatant from Cdc25A-expressing cells was higher than that for control groups (Fig. 7F and H). Consistently, the ratio of VSV-GFP-positive cells for A549 cells treated with supernatant from sh-Cdc25A-transfected cells was lower than that for control cells (Fig. 7G and I). Taken together, these results indicate that Cdc25A negatively regulates the antiviral immune response.

FIG 7.

Cdc25A restrains the antiviral response. (A) Flowchart of the experimental procedure. (B to I) 293T cells transfected with pCMV myc-Cdc25A, pCMV myc-Cdc25A (C431S), or the control plasmid and Cdc25A knockdown 293T cells were transfected with poly(I·C) for 12 h. The supernatants were collected and added to A549 cells for 6 h. (B and C) One set of cells was collected, and total RNA was extracted for RT-qPCR to quantify expression of the indicated genes. The second set of A549 cells was infected with VSV-GFP (MOI = 0.01) for 8 h. The cells were then collected for total RNA extraction to quantify VSV M gene expression by RT-qPCR (D or E) or subjected to flow cytometry analysis (F and G), and the ratios of GFP-positive cells were calculated (H and I). The results shown in the graphs are means and SD for three independent biological replicates per group. *, P < 0.05; **, P < 0.01; ***, P < 0.001; NS, not significant. The data are representative of three independent experiments.

In summary, we demonstrated that the phosphatase Cdc25A negatively regulates the virus-induced innate immune response. Our studies show that Cdc25A inhibits RLR-mediated signaling mainly via its phosphatase activity. We found that Cdc25A reduces TBK1 activity by inhibiting its active phosphorylation and consequently attenuates the activation of IFN-β transcription as well as the antiviral status of nearby cells. These findings reveal Cdc25A as an important negative regulator of antiviral immunity.

DISCUSSION

Several phosphatases have been identified as being involved in the innate immune response or the cross talk between innate and adaptive immunity. For example, the phosphatase PP1γ dephosphorylates RIG-I and MDA5 and plays an essential role in innate immune signaling (16). The phosphatase SHP-1 negatively regulates TLR-mediated production of proinflammatory cytokines but, conversely, increases TLR- and RIG-I-activated production of type I interferon by inhibiting the kinase IRAK1 (28). SHP-1 was also found to be involved in the regulation of cross-presentation in adaptive immunity by dephosphorylating the NADPH oxidase component p47 and inhibiting the activation of NOX2 on phagosomes (29). The phosphatase SHP-2 inhibits TLR4- and TLR3-mediated activation of IFN-β production as well as the TLR3-activated proinflammatory cytokines interleukin-6 (IL-6) and tumor necrosis factor alpha (TNF-α) (30). Here we demonstrate that the phosphatase Cdc25A negatively regulates the antiviral immune response by inhibiting TBK1 activity. This finding suggests that the dephosphorylation of key components in antiviral signaling plays an important role in balancing the innate immune response or maintaining homeostasis in host cells to prevent overreaction during viral infection. Thus, it may represent a precisely controlled self-protection system generated during the evolution of host cells.

Cdc25A has long been known as a key phosphatase in cell cycle control (31, 32); it removes inhibitory phosphorylation of cyclin-dependent kinases (CDKs) to activate their activity, consequently promoting cell cycle progression (33–35). We compared the sequences around the phosphorylation sites of CDKs and TBK1 and did not observe the conserved motif, suggesting that Cdc25A may pick up substrates with a relatively broad spectrum of primary sequences. However, whether the specificity of Cdc25A is determined by the secondary/tertiary structure of the substrate needs further clarification.

In response to stress, such as DNA damage, eukaryotic cells activate cell cycle checkpoints to prevent cell cycle progression through complex kinase signaling pathways. In this process, the kinases ATM/Chk2 and ATR/Chk1 are activated and phosphorylate Cdc25A, promoting its ubiquitin-mediated degradation (36, 37). Viral infection is a stress to host cells that causes dysregulation of the cell cycle. It has been reported that influenza A virus (IAV) infection causes a cell cycle arrest in G₁ phase that is favorable to viral transcription and replication (38). However, it is unclear whether IAV infection causes cell cycle arrest by promoting Cdc25A degradation or inhibiting Cdc25A phosphatase activity. If either is so, IAV infection may release the inhibitory effect of Cdc25A on antiviral immunity. In some cases, viral infection promotes cell cycle progression (39). It is possible that the virus interferes with cell cycle checkpoints to push cell cycle progression even if the cells are in a stressful situation. It will be interesting to study how Cdc25A balances its roles in cell cycle checkpoints and the antiviral immune response.

We previously reported that CK1γ1 inhibits both RIG-I and TLR signaling by phosphorylating NF-κB p65 and promoting its degradation (18). Because TBK1 is the common downstream component in RLRs/TLRs as well as DNA sensors, such as AIM2, cGAS, and DDX41-mediated signaling pathways (40), we postulate that Cdc25A may also inhibit TLR- and DNA sensor-mediated innate immune responses. It will be interesting to further investigate the function of Cdc25A in response to different pathogens.

MATERIALS AND METHODS

Plasmids and antibodies.

Cdc25A (accession number NM_001789) cDNA was cloned into pCMV-myc. RIG-I (accession number NM_014314), TBK1 (accession number NM_013254), IRF3 (accession number NM_001571), and p65 (accession number NM_021975) cDNAs were cloned into pCMV FLAG. pCMV FLAG-IRF3 (5D) was provided by Wenjun Liu (Institute of Microbiology, Chinese Academy of Sciences [CAS], Beijing, China). pCMV FLAG-VISA was provided by Hong-bin Shu (College of Life Sciences, Wuhan University, China). Cdc25A shRNA was cloned into PSIH1-GFP, which was provided by Jilong Chen (Institute of Microbiology, CAS, Beijing, China). The IFN-β luciferase reporter plasmid was provided by Hong Tang (Institut Pasteur of Shanghai, Chinese Academy of Science, China).

Horseradish peroxidase (HRP)-linked goat anti-rabbit (111-035-003) and goat anti-mouse (115-035-003) IgGs were obtained from Jackson ImmunoResearch. Mouse control IgG (sc-2025), mouse anti-Cdc25A (sc-70823), and rabbit anti-IRF3 (sc-9082) were purchased from Santa Cruz Biotechnology. Rabbit anti-myc (C3956) and anti-FLAG M2 affinity gel (A2220) were purchased from Sigma-Aldrich. Rabbit anti-TBK1 (CST-D1B4), rabbit anti-phospho-TBK1 (Ser172) (CST-5483), rabbit anti-phospho-IRF3 (Ser396) (CST-4947), and rabbit anti-phospho-Cdc2 (Tyr15) (CST-4539) were obtained from Cell Signaling Technology. Mouse anti-FLAG and mouse anti-myc agarose-conjugated beads were purchased from Sungene Biotech and Abmart, respectively.

Cell culture and virus infection.

HEK293T cells and A549 cells were maintained in Dulbecco's modified Eagle's medium (DMEM; Gibco) with 10% fetal bovine serum (Pan Biotech) at 37°C with 5% CO₂. HEK293T cells were infected with Sendai virus (SeV) at a multiplicity of infection (MOI) of 0.25 for 8 h and collected for further analysis. A549 cells were infected with a GFP-modified vesicular stomatitis virus (GFP-VSV) at an MOI of 0.01 for the time indicated in the relevant figure legend. The cells were collected, fixed with 1% paraformaldehyde, and subjected to flow cytometry (BD FACSCalibur).

GFP-VSV was provided by Wenjun Liu (Institute of Microbiology, CAS, Beijing, China) and was propagated in MDCK cells. The GFP-modified herpes simplex virus 1 (HSV-1) strain was from Jinghua Yan's lab (Institute of Microbiology, CAS, Beijing, China) and was propagated in Vero cells. SeV was propagated in 9-day-old specific-pathogen-free embryonated eggs (Merial) at 37°C for 3 days. Influenza virus RNA was isolated from A/WSN/33 (H1N1; WSN)-infected A549 cells by use of TRIzol reagent. Poly(I·C) was purchased from Sigma-Aldrich.

Coimmunoprecipitation.

HEK293T cells were lysed with lysis buffer (1% Triton, 10% glycerol, 150 mM NaCl, 20 mM HEPES, 1 mM EDTA, pH 7.4) containing protease inhibitors for 15 min on ice. The lysates were collected for immunoprecipitation with the corresponding antibody for 4 h and subjected to immunoblotting with the antibodies indicated in the figures.

RNA extraction and real-time quantitative PCR.

Total RNA of the cells was extracted by use of TRIzol (ToYoBo). cDNAs were synthesized by use of a reverse transcription kit (TRANs). The mRNA levels for the indicated genes were analyzed by use of a SYBR green qPCR master mix (Yeasen). The primers used for RT-qPCR were as follows: for human actin, 5′-GGATCAGCAAGCAGGAGTATG-3′ and 5′-AGAAAGGGTGTAACGCAACTAA-3′; for human IFN-β (accession number NM_002176), 5′-AGGACAGGATGAACTTTGAC-3′ and 5′-TGATAGACATTAGCCAGGAG-3′; for human MX1 (accession number NM_001144925), 5′-GGTGGTGGTCCCCAGTAATG-3′ and 5′-ACCACGTCCACAACCTTGTCT-3′; for human ISG15 (accession number NM_005101), 5′-GAGAGGCAGCGAACTCATCTT-3′ and 5′-CCAGCATCTTCACCGTCAGG-3′; for human ISG20 (accession number NM_001303236), 5′-TGGACTGCGAGATGGTGG-3′ and 5′-GGGTTCTGTAATCGGTGAT-3′; for human IFIT1 (accession number NM_001270930), 5′-GCCATTTTCTTTGCTTCCCCTA-3′ and 5′-TGCCCTTTTGTAGCCTCCTTG-3′; for human IFITM1 (accession number NM_003641), 5′-ACAGGAAGATGGTTGGCGAC-3′ and 5′-GTAGACTGTCACAGAGCCGAA-3′; and for VSV M, 5′-CGAGCGCTCCAATTGACAAA-3′ and 5′-GATCTGCCAATACCGCTGGA-3′.

RNA interference.

HEK293T cells were transfected with siRNAs specific for Cdc25A (si-Cdc25A#1 and si-Cdc25A#2) or with control siRNA by use of Lipofectamine 2000. The sequences of si-Cdc25A#1 and si-Cdc25A#2 are 5′-GCUGGGAAACAUCAGGAUU-3′ and 5′-GCAUGGACAUGACUGGAUA-3′, respectively. si-Cdc25A#2 specifically targets the 3′ untranslated region (3′UTR) of Cdc25A mRNA. The sequence of the control siRNA is 5′-UUCUCCGAACGUGUCACGU-3′. To generate Cdc25A knockdown cell lines, HEK293T cells were infected with a recombinant lentivirus carrying GFP and sh-Cdc25A. GFP-positive cells were sorted by FACS analysis 48 h after infection. The sequences of sh-Cdc25A#1 and sh-Cdc25A#2 are 5′-TGCTGGGAAACATCAGGATTT-3′ and 5′-GGCATGGACATGACTGGATAG-3′, respectively. sh-Cdc25A#2 targets the 3′UTR of Cdc25A mRNA. For the rescue experiment, sh-Cdc25A#2 was used. The sequence of the control shRNA is 5′-CAACAAGATGAAGAGCACCAA-3′.

Luciferase reporter assay.

HEK293T cells were seeded into 24-well plates (3 × 10⁵ per well) and transfected with the IFN-β luciferase reporter plasmid, the pRL-TK Renilla luciferase reporter plasmid as an internal control, and the indicated plasmids. The cell lysates were harvested and subjected to a luciferase assay using a dual-luciferase reporter assay system (Promega).

In vitro phosphatase assay.

GST-Cdc25A was induced with 0.2 mM isopropyl-β-d-thiogalactopyranoside (IPTG) at 18°C for 16 h, and Escherichia coli was lysed by use of lysozyme with grinding on ice.

HEK293T cells were transfected with pCMV FLAG-TBK1 or pCMV myc-Cdc2 for 24 h. FLAG-TBK1 (or myc-Cdc2) was immunoprecipitated with FLAG M2 affinity gel (or anti-myc agarose-conjugated beads) and washed with lysis buffer three times and with phosphatase buffer (20 mM Tris-HCl [pH 8.0], 2 mM EDTA, 2 mM dithiothreitol [DTT]) twice. The beads were then incubated with E. coli-expressed GST-Cdc25A (5 μg) or GST (5 μg) in phosphatase buffer at 30°C for 30 min. FLAG-TBK1 and myc-Cdc2 were resolved in SDS loading buffer and subjected to immunoblotting with anti-phospho-S172 TBK1 antibody and anti-phospho-Y15 Cdc2 antibody, respectively.