Skip to main content
NCBI home page
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2025 Jan 15;122(3):e2320644122. doi: 10.1073/pnas.2320644122

Direct lysine dimethylation of IRF3 by the methyltransferase SMYD3 attenuates antiviral innate immunity

Zixuan Wang a,b,c,d,e,1, Xiaoyun Chen a,b,c,d,e,1, Chunchun Zhu a,b,c,d,e, Sijia Fan a,b, Jinhua Tang a,b, Hongyan Deng a,b, Xueyi Sun a,b,c,d,e, Xing Liu a,b,c,d,e,2, Wuhan Xiao a,b,c,d,e,2
PMCID: PMC11761311  PMID: 39813248

Significance

As a key factor in the type I interferon signaling, Interferon regulatory factor 3 (IRF3) activation must be tightly controlled to efficiently activate innate immunity and avoid overactivation of innate immunity. However, how the IRF3 activation is controlled in the resting state and during viral infection remains largely unknown. In this study, we found that SMYD3 directly catalyzes the dimethylation of IRF3 at lysine 39, resulting in the inhibition of its activation. As a result, disruption of Smyd3 in mice and zebrafish can protect the host from viral infection. Our results suggest that SMYD3 is a negative regulator of the type I interferon signaling and reveal a specific modulation of IRF3.

Keywords: SMYD3, IRF3, innate immunity, methylation

Abstract

Interferon regulatory factor 3 (IRF3) is the key transcription factor in the type I IFN signaling pathway, whose activation is regulated by multiple posttranslational modifications. Here, we identify SMYD3, a lysine methyltransferase, as a negative regulator of IRF3. SMYD3 interacts with IRF3 and catalyzes the dimethylation of IRF3 at lysine 39. This modification reduces IRF3 phosphorylation, dimerization, and subsequent nuclear translocation, leading to the inhibition of downstream type I interferon production. In addition, Smyd3-deficient mice are more resistant to RNA and DNA viral infections. Zebrafish lacking smyd3 or treated with the inhibitor BCI121 are also more resistant to viral infection. Our findings reveal a role for SMYD3 in the regulation of antiviral innate immunity and provide insight into a specific modulation of IRF3 that affects its activation.

Innate immunity is the first line for the host to defend against invading viruses. Upon recognition of pathogen-associated molecular patterns and host damage-associated molecular patterns, pattern recognition receptors trigger innate immunity via activation of NF-κB, inflammasomes and type I interferons (IFNs) (1–4). As a key transcription factor in the antiviral innate immune response, interferon regulatory factor 3 (IRF3) is phosphorylated on multiple serine and threonine residues by the upstream kinases in response to viral infection (5). Phosphorylated IRF3 then activates the expression of type I IFNs and subsequent interferon-stimulated genes (ISGs) (3, 6). To efficiently activate innate immunity and to avoid overactivation of innate immunity leading to deleterious immunopathology, the activity of IRF3 is tightly regulated by multiple posttranslational modifications (PTMs) (5, 7–17). In fact, methylation is one of the PTMs that modulates IRF3 function, but it is poorly understood (18, 19).

The Su (Var)3–9, Enhancer-of-zeste and Trithorax (SET) and Myeloid, Nervy, and DEAF-1 (MYND) domain-containing (SMYD) proteins are a family of methyltransferases consisting of SMYD1, SMYD2, SMYD3, SMYD4, and SMYD5. Due to the lysine-specific methyltransferase activity in their SET domain, they act on histone and nonhistone targets to catalyze lysine methylation involved in a wide range of biological processes (20–23). SMYD3 is a member of the SMYD family and has been shown to play an important role in tumorigenesis and affecting transcriptional regulation (24–31). In addition, SMYD3 has been found to directly interact with the human T cell lymphotropic virus type 1 (HTLV-1) tax and support its cytoplasmic localization (32). SMYD3 can also be recruited by the Ebola virus nucleoprotein, resulting in facilitation of viral mRNA transcription (33). SMYD3-deficient mice infected with the respiratory syncytial virus exhibit exaggerated inflammatory responses and disease exacerbation (34). These observations link SMYD3 to host–pathogen interactions and host immunity. However, the role of SMYD3 in antiviral innate immunity remains elusive.

Here, we found that SMYD3 negatively regulates both RNA- and DNA virus–induced innate antiviral responses. Mechanistic studies indicate that SMYD3 catalyzes dimethylation of K39 on IRF3. SMYD3-mediated methylation attenuates phosphorylation, dimerization, and nuclear localization of IRF3 in response to viral infection, resulting in the attenuation of type I IFN signaling.

Results

SMYD3 Attenuates the Innate Antiviral Response Induced by RNA and DNA Viruses.

To determine whether the SMYD family has an influence on the cellular antiviral response, we overexpressed each SMYD gene in H1299 cells (p53-deficient) and then infected the cells with VSV-GFP virus (Fig. 1A). Compared to the empty vector (EV) control transfection, SMYD3 overexpression apparently promoted VSV-GFP virus replication (Fig. 1 A and B). Notably, SMYD2 has been also reported to inhibit interferon production during VSV infection in HEK293T cells and macrophages (35). After comparison, we found that SMYD3 has a stronger suppressive role than SMYD2 in IFN production during viral infection in H1299 cells (SI Appendix, Fig. S1A). In addition, Smyd3 was induced by VSV infection (SI Appendix, Fig. S1B). Luciferase activity assays indicated that overexpression of SMYD3 suppressed IFNβ promoter reporter activity induced by Sendai virus (SeV) infection (Fig. 1C). Consistently, overexpression of SMYD3 in HEK293T cells suppressed SeV-induced IFNβ, CXCL10, and IFIT1 mRNA expression compared to EV control transfection (Fig. 1D). Conversely, SeV-induced IFNβ, CXCL10, and CCL5 mRNA expression was higher in SMYD3-disrupted HEK293T cells (SMYD3−/−) than in wild-type (WT) HEK293T cells (SMYD3+/+) (Fig. 1E). Furthermore, induction of IFNβ promoter activity and the expression of IFNβ, CXCL10, CCL5, and IFIT1 mRNA by the transfection of poly I:C or poly A:T in HEK293T cells was also suppressed by SMYD3 overexpression and enhanced by SMYD3 knockout (SI Appendix, Fig. S1 C and D). Similar results were obtained in A549 cells (p53-intact) infected with herpes simplex virus 1 (HSV-1) or vesicular stomatitis virus (VSV) (Fig. 1 F and G and SI Appendix, Fig. S1E).

Fig. 1.

Fig. 1.

Open in a new tab

SMYD3 attenuates the cellular antiviral immune response. (A and B) The effect of ectopic expression of SMYDs on VSV-GFP replication in H1299 cells. (C) Luciferase activity of IFNβ promoter reporter in H1299 cells transfected with empty pCMV vector (EV) or pCMV-SMYD3 for 24 h, followed by uninfected (UI) or infected with SeV for 8 h. (D) qPCR analysis of IFNβ, CXCL10, and IFIT1 mRNA in HEK293T cells transfected with empty pCMV EV or pCMV-SMYD3, followed by UI or infected with SeV. (E) qPCR analysis of IFNβ, CXCL10, and CCL5 mRNA in SMYD3-intact (SMYD3+/+) or SMYD3-null (SMYD3−/−) HEK293T cells. (F) qPCR analysis of IFNβ, CXCL10, and CCL5 mRNA in A549 cells infected with pHAGE lentivirus (control) or pHAGE-SMYD3 lentivirus, followed by UI or infected with HSV-1 for 8 h. (G) qPCR analysis of IFNβ, CXCL10, and CCL5 mRNA in SMYD3-intact (SMYD3+/+) or SMYD3-null (SMYD3−/−) A549 cell UI or infected with HSV-1 for 8 h. (H) Microscopic imaging (Left panels) and viral mRNA detection (Right panels) of VSV replication in SMYD3-intact (SMYD3+/+) or SMYD3-null (SMYD3−/−) A549 cells. (I) Microscopy imaging of HSV-1-GFP virus replication in SMYD3-intact (SMYD3+/+) or SMYD3-null (SMYD3−/−) A549 cells. (J) Flow cytometric analysis of HSV-1 GFP virus replication in SMYD3-intact (SMYD3+/+) or SMYD3-null (SMYD3−/−) A549 cells UI or infected with HSV-1-GFP virus for 8 h. (K) Phosphorylation of IRF3 in A549 cells infected with pHAGE lentivirus (control) or pHAGE-SMYD3 lentivirus, followed by HSV-1 infection for 0, 4, and 8 h. Quantification for phosphorylation of IRF3 was shown in red. (L) Microscopic imaging of IRF3 localization in SMYD3-intact (SMYD3+/+) or SMYD3-null (SMYD3−/−) A549 cells UI or infected with HSV-1-GFP for 8 h. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. Graphs represent fold induction relative to the untreated cells. Data are presented as the mean values of a representative experiment performed in triplicate (C–G, and H) or as representative data (A, B, H, I, J, and L); these experiments were repeated independently at least three times, and error bars indicate S.D.

Consistently, HSV-1-GFP and VSV-GFP virus replication was reduced in SMYD3−/− A549 cells compared to in SMYD3 +/+ A549 cells as demonstrated by fluorescence microscopy imaging, viral mRNA expression, and flow cytometric analysis (Fig. 1 H–J). Furthermore, HSV-1 and VSV-induced IRF3 phosphorylation was reduced in A549 cells ectopically expressing SMYD3 by lentivirus infection (Fig. 1K and SI Appendix, Fig. S1F). As expected, HSV-1-induced IRF3 nuclear localization was enhanced in SMYD3−/− A549 cells compared to in SMYD3+/+ A549 cells (Fig. 1L).

We then examined whether the effect of SMYD3 on the type I interferon signaling was dependent on type I IFN using Vero cells (deficient for type I IFN production). Overexpression of SMYD3 in Vero cells had no apparent effect on VSV-GFP and HSV-1-GFP virus replication as demonstrated by fluorescence microscopy imaging, flow cytometric analysis, and immunoblotting with anti-GFP antibody (SI Appendix, Fig. S1 G–L).

Taken together, these data suggest that SMYD3 attenuates RNA- and DNA virus–induced antiviral innate immunity.

SMYD3 Attenuates Antiviral Innate Immunity by Targeting IRF3 for Lysine Dimethylation at Lysine 39.

To determine which molecule in the IRF3 signaling pathway is targeted by SMYD3, we first used the promoter luciferase assay. In HEK293T cells, cotransfection of SMYD3 significantly suppressed MAVS, TBK1, and IRF3-stimulated IFNβ promoter activity (SI Appendix, Fig. S2A). Given that SMYD3 attenuates both RNA- and DNA virus–induced antiviral innate immunity, SMYD3 may target IRF3 to negatively regulate the IRF3 signaling pathway. Coimmunoprecipitation assays showed that ectopically expressed SMYD3 interacted with ectopically expressed IRF3 (SI Appendix, Fig. S2 B and C). Endogenous SMYD3 also interacted with endogenous IRF3 (SI Appendix, Fig. S2D). Exogenous SMYD3 colocalized with IRF3 mainly in the cytosol, even upon VSV infection (SI Appendix, Fig. S2E). However, in response to VSV infection, the interaction between SMYD3 and IRF3 was enhanced as indicated by the in situ PLA and endogenous coimmunoprecipitation assays (SI Appendix, Fig. S2 F and G). Domain mapping revealed that the N terminus (1–133 aa) of IRF3 was required for IRF3 binding to SMYD3 (SI Appendix, Fig. S2 H–J) and that the C terminus (219–428 aa) of SMYD3 was required for SMYD3 binding to IRF3 (SI Appendix, Fig. S2 K and L). However, both the N terminus and the C terminus of SMYD3 were required for its suppressive roles in the innate antiviral response (SI Appendix, Fig. S3).

It appeared that disruption of SMYD3 or overexpression of the WT or its catalytically inactivated mutant, SMYD3-F183A (27), does not affect the stability of the IRF3 protein (Fig. 2 A–C). We then determined whether the enzymatic activity of SMYD3 is required for SMYD3 to inhibit antiviral innate immunity. Overexpression of the catalytically inactivated SMYD3 mutant, SMYD3-F183A, had no apparent effect on HSV-1 induced IFNβ or SeV-induced CXCL10 and CCL5 mRNA expression (Fig. 2 D and E). Upon poly I:C stimulation, reconstitution of SMYD3-F183A in SMYD3 knockout cells had no apparent effects on the ISRE reporter activity, ISG15 and IFIT1 mRNA expression, or IRF3 phosphorylation and dimerization (Fig. 2 F–H).

Fig. 2.

Fig. 2.

Open in a new tab

SMYD3 catalyzes the methylation of IRF3 at lysine 39. (A) Immunoblotting of endogenous IRF3 in SMYD3+/+ or SMYD3−/− HEK293T cells. (B) Immunoblotting of ectopic IRF3 protein in HEK293T cells transfected with WT SMYD3 or the enzymatically inactive mutant of SMYD3 (F183A) at increasing doses. (C) Immunoblotting of endogenous IRF3 in A549 cells infected with pHAGE-SMYD3, or pHAGE-SMYD3(F183A) lentiviruses. (D) qPCR analysis of IFNβ in A549 cells transfected with empty vector (EV), WT SMYD3 (WT), or the enzymatically inactive mutant of SMYD3 (F183A), followed by UI or infected with HSV-1 for 8 h. (E) qPCR analysis of CXCL10 and CCL5 in HEK293T cells transfected with empty vector (EV), WT SMYD3 (WT), or the enzymatically inactive mutant of SMYD3 (F183A), followed by UI or infected with SeV for 8 h. (F) Luciferase activity of the ISRE reporter in SMYD3−/− HEK293T cells transfected with empty vector (EV), WT SMYD3 (WT), or the enzymatically inactive mutant of SMYD3 (F183A), followed by untransfected (UT) or transfected with poly I:C for 8 h. (G) qPCR analysis of ISG15 and IFIT1 in SMYD3−/− HEK293T cells transfected with empty vector (EV), WT SMYD3 (WT), or the enzymatically inactive mutant of SMYD3 (F183A), followed by untransfected (UT) or transfected with poly I:C for 8 h. (H) Phosphorylation and dimerization of IRF3 in SMYD3−/− HEK293T cells transfected with empty vector (EV), WT SMYD3 (WT), or the enzymatically inactive mutant of SMYD3 (F183A), followed by untransfected (UT) or transfected with poly I:C for 0, 4, and 8 h. (I) Scheme of detection of IRF3 methylation by mass spectrometry analysis. (J) Mass spectrometry analysis revealed that lysine 39 in IRF3 was dimethylated by SMYD3. (K) Luciferase activity of the ISRE reporter in IRF3−/− H1299 cells transfected with SMYD3 together with WT IRF3 (WT) or the lysine to arginine mutant (all the seven lysine residues, including K29, K39, K70, K193, K313, K315, and K409, were simultaneously mutated to arginine residues). (L) Luciferase activity of the ISRE reporter in HEK293T cells transfected with SMYD3 together with WT IRF3 (WT) or the lysine to arginine mutants, including K29R, K39R, K70R, K193R, K313R, K315R, or K409R. (M) Luciferase activity of the IFNβ promoter reporter in IRF3−/− H1299 cells transfected SMYD3 together with WT IRF3 (WT), or its lysine to arginine mutants. (N) Luciferase activity of the ISRE reporter in IRF3−/− H1299 cells transfected with SMYD3 together with WT IRF3 (WT), or its lysine to arginine mutants, including K29, K39, K70, K193, K313, K315, or K409 (In which only K29, K39, K70, K193, K313, K315, or K409, was kept intact respectively and other six lysine residues were simultaneously mutated to arginine residues). (O) qPCR analysis of IFIT1 in SMYD3+/+ and SMYD3−/− HEK293T cells transfected with empty vector (EV), WT IRF3 (WT), or IRF3-K39R mutant. (P) Sequence alignment of partial IRF3 (34-50 amino acids) from human, mouse, rat, cow, pig, macaque, dog, and zebrafish. The red box indicates a conserved lysine (K39). (Q) Dot blot assay for the specificity of the anti-IRF3-K39Me2 antibody. Equal amounts of dimethylated peptides or the control peptides were immunoblotted with the indicated dilutions of anti-IRF3-K39Me2 antibody. (R) HEK293T cells were transfected with empty Flag vector (−), Flag-IRF3 (WT), or Flag-IRF3-K39R, followed by immunoprecipitating with anti-Flag antibody, and immunoblotting with anti-IRF3-K39Me2 antibody. (S) Cell lysates from IRF3-intact or IRF3-null H1299 cells (IRF3+/+ or IRF3−/−) were extracted, followed by immunoprecipitation with anti-IRF3 antibody, and immunoblotting with anti-IRF3 and anti-IRF3-K39Me2 antibodies. (T) Cell lysates from SMYD3-intact or SMYD3-null HEK293T cells (SMYD3+/+ or SMYD3−/−) were extracted, followed by immunoprecipitation with anti-IRF3 antibody, and immunoblotting with anti-IRF3 and anti-IRF3-K39Me2 antibodies. (U) Scheme for the in vitro methylation assay and detection of IRF3 methylation by mass spectrometry analysis. (V) Mass spectrometry analysis revealed that lysine 39 in IRF3 was methylated by bacterially expressed SMYD3. (W) IRF3 methylation at lysine 39 was steadily increased after VSV infection. Anti-IRF3 antibody was used for Co-IP, and anti-IRF3-K39Me2 antibody was used for detection. IP, immunoprecipitation; TCL, total cell lysate; empty vector (EV); uninfected (UI); UT, untransfected; ns, not significant, *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. Graphs represent fold induction relative to the untreated cells. Data are presented as the mean values of a representative experiment performed in triplicate (D–G and K–O) or as representative data (A–C, H, R–T, and W); these experiments were repeated independently at least three times, and error bars indicate S.D.

To further validate that the enzymatic activity is required for the role of SMYD3 on antiviral innate immunity, we used a specific inhibitor of SMYD3 methyltransferase, BCI121, for assays (36). BCI121 did not affect HEK293T apoptosis and viability when used in the range of 20 to 200 μM (SI Appendix, Fig. S4 A and B). Upon challenge with SeV, the expression of IFNβ, CXCL10, and ISG15 was increased by BCI121 treatment, reaching the highest level at 50 μM (SI Appendix, Fig. S4C). In response to SeV or VSV infection, treatment with BCI121 (50 μM) significantly promoted IFNβ, CXCL10, or Isg15 expression and inhibited VSV-GFP virus replication in THP-1 or BMDCs (SI Appendix, Fig. S4 D–G). In Vero cells, addition of BCI121 (50 μM) had no apparent effect on VSV-GFP and HSV-GFP virus replication (SI Appendix, Fig. S4 H–K), further confirming the requirement of type I IFN in mediating the role of SMYD3. These data suggest that the repressive role of SMYD3 on the innate antiviral response is dependent on the enzymatic activity of SMYD3 as a methyltransferase. Furthermore, knockdown of SMYD3 by siRNA, inhibition of SMYD3 by BCI121 treatment, or knockout of SMYD3 by CRISPR/Cas9, induced IFNβ, ISG15, and CXCL10 mRNA expression in IRF3-intact cells, but not in IRF3-deficient cells (SI Appendix, Fig. S5 A–C), further confirming that the suppressive role of SMYD3 on the innate antiviral response is dependent on IRF3.

We then used mass spectrometry analysis to identify the methylated residues in IRF3 targeted by SMYD3 (Fig. 2I). The results showed that the seven lysine residues of IRF3 might be dimethylated or trimethylated by SMYD3 (Fig. 2J and SI Appendix, Fig. S6 A–I). To validate which lysine of IRF3 was actually targeted by SMYD3 to mediate the repressive function of SMYD3 on IRF3 activation, we mutated all of these seven lysine residues to arginine residues simultaneously or respectively and examined the effect of SMYD3 on these mutant-induced innate antiviral response. Overexpression of SMYD3 could not suppress IRF3-K39R-induced luciferase reporter activity or target gene expression, but could suppress IRF3-K39 (in which only Lysine 39 was kept intact and other six lysine residues, K29, K70, K193, K313, K315, and K409, were simultaneously mutated to arginine residues)-induced ISRE reporter activity, suggesting that SMYD3 may target lysine 39 of IRF3 to catalyze its dimethylation (Fig. 2 K–O). This lysine residue is evolutionarily conserved (Fig. 2P). We then generated an antibody against IRF3-K39 Me2 and validated its specificity by the dot blot assay (Fig. 2Q). Overexpression of WT IRF3 but not IRF3-K39R was detected by the anti-IRF3-K39 Me2 antibody, further confirming the specificity of the antibody (Fig. 2R). As expected, dimethylation of IRF3 was detected by anti-IRF3-K39 Me2 antibody in WT H1299 cells (IRF3+/+), but not in IRF3-deficient H1299 cells (IRF3−/−) (Fig. 2S). Furthermore, dimethylation of IRF3 was detected in WT HEK293T cells (SMYD3+/+), but not in SMYD3-deficient HEK293T cells (SMYD3−/−) (Fig. 2T).

To validate that SMYD3 can directly catalyze IRF3 methylation, we performed in vitro methylation assays using bacterially expressed SMYD3 and IRF3. Mass spectrometry analysis showed that SMYD3 could directly catalyze the dimethylation of IRF3 at lysine 39 (Fig. 2 U and V). Furthermore, mass spectrometry analysis showed that dimethylation of IRF3 at lysine 39 was detected in WT HEK293T cells (SMYD3+/+), but not in SMYD3-deficient HEK293T cells (SMYD3−/−) (SI Appendix, Fig. S6 J and K), confirming that the endogenous SMYD3 could catalyze the dimethylation of endogenous IRF3. In addition, it appeared that the dimethylation of IRF3 at lysine 39 was enhanced with the prolongation of VSV infection time (Fig. 2W).

To determine how viral infection triggers the de-dimethylation of IRF3, we sought to identify IRF3-associated proteins and then searched for proteins with de-dimethylase activity. Mass spectrometry analysis revealed the presence of two KDM family proteins, KDM2A and KDM8, which were identified as putative IRF3-associated proteins (SI Appendix, Fig. S7A). By knockdown by siRNA or overexpression, we found that KDM2A suppressed innate immunity, whereas KDM8 enhanced innate immunity upon VSV infection, suggesting that KDM8 may be the de-dimethylase responsible for IRF3 de-dimethylation (SI Appendix, Fig. S7 B to D). Coimmunoprecipitation assays revealed that KDM8 interacts with IRF3 in response to poly I:C stimulation (SI Appendix, Fig. S7E). Upon VSV infection, overexpression of KDM8 suppressed the level of IRF3 dimethylation at K39 (SI Appendix, Fig. S7F). In addition, overexpression of KDM8 increased the mRNA levels of IFNβ, CXCL10, and IFIT1 induced by WT IRF3, but not by IRF3-K39R (SI Appendix, Fig. S7G). Furthermore, KDM8 did not interact with IRF3 in the absence of VSV infection. However, in response to VSV infection, the endogenous KDM8 interacted with endogenous IRF3 (SI Appendix, Fig. S7H). Taken together, these data suggest that viral infection induces the interaction between KDM8 and IRF3, leading to the initiation of the de-dimethylation of IRF3 at K39, thereby activating IRF3.

To evaluate whether the Lys-39 methylation of IRF3 mediates the inhibitory role of SMYD3 in antiviral innate immunity, we used the methylation mimic mutant of IRF3 at lysine 39, IRF3-K39F (19), to perform a series of assays. Overexpression of WT IRF3 significantly enhanced IFNβ promoter reporter and ISRE reporter activity in WT HEK293T cells and IRF3−/− H1299 cells, but overexpression of IRF3-K39F did not (Fig. 3 A and B). In addition, overexpression of WT IRF3 promoted poly I:C-induced IFNβ and CXCL10 mRNA expression in HEK293T cells, but overexpression of IRF3-K39F did not (Fig. 3C). Furthermore, overexpression of WT IRF3 promoted HSV-1-stimulated IFNβ and CXCL10 mRNA expression in IRF3−/− H1299 cells, but overexpression of IRF3-K39F did not (Fig. 3D). Consistently, overexpression of WT IRF3 suppressed HSV-1-GFP and VSV-GFP virus replication in IRF3−/− H1299 cells, but overexpression of IRF3-K39F did not, as shown by flow cytometry analysis, fluorescence microscopy imaging and immunoblotting with anti-GFP antibody (Fig. 3 E–G). Similar results were obtained when lysine 39 was mutated to methionine or leucine (Fig. 3 H–M). Consistently, in response to VSV infection, WT IRF3 was greatly phosphorylated and predominantly detected in the nuclei of IRF3−/− H1299 cells, whereas IRF3-K39F was barely phosphorylated and mainly detected in the cytosol of IRF3−/− H1299 cells (Fig. 3 N and O). SMYD3 could not interact with TBK1 (SI Appendix, Fig. S8 A and B), whereas in response to VSV infection, IRF3 methylation at K39 inhibits its interaction with TBK1 (SI Appendix, Fig. S8C), which may account for the inhibitory effect of IRF3 methylation at K39 on IRF3 phosphorylation. Therefore, methylation of IRF3 at lysine 39 results in the loss of its activity.

Fig. 3.

Fig. 3.

Open in a new tab

IRF3 is inactivated by methylation at lysine 39. (A) Luciferase activity of IFNβ promoter reporter and ISRE reporter in HEK293T cells transfected with empty Flag vector, WT IRF3 (Flag-IRF3-WT), or the methylation mimic mutant (Flag-IRF3-K39F). (B) Luciferase activity of the IFN-β promoter reporter and ISRE reporter in IRF3−/− H1299 cells transfected with empty Flag vector, WT IRF3 (Flag-IRF3-WT), or the methylation mimic mutant (Flag-IRF3-K39F). (C) qPCR analysis of IFNβ and CXCL10 mRNA in HEK293T cells transfected with empty Flag vector, WT IRF3 (Flag-IRF3-WT), or the methylation mimic mutant (Flag-IRF3-K39F), followed by transfection with (+) or without (−) poly I: C. (D) qPCR analysis of IFNβ and CXCL10 mRNA in IRF3−/− H1299 cells transfected with empty Flag vector, WT IRF3 (Flag-IRF3-WT), or the methylation mimic mutant (Flag-IRF3-K39F), followed by UI or infected with HSV-1. (E) Flow cytometric analysis of HSV-1-GFP virus replication in IRF3−/− H1299 cells transfected with empty Flag vector, WT IRF3 (Flag-IRF3-WT), or the methylation mimic mutant (Flag-IRF3-K39F), followed by UI or infected with HSV-1-GFP virus. (F) Microscopic imaging of VSV-GFP virus replication in IRF3−/− H1299 cells transfected with empty Flag vector, WT IRF3 (Flag-IRF3-WT), or the methylation mimic mutant (Flag-IRF3-K39F), followed by infection with VSV-GFP virus. (G) Immunoblotting analysis of VSV-GFP virus replication in IRF3−/− H1299 cells transfected with empty Flag vector, WT IRF3 (Flag-IRF3-WT), or the methylation mimic mutant (Flag-IRF3-K39F), followed by infection with VSV-GFP virus. (H) Luciferase activity of the IFN-β promoter reporter and ISRE reporter in IRF3−/− H1299 cells transfected with empty HA vector, WT IRF3 (HA-IRF3-WT), or the methylation mimic mutant (HA-IRF3-K39M or HA-IRF3-K39L mutant). (I) qPCR analysis of IFNβ and CXCL10 mRNA in IRF3−/− H1299 cells transfected with empty HA vector, WT IRF3 (HA-IRF3-WT), or the methylation mimic mutant (HA-IRF3-K39M or HA-IRF3-K39L mutant), followed by UI or infected with VSV. (J and K) Flow cytometric analysis of VSV-GFP virus replication in IRF3−/− H1299 cells transfected with empty HA vector, WT IRF3 (HA-IRF3-WT), or the methylation mimic mutant (HA-IRF3-K39M or HA-IRF3-K39L mutant), followed by UI or infected with VSV-GFP virus. (L) Immunoblotting analysis of VSV-GFP virus replication in IRF3−/− H1299 cells transfected with empty HA vector, WT IRF3 (HA-IRF3-WT), or the methylation mimic mutant (HA-IRF3-K39M or HA-IRF3-K39L mutant), followed by infection with VSV-GFP virus. (M) Microscopic imaging of VSV-GFP virus replication in IRF3−/− H1299 cells transfected with empty HA vector, WT IRF3 (HA-IRF3-WT), or the methylation mimic mutant (HA-IRF3-K39M or HA-IRF3-K39L mutant), followed by infection with VSV-GFP virus. (N) Phosphorylation of IRF3 in IRF3−/− H1299 cells transfected with empty Flag vector, WT IRF3 (Flag-IRF3-WT), or the methylation mimic mutant (Flag-IRF3-K39F), followed by VSV infection for 8 h.(O) Localization of WT IRF3 (HA-IRF3-WT) and the methylation mimic mutant (HA-IRF3-K39F) in IRF3−/− H1299 cells UI or infected with VSV. ****P < 0.0001. Graphs represent fold induction relative to the untreated cells. Data are presented as the mean values of a representative experiment performed in triplicate (A–D, H, I, and K) or as representative data (E–G, J, and L–O); these experiments were repeated independently at least three times, and error bars indicate S.D.

Taken together, these data suggest that SMYD3 catalyzes the dimethylation of IRF3 at lysine 39, resulting in the inactivation of IRF3, thereby suppressing the type I interferon signaling. In addition, KDM8 may serve as a de-dimethylase to remove the dimethylation of IRF3 at lysine 39 upon viral infection, resulting in the release of SMYD3 suppression of IRF3 activation.

Smyd3-Deficient Mice Are More Resistant to Viral Infection.

To further evaluate the function of Smyd3 in the type I IFN signaling in vivo, we generated Smyd3-deficient mice using CRISPR-Cas9 (SI Appendix, Fig. S9A). After intercrossing Smyd3+/- mice, Smyd3−/− mice were born in the expected Mendelian ratio (SI Appendix, Fig. S9B). No gross defects were observed in Smyd3−/− mice, including body weight (SI Appendix, Fig. S9C). In addition, the lymphocyte numbers or proportions in the spleen and intestinal lymph nodes were similar between Smyd3+/+ and Smyd3−/− mice as revealed by flow cytometry analysis (SI Appendix, Fig. S9 D and E).

To evaluate the importance of Smyd3 in the host against virus infection, we intraperitoneally injected 6 to 8-wk-old Smyd3+/+ and Smyd3−/− mice with VSV and monitored their survival. The results showed that Smyd3−/− mice were more resistant to VSV-induced death than the Smyd3+/+ controls (Fig. 4A). Hematoxylin and eosin (H & E) staining of lung tissues revealed greater immune cell infiltration and injury in the lungs of Smyd3+/+ mice compared with those of Smyd3−/− mice, after infection with VSV or HSV-1 (Fig. 4B). Consistently, the Ifn-β level in sera or BMDCs of Smyd3−/− mice infected with VSV for 24 h was higher than that of Smyd3+/+ mice (Fig. 4 C and D). Furthermore, the mRNA level of Ifnβ, Cxcl10, Isg15, or Ifit1 in the spleen, liver, and lung of Smyd3−/− mice was higher than that of Smyd3+/+ mice after HSV-1 or VSV infection for 24 h (Fig. 4 E–H).

Fig. 4.

Fig. 4.

Open in a new tab

Disruption of Smyd3 in mice results in increased susceptibility to lethal viral infection. (A) Survival (Kaplan–Meier curve) of Smyd3+/+ (n = 10) and Smyd3−/− (n = 10) mice by intraperitoneal injection of VSV (1 × 107 plaque-forming units (PFU)/ per mouse) and monitored for 10 d. **P < 0.01. (B) Hematoxylin and eosin-stained images of lung sections from Smyd3+/+ or Smyd3−/− mice injected intraperitoneally with phosphate buffer saline (PBS) (vehicle control), VSV (1 × 107 PFU/per mouse) or HSV-1 (5 × 107 PFU/per mouse) for 24 h. (C) ELISA of Ifnβ in the serum of Smyd3+/+ or Smyd3−/− mice injected intraperitoneally with VSV (1 × 107 PFU/per mouse) for 24 h. (D) ELISA of Ifn-β in the supernatant of Smyd3+/+ or Smyd3−/− BMDCs infected with VSV for 8 h. (E) qPCR analysis of Ifnβ, Cxcl10, and Isg15 mRNA in spleens of Smyd3+/+ or Smyd3−/− mice injected intraperitoneally with PBS or HSV-1 (5 × 107 PFU/per mouse) for 24 h. (F) qPCR analysis of Ifnβ, Cxcl10, and Isg15 mRNA in spleens of Smyd3+/+ or Smyd3−/− mice injected intraperitoneally with PBS or VSV (1 × 107 PFU/per mouse) for 24 h. (G) qPCR analysis of Cxc10, Ifit1, and Isg15 mRNA in the liver of Smyd3+/+ or Smyd3−/− mice injected intraperitoneally with PBS or VSV (1 × 107 PFU/per mouse) for 24 h. (H) qPCR analysis of Cxcl10, Ifit1, and Isg15 mRNA in the lungs of Smyd3+/+ or Smyd3−/− mice injected intraperitoneally with PBS or VSV (1 × 107 PFU/per mouse) for 24 h. PBS, phosphate buffer saline; *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. Data are presented as the mean values of a representative experiment performed in triplicate (E–H) or as representative data (B); these experiments were repeated independently at least three times, and error bars indicate S.D.

Taken together, these data suggest that Smyd3 deficiency protects the host against RNA and DNA viral infection in vivo.

Smyd3-Deficient Cells Are More Resistant to Viral Infection.

We took full advantage of the Smyd3 knockout mice to perform further analyses. We established bone marrow–derived dendritic cells (BMDCs), bone marrow–derived macrophages (BMDMs), and mouse embryonic fibroblast (MEF) cells from Smyd3+/+ and Smyd3−/− mice and then examined cellular antiviral immune responses. Upon VSV infection, the expression of Ifnβ, Isg15, and Ifit1 mRNA was higher in Smyd3−/− BMDCs than in Smyd3+/+ BMDCs (Fig. 5A). Upon HSV-1 infection, the expression of Ifnβ, Cxcl10, and Ifnα mRNA was also higher in Smyd3−/− BMDCs than in Smyd3+/+ BMDCs (Fig. 5B). Similar results were obtained in BMDM cells (SI Appendix, Fig. S10). Consistently, the replication of VSV-GFP and HSV-1-GFP viruses was higher in Smyd3+/+ BMDCs than in Smyd3−/− BMDCs as revealed by fluorescence microscopy imaging (Fig. 5 C and D). Furthermore, VSV-induced phosphorylation and dimerization of Irf3 were enhanced in Smyd3−/− BMDCs compared to Smyd3+/+ BMDCs (Fig. 5E).

Fig. 5.

Fig. 5.

Open in a new tab

Loss of Smyd3 enhances the cellular antiviral immune response. (A) qPCR analysis of Ifnβ, Isg15, and Ifit1 mRNA in Smyd3+/+ and Smyd3−/− BMDCs UI or infected with VSV for 8 h. (B) qPCR analysis of Ifnβ, Cxcl10, and Ifnα mRNA in Smyd3+/+ and Smyd3−/− BMDCs UI or infected with HSV-1 for 8 h. (C) Microscopic imaging of VSV-GFP virus replication in Smyd3+/+ and Smyd3−/− BMDCs. (D) Microscopic imaging of HSV-1-GFP virus replication in Smyd3+/+ and Smyd3−/− BMDCs. (E) Phosphorylation and dimerization of Irf3 in Smyd3+/+ and Smyd3−/− BMDCs infected with VSV for 0, 4, and 8 h. (F) qPCR analysis of Ifnβ, Cxcl10, and Ccl5 mRNA in Smyd3+/+ and Smyd3−/− MEF cells transfected with (+) or without (−) poly I:C for 8 h. (G) qPCR analysis of Ifnβ, Cxcl10, and Ccl5 mRNA in Smyd3+/+ and Smyd3−/− MEF cells transfected with (+) or without (−) poly A:T for 8 h. (H) qPCR analysis of Ifnβ and Cxcl10 mRNA in Smyd3+/+ and Smyd3−/− MEF cells UI or infected with VSV for 8 h. (I) qPCR analysis of Ifnβ and Cxcl10 mRNA in Smyd3+/+ and Smyd3−/− MEF cells UI or infected with SeV for 8 h. (J) qPCR analysis of Ifnβ and Cxcl10 mRNA in Smyd3+/+ and Smyd3−/− MEF cells UI or infected with HSV-1 for 8 h. (K) Microscopic imaging (Left panels) and immunoblotting (Right panels) of VSV-GFP virus replication in Smyd3+/+ and Smyd3−/− MEF cells. (L) Phosphorylation and dimerization of Irf3 in Smyd3+/+ and Smyd3−/− MEF cells transfected with poly I:C for 0, 4, and 8 h. ***P < 0.001 and ****P < 0.0001. Graphs represent fold induction relative to the untreated cells. Data are presented as the mean values of a representative experiment performed in triplicate (A, B, and F–J) or as representative data (C–E, K, and L); these experiments were repeated independently at least three times, and error bars indicate S.D.

In MEF cells, poly I:C and poly A:T-induced Ifnβ, Cxcl10, and Ccl5 mRNA expression was higher in Smyd3 −/− MEF cells than in Smyd3+/+ MEF cells (Fig. 5 F and G). VSV, SeV, and HSV-1-induced Ifnβ and Cxcl10 mRNA expression was higher in Smyd3 −/− MEF cells than in Smyd3+/+ MEF cells (Fig. 5 H–J). Consistently, VSV-GFP virus replication was lower in Smyd3 −/− MEF cells than in Smyd3+/+ MEF cells as revealed by fluorescence microscopy imaging and immunoblotting with anti-GFP antibody (Fig. 5K). Poly I:C-induced phosphorylation and dimerization of Irf3 were enhanced in Smyd3−/− MEF cells compared to Smyd3+/+ MEF cells (Fig. 5L).

We then reconstituted WT Smyd3 in Smyd3−/− MEFs and examined antiviral immune responses. Reconstitution of Smyd3 in Smyd3−/− MEFs suppressed HSV-1-induced Ifnβ, Ifnα4, Ifit1, and Cxcl10 mRNA expression (SI Appendix, Fig. S11A). Consistently, the replication of VSV-GFP and HSV-1-GFP was increased in Smyd3-reconstituted Smyd3−/− MEFs as revealed by fluorescence microscopy imaging, immunoblotting, and flow cytometry analysis (SI Appendix, Fig. S11 B–E).

To further determine whether Smyd3 specifically modulates antiviral responses in response to viral infection, rather than having a broad transcriptional role, we performed genome-wide RNA sequencing (RNA-seq) analysis using Smyd3+/+ and Smyd3−/− MEFs infected without or with VSV. SMYD3 deficiency has no apparent effect on innate antiviral immunity in the absence of viral infection (SI Appendix, Fig. S12 A–C). However, VSV-induced antiviral genes were apparently up-regulated in Smyd3−/− MEFs compared to those in Smyd3+/+ MEF cells (SI Appendix, Fig. S12D). Furthermore, gene ontology enrichment analysis revealed that the enhanced genes resulting from Smyd3 loss were enriched in the category of innate immune response genes and virus response genes (SI Appendix, Fig. S12E). Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis revealed that these genes were enriched in the Toll-like receptor and RLR signaling pathways (SI Appendix, Fig. S12F). Thus, upon viral challenge, Smyd3 mainly affected the expression of antiviral response genes.

In addition, Smyd3 disruption induced Ifnβ, Cxcl10, and Ifit1 mRNA expression and inhibited virus replication in Irf3-intact MEF cells (Irf3+/+), but not in Irf3-deficient MEF cells (Irf3−/−) (SI Appendix, Fig. S13 A–F), further confirming that the suppressive role of Smyd3 on the innate antiviral response is dependent on Irf3.

Taken together, these data suggest that Smyd3-deficient BMDCs and MEFs are more resistant to viral infection than WT BMDCs and MEFs due to enhanced antiviral innate immunity in Smyd3-deficient BMDCs and MEFs.

Smyd3-Deficient Zebrafish Are More Resistant to Viral Infection.

To further elucidate the function of smyd3 in the type I IFN signaling pathway in vivo, we have used the zebrafish as a model organism. Smyd3 is evolutionarily conserved (SI Appendix, Fig. S14A). Overexpression of zebrafish smyd3 in EPC cells suppressed poly I:C and SVCV-induced ifn1, isg15, and viperin mRNA expression (SI Appendix, Fig. S14 B and C). Overexpression of smyd3 also suppressed irf3-activated the ISRE reporter activity (SI Appendix, Fig. S14D). Like mammalian Smyd3, ectopically expressed smyd3 interacted with ectopically expressed irf3 (SI Appendix, Fig. S14E). Zebrafish smyd3 had no effect on zebrafish irf3 protein stability (SI Appendix, Fig. S14F). Furthermore, zebrafish smyd3 was induced by SVCV infection in zebrafish larvae (SI Appendix, Fig. S14G).

We then knocked out smyd3 in zebrafish using CRISPR/Cas9 and obtained two mutant lines (SI Appendix, Fig. S14H). For mutant line 1 (-7 bp deletion), more smyd3+/+ zebrafish larvae were dead upon SVCV infection compared to smyd3−/− zebrafish larvae (SI Appendix, Fig. S15A). Consistently, the expression of ifn1 and lta mRNA was higher in smyd3−/− zebrafish larvae than in smyd3+/+ zebrafish larvae (SI Appendix, Fig. S15B). A similar survival curve was obtained in larvae of mutant line 2 (-19 bp deletion) upon SVCV infection (SI Appendix, Fig. S15C).

In adult zebrafish, SVCV injection also caused more severe pathological symptoms and a higher mortality rate in smyd3+/+ zebrafish than in smyd3−/− zebrafish (SI Appendix, Fig. S15 D and E). SVCV-induced ifn1 and lta mRNA expression in the spleen, liver, and kidney was higher in smyd3−/− zebrafish than in smyd3+/+ zebrafish (SI Appendix, Fig. S15 F–H). In contrast, the expression of G-, N-, and P-protein mRNA of SVCV was higher in smyd3+/+ zebrafish spleen and liver than in smyd3−/− zebrafish spleen and liver (SI Appendix, Fig. S15 I and J). Consistently, P and G protein of SVCV was higher in smyd3+/+ zebrafish liver and spleen than in smyd3−/− zebrafish liver and spleen as revealed by immunofluorescence staining with anti-P and anti-G antibodies (SI Appendix, Fig. S15 K and L).

To further evaluate the function of smyd3 in zebrafish, we used the SMYD3 inhibitor, BCI121. Similar to the effect on HEK293T cells, BCI121 did not affect ZFL apoptosis and viability when used in the range of 10 to 200 μM (SI Appendix, Fig. S16 A and B). BCI121 treatment increased poly I:C and SVCV-induced ifn1, Ita, or pkz mRNA expression in ZFL cells (SI Appendix, Fig. S16 C and D). For zebrafish larvae (3 dpf), treatment with BCI121 in the range of 10 to 200 μM did not affect the gross mortality (SI Appendix, Fig. S16E). Consistently, treatment with BCI121 significantly reduced the mortality of zebrafish larvae after challenge with SVCV (SI Appendix, Fig. S16 F and G). The expression of ifn1, ifn2, mxb, mxc, lta, pkz, and rig-i was promoted in zebrafish larvae treated with BCI121 (SI Appendix, Fig. S16H). Furthermore, ifnφ 1 promoter-driven mCherry expression was increased in Tg (infφ 1:mCherry) zebrafish larvae (8 dpf) after treatment with BCI121 (SI Appendix, Fig. S16I). Moreover, smyd3 deficiency or inhibition by BCI121 in zebrafish also resulted in enhanced resistance to infection of GCRV II (SI Appendix, Fig. S17 A–C), a virus causing severe disease in grass carp (37). In contrast, RFP-labeled macrophages and granulocytes and GFP-labeled neutrophils were not changed in Tg (lyz:DsRed2) or Tg(mpx:EGFP) zebrafish larvae after treatment with BCI121 (SI Appendix, Fig. S18 A and B). In addition, we knocked out irf3 in zebrafish using CRISPR/Cas9 (SI Appendix, Fig. S19A), and generated smyd3 and irf3 double knockout zebrafish (SI Appendix, Fig. S19B). In response to SVCV infection, no apparent difference was detected between smyd3+/+irf3−/− and smyd3−/−irf3−/− zebrafish (SI Appendix, Fig. S19C), indicating that the enhancement of antiviral ability by disruption of smyd3 was dependent on irf3.

Collectively, these data suggest that smyd3 deficiency protects zebrafish against SVCV or GCRV infection in vivo and that zebrafish smyd3 negatively regulates antiviral innate immunity depending on its enzymatic activity.

Notably, lysine 39 is also conserved between IRF3 and IRF7, as well as within IRF7 (lysine 58) (SI Appendix, Fig. S20 A and B). Although SMYD3 did not affect IRF7 protein stability, it did interact with IRF7 (SI Appendix, Fig. S20 C and D). Similar to the effect of SMYD3 on IRF3, SMYD3 also suppressed IRF7-activated ISRE reporter activity (SI Appendix, Fig. S20E). It appeared that the conserved lysine 58 of IRF7 is also the key target site of SMYD3 that modulates IRF7 activity (SI Appendix, Fig.S20 E and F). Therefore, SMYD3 may affect the type I interferon signaling pathway by targeting both IRF3 and IRF7. Further investigation of the effect of SMYD3 on IRF7 will provide a complete picture of the function of SMYD3 in antiviral innate immunity.

Discussion

As a family of lysine methyltransferases, SMYD proteins have been poorly characterized in antiviral innate immunity, although they have been shown to methylate nonhistone proteins and participate in host immunity against pathogen infection (20, 21, 33). In this study, we found that SMYD3, one of the SMYD family members, targets IRF3, possibly also IRF7, to directly catalyze the dimethylation of lysine residue in IRF3 and IRF7, thereby suppressing the type I IFN signaling. Thus, we reveal a specific function of SMYD3 and a specific modulator of the type I IFN signaling. In the resting state, SMYD3, which is constitutively associated with IRF3, may serve as a guard to prevent abnormal activation of IRF3. Upon viral infection, SMYD3 changes its role to a brake to limit overactivation of IRF3 and keep IRF3 activation at appropriate levels, as the association between SMYD3 and IRF3 is increased in this state. In addition, since SMYD3 is up-regulated during viral infection, the virus may escape the antiviral immune response by stimulating SMYD3 expression.

As a key factor in the type I IFN signaling, PTMs of IRF3 have been extensively studied, causing either enhancement or suppression of IRF3 activation (5, 9–11, 13, 16, 17, 38–43). Some of the PTMs even cross-react and influence each other to ultimately suppress or activate IRF3. However, it is still unclear whether SMYD3-mediated methylation can affect other modifications of IRF3 and how much SMYD3-mediated methylation contributes to the regulation of IRF3 activation. Because IRF3 activation requires to undergo a sequential process, including phosphorylation, dimerization, and subsequent nuclear translocation (5), PTMs usually affect IRF3 activation by affecting these steps. In this study, we found that SMYD3-mediated methylation of IRF3 attenuated IRF3 phosphorylation, dimerization, and nuclear translocation, thereby suppressing IRF3 activation. However, the methylated lysine of IRF3 by SMYD3 is located at its N terminus (lysine 39), which is located in the N-terminal DNA-binding domain of IRF3, whereas the major phosphorylation sites of IRF3 are located at the C terminus (6, 44–46). How IRF3 K39 methylation affects IRF3 phosphorylation to repress its transcriptional activity remains a mystery. Notably, the crystal structure of the DNA-binding domain of IRF3 bound to the enhancer of the IFN-β gene indicates that a hydrogen bond is formed between the IFN-β enhancer site and the main carbonyl from Arg7 to Leu42 of the upstream of IRF3 (47). Lys39 is localized in the region from Arg7 to Leu 42 and is closer to Leu42. Thus, the methylation of IRF3 at Lys39 may also interfere with the binding of IRF3 to its target DNA, resulting in the attenuation of target gene expression. Further investigation of the fine process of SMYD3-mediated IRF3 methylation in affecting IRF3 function will provide insights into the molecular mechanisms of SMYD3 in antiviral innate immunity.

Notably, SMYD3 not only catalyzes dimethylation at lysine residues of its targets but also catalyzes trimethylation of its targets (24). Indeed, in this study, we identified both trimethylation and dimethylation at lysine 39 of IRF3 by mass spectrometry analysis. Due to the unsuccessful development of the anti-IRF3-K39Me3 antibody, we were unable to provide further evidence that SMYD3 also catalyzes trimethylation of IRF3 lysine 39. Therefore, we cannot exclude the possibility that SMYD3-catalyzed trimethylation of IRF3 lysine 39 also contributes to the regulation of IRF3 activity.

In addition, we show that the suppressive role of SMYD3 on antiviral innate immunity is dependent on its enzymatic activity and the inhibitor BCI121 can promote cellular and host antiviral ability, suggesting that the specific inhibitor of SMYD3 with small molecule can be developed as a therapeutic drug to treat RNA and DNA virus infection.

It is well known that IRF3 is activated in response to viral infection. Here, we show that IRF3 K39 is constitutively dimethylated in UI and early infected cells, how does viral infection trigger the de-dimethylation of IRF3 to release the inhibition of IRF3 by SMYD3? Our further assays showed that KDM8 is responsible for the de-dimethylation of IRF3 at K39 upon viral infection, which may explain this confusion. Of course, we still cannot rule out other reasons responsible for this mechanism.

Materials and Methods

The following is a brief description of Materials and Methods, and the description of Materials and methods is detailed in SI Appendix.

Mice.

6 to 8-wk-old mice of the same sex were randomly assigned to the experiments.

Zebrafish.

The smyd3ihbsm3/ihbsm3 and the smyd3ihbsm3b/ihbsm3b mutant zebrafish were used in the experiments.

CRISPR-Cas9 Knockout Cell Lines.

Double-stranded oligonucleotides corresponding to the target sequences were cloned into the LentiCRISPRv2 plasmid and then cotransfected with viral packaging plasmids into HEK293T cells. Infected cells were selected with puromycin (1 μg/mL) for 2 wk.

Lentivirus-Mediated Gene Transfer.

Cells were transfected with pHAGE-SMYD3, pHAGE-SMYD3-F183A, or pHAGE EV together with the packaging vectors pSPAX2 and pMD2G.

Flow Cytometry Analysis for VSV-GFP- or HSV-1-GFP-Infected Cells.

Cells were infected with VSV-GFP or HSV-1-GFP viruses for the indicated time. Cells were then counted and analyzed using Beckman CytoFLEXS.

Immunofluorescence Confocal Microscopy.

Cells were stained with the indicated primary antibodies followed by incubation with fluorescence dye-conjugated secondary antibodies.

Proximity Ligation Assay.

Cells were incubated with primary antibodies for 1 h at room temperature and nuclei were counterstained with DAPI. Red fluorescent spots were detected using a Leica laser scanning confocal microscope.

Identification of IRF3 Methylated Lysine Site (s) by Mass Spectrometry.

HEK293T cells were cotransfected with Flag-IRF3 and Myc-SMYD3. The purified proteins were then digested with trypsin and analyzed by online nanoflow LC–MS/MS.

Generation of Site-Specific Methylation Antibody.

An IRF3 K39 site-specific dimethylation antibody (anti-IRF3-K39Me2) was generated by using a human IRF3 dimethylated peptide [RIPW-(K-me2)-HGLR-C] as an antigen.

In Vitro Methylation Assays.

In a 50 μL reaction volume, purified GST-IRF3 (5 μg) together with GST-SMYD3 (1 μg) or GST (1 μg) as control was mixed with reaction buffer (20 mM Tris–HCl, pH 8.0, 200 mM NaCl, and 0.4 mM EDTA) containing 100 mM unlabeled S-(50-adenosyl)-L-methionine chloride dihydrochloride (SAM) for 1 h at 37 °C.

Identification of IRF3-Associated Proteins by Mass Spectrometry.

HEK293T cells were cotransfected with Flag-IRF3 for 24 h, followed by poly I:C stimulation for 8 h. The specific bands were then digested with trypsin and analyzed by online nanoflow LC-MS/MS.

RNA Sequencing.

Whole RNA from MEF cells infected with or without VSV was purified using the RNeasy Mini Kit. The transcriptome library for sequencing was prepared using the NEBNext Ultra RNA Library Prep Kit for Illumina.

Viral Infection in Cells and Mice.

Cells were infected with SeV, VSV, VSV-GFP, HSV-1, or HSV-GFP for the indicated time. Age- and sex-matched male mice were infected with VSV (1 × 107 PFU/ per mouse) or HSV-1 (5 × 107 PFU/ per mouse) by intraperitoneal injection.

Viral Infection of Zebrafish.

Thirty larvae (3–8 dpf) were pooled in a 60 mm disposable cell culture dish filled with 4 mL egg water and 1 mL SVCV (∼6 × 107 TCID50 per milliliter). Three-month-old zebrafish were injected intraperitoneally per individual with 10 μL SVCV (∼6 × 107 TCID50 per milliliter) or cell culture medium as a control.

Statistical Analysis.

Statistical analyses were performed with GraphPad Prism 8.0.

Supplementary Material

Appendix 01 (PDF)

pnas.2320644122.sapp.pdf (5.3MB, pdf)

Acknowledgments

We thank Cyagen Biosciences Inc. (Guangzhou, China) for assistance with generation of Smyd3-deficient mice. We thank Drs. Hong-Bing Shu, Bo Zhong, Mingzhou Chen, Chunfu Zheng, and Hong Tang for providing reagents. We thank Yan Wang and Fang Zhou at the Core Facility of Institute of Hydrobiology for FACS and confocal microscope. This work was supported by grants from the Strategic Priority Research Program of the Chinese Academy of Sciences (XDB0730300 and XDA24010308), the National Natural Science Foundation of China (32430108), the National Key Research and Development Program of China (2022YFF1000302), the Natural Science Foundation of Hubei Province of China (2022CFA110), and “Agricultural Biological Breeding-2030” major project (2023ZD04065).

Author contributions

Z.W., X.L., and W.X. designed research; Z.W., X.C., C.Z., J.T., and H.D. performed research; C.Z., S.F., J.T., H.D., and X.S. contributed new reagents/analytic tools; Z.W., X.C., X.L., and W.X. analyzed data; and Z.W., X.L., and W.X. wrote the paper.

Competing interests

The authors declare no competing interest.

Footnotes

This article is a PNAS Direct Submission. P.F. is a guest editor invited by the Editorial Board.

Contributor Information

Xing Liu, Email: liuxing@ihb.ac.cn.

Wuhan Xiao, Email: w-xiao@ihb.ac.cn.

Data, Materials, and Software Availability

All study data are included in the article and/or SI Appendix and research datasets (48, 49). Raw mass spectrometry data were deposited with the ProteomeXchange Consortium via the iProX partner repository as indicated in reference (48). The original RNA-seq data were uploaded to the GEO datasets as indicated in reference (49).

Supporting Information

References

  • 1.Chen I. Y., Ichinohe T., Response of host inflammasomes to viral infection. Trends Microbiol. 23, 55–63 (2015). [DOI] [PubMed] [Google Scholar]
  • 2.Tan X., Sun L., Chen J., Chen Z. J., Detection of microbial infections through innate immune sensing of nucleic acids. Annu. Rev. Microbiol. 72, 447–478 (2018). [DOI] [PubMed] [Google Scholar]
  • 3.Takeuchi O., Akira S., Pattern recognition receptors and inflammation. Cell 140, 805–820 (2010). [DOI] [PubMed] [Google Scholar]
  • 4.Liu X., et al. , Oxygen enhances antiviral innate immunity through maintenance of EGLN1-catalyzed proline hydroxylation of IRF3. Nat. Commun. 15, 3533 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Petro T. M., IFN regulatory factor 3 in health and disease. J. Immunol. 205, 1981–1989 (2020). [DOI] [PubMed] [Google Scholar]
  • 6.Fitzgerald K. A., et al. , IKKepsilon and TBK1 are essential components of the IRF3 signaling pathway. Nat. Immunol. 4, 491–496 (2003). [DOI] [PubMed] [Google Scholar]
  • 7.Taniguchi T., Ogasawara K., Takaoka A., Tanaka N., IRF family of transcription factors as regulators of host defense. Annu. Rev. Immunol. 19, 623–655 (2001). [DOI] [PubMed] [Google Scholar]
  • 8.Lin R., Heylbroeck C., Pitha P. M., Hiscott J., Virus-dependent phosphorylation of the IRF-3 transcription factor regulates nuclear translocation, transactivation potential, and proteasome-mediated degradation. Mol. Cell Biol. 18, 2986–2996 (1998). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Huai W., et al. , KAT8 selectively inhibits antiviral immunity by acetylating IRF3. J. Exp. Med. 216, 772–785 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Long L., et al. , Recruitment of phosphatase PP2A by RACK1 adaptor protein deactivates transcription factor IRF3 and limits type I interferon signaling. Immunity 40, 515–529 (2014). [DOI] [PubMed] [Google Scholar]
  • 11.Wang C. M., et al. , The methyltransferase NSD3 promotes antiviral innate immunity via direct lysine methylation of IRF3. J. Exp. Med. 214, 3597–3610 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Li S., et al. , The tumor suppressor PTEN has a critical role in antiviral innate immunity. Nat. Immunol. 17, 241–249 (2016). [DOI] [PubMed] [Google Scholar]
  • 13.Zhang Z. M., Wang D. D., Wang P. Y., Zhao Y. C., You F. P., OTUD1 negatively regulates type I IFN induction by disrupting noncanonical ubiquitination of IRF3. J. Immunol. 204, 1904–1918 (2020). [DOI] [PubMed] [Google Scholar]
  • 14.Maarifi G., et al. , MxA mediates SUMO-induced resistance to vesicular stomatitis virus. J. Virol. 90, 6598–6610 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Suhara W., Yoneyama A., Kitabayashi I., Fujita T., Direct involvement of CREB-binding protein/p300 in sequence-specific DNA binding of virus-activated interferon regulatory factor-3 holocomplex. J. Biol. Chem. 277, 22304–22313 (2002). [DOI] [PubMed] [Google Scholar]
  • 16.Li Z., et al. , Zebrafish F-box protein fbxo3 negatively regulates antiviral response through promoting K27-linked polyubiquitination of the transcription factors irf3 and irf7. J. Immunol. 205, 1897 (2020). [DOI] [PubMed] [Google Scholar]
  • 17.Chattopadhyay S., Kuzmanovic T., Zhang Y., Wetzel J. L., Sen G. C., Ubiquitination of the transcription factor IRF-3 activates RIPA, the apoptotic pathway that protects mice from viral pathogenesis. Immunity 44, 1151–1161 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Wang J., et al. , Arginine methylation by PRMT2 promotes IFN-beta production through TLR4/IRF3 signaling pathway. Mol. Immunol. 139, 202–210 (2021). [DOI] [PubMed] [Google Scholar]
  • 19.Wang C., et al. , The methyltransferase NSD3 promotes antiviral innate immunity via direct lysine methylation of IRF3. J. Exp. Med. 214, 3597–3610 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Rueda-Robles A., Audano M., Alvarez-Mercado A. I., Rubio-Tomas T., Functions of SMYD proteins in biological processes: What do we know? An updated review. Arch Biochem. Biophys. 712, 109040 (2021). [DOI] [PubMed] [Google Scholar]
  • 21.Rubio-Tomas T., The SMYD family proteins in immunology: An update of their obvious and non-obvious relations with the immune system. Heliyon 7, e07387 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Padilla A., et al. , Targeting epigenetic changes mediated by members of the SMYD family of lysine methyltransferases. Molecules 28, 2000 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Tracy C., et al. , The Smyd family of methyltransferases: Role in cardiac and skeletal muscle physiology and pathology. Curr. Opin. Physiol. 1, 140–152 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Hamamoto R., et al. , SMYD3 encodes a histone methyltransferase involved in the proliferation of cancer cells. Nat. Cell Biol. 6, 731–740 (2004). [DOI] [PubMed] [Google Scholar]
  • 25.Van Aller G. S., et al. , Smyd3 regulates cancer cell phenotypes and catalyzes histone H4 lysine 5 methylation. Epigenetics 7, 340–343 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Kunizaki M., et al. , The lysine 831 of vascular endothelial growth factor receptor 1 is a novel target of methylation by SMYD3. Cancer Res. 67, 10759–10765 (2007). [DOI] [PubMed] [Google Scholar]
  • 27.Mazur P. K., et al. , SMYD3 links lysine methylation of MAP3K2 to Ras-driven cancer. Nature 510, 283–287 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Yoshioka Y., et al. , SMYD3-mediated lysine methylation in the PH domain is critical for activation of AKT1. Oncotarget 7, 75023–75037 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Kim H., et al. , Requirement of histone methyltransferase SMYD3 for estrogen receptor-mediated transcription. J. Biol. Chem. 284, 19867–19877 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Yoshioka Y., et al. , Protein lysine methyltransferase SMYD3 is involved in tumorigenesis through regulation of HER2 homodimerization. Cancer Med. 6, 1665–1672 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Bottino C., Peserico A., Simone C., Caretti G., SMYD3: An oncogenic driver targeting epigenetic regulation and signaling pathways. Cancers (Basel) 12, 142 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Yamamoto K., et al. , SMYD3 interacts with HTLV-1 Tax and regulates subcellular localization of Tax. Cancer Sci. 102, 260–266 (2011). [DOI] [PubMed] [Google Scholar]
  • 33.Chen J., et al. , Host factor SMYD3 is recruited by Ebola virus nucleoprotein to facilitate viral mRNA transcription. Emerg. Microbes Infect. 8, 1347–1360 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Nagata D. E., et al. , Epigenetic control of Foxp3 by SMYD3 H3K4 histone methyltransferase controls iTreg development and regulates pathogenic T-cell responses during pulmonary viral infection. Mucosal Immunol. 8, 1131–1143 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Wu J., et al. , Lysine methyltransferase SMYD2 inhibits antiviral innate immunity by promoting IRF3 dephosphorylation. Cell Death Dis. 14, 592 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Peserico A., et al. , A SMYD3 small-molecule inhibitor impairing cancer cell growth. J. Cell Physiol. 230, 2447–2460 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Rao Y., Su J., Insights into the antiviral immunity against grass carp (Ctenopharyngodon idella) reovirus (GCRV) in grass carp. J. Immunol. Res. 2015, 670437 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Wang Z., et al. , The EF-hand protein CALML6 suppresses antiviral innate immunity by impairing IRF3 dimerization. Cell Rep. 26, 1273–1285.e1275 (2019). [DOI] [PubMed] [Google Scholar]
  • 39.Sato M., et al. , Distinct and essential roles of transcription factors IRF-3 and IRF-7 in response to viruses for IFN-alpha/beta gene induction. Immunity 13, 539–548 (2000). [DOI] [PubMed] [Google Scholar]
  • 40.Cai Z., et al. , USP22 promotes IRF3 nuclear translocation and antiviral responses by deubiquitinating the importin protein KPNA2. J. Exp. Med. 217, e20191174 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Xiao J., et al. , Targeting 7-dehydrocholesterol reductase integrates cholesterol metabolism and IRF3 activation to eliminate infection. Immunity 52, 109–122.e106 (2020). [DOI] [PubMed] [Google Scholar]
  • 42.Yang Q., et al. , Host HDAC4 regulates the antiviral response by inhibiting the phosphorylation of IRF3. J. Mol. Cell Biol. 11, 158–169 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Zhou Y., et al. , Interferon-inducible cytoplasmic lncLrrc55-AS promotes antiviral innate responses by strengthening IRF3 phosphorylation. Cell Res. 29, 641–654 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Yoneyama M., Suhara W., Fujita T., Control of IRF-3 activation by phosphorylation. J. Interferon Cytokine Res. 22, 73–76 (2002). [DOI] [PubMed] [Google Scholar]
  • 45.Mori M., et al. , Identification of Ser-386 of interferon regulatory factor 3 as critical target for inducible phosphorylation that determines activation. J. Biol. Chem. 279, 9698–9702 (2004). [DOI] [PubMed] [Google Scholar]
  • 46.Qin B. Y., et al. , Crystal structure of IRF-3 in complex with CBP. Structure 13, 1269–1277 (2005). [DOI] [PubMed] [Google Scholar]
  • 47.Panne D., Maniatis T., Harrison S. C., An atomic model of the interferon-beta enhanceosome. Cell 129, 1111–1123 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Wang Z. X., et al. , Data from “Direct lysine dimethylation of IRF3 by the methyltransferase SMYD3 attenuates antiviral innate immunity.” iProX partner repository. https://www.iprox.cn/page/project.html?id=IPX0010737000. Deposited 2 Jan 2025. [DOI] [PMC free article] [PubMed]
  • 49.Wang Z. X., et al. , Data from “Direct lysine dimethylation of IRF3 by the methyltransferase SMYD3 attenuates antiviral innate immunity.” The GEO Datasets. https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE285658. Deposited 1 Jan 2025. [DOI] [PMC free article] [PubMed]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Appendix 01 (PDF)

pnas.2320644122.sapp.pdf (5.3MB, pdf)

Data Availability Statement

All study data are included in the article and/or SI Appendix and research datasets (48, 49). Raw mass spectrometry data were deposited with the ProteomeXchange Consortium via the iProX partner repository as indicated in reference (48). The original RNA-seq data were uploaded to the GEO datasets as indicated in reference (49).

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES