A systematic screen for regulators of the type I Interferon signaling pathway

To identify host factors that modulate the type I IFN signaling pathway, we conducted a cDNA library screen using HEK293A cells (Fig 1A). Cells were transfected with individual plasmids encoding 15,150 human genes. Twenty-four hours post-transfection, cells were stimulated with IFNβ for 12 hours to activate the IFN pathway, followed by infection with Newcastle disease virus (NDV) expressing green fluorescence protein (GFP) at an MOI of 1 for an additional 24 hours. NDV-GFP replication was calculated based on the proportion of GFP-positive cells via high-content screening, with the aim of identifying genes whose overexpression counteracts the antiviral effect of IFN. Using a Z-score threshold greater than 3, we identified 257 genes that counteracted IFNβ-mediated inhibition of NDV-GFP infection (Fig 1B and S1 Table). As expected, overexpression of IFNAR2, a positive regulator of IFN signaling, markedly reduced NDV-GFP infection, whereas SOCS1, a known negative regulator of IFN pathways, enhanced viral replication (Fig 1C).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Genome-wide cDNA screening for antagonists of the type I interferon downstream signal pathway.

(A) Schematic of the screening workflow. Plasmids containing the cDNA library (15,150 plasmids) were transfected individually into HEK293A cells. After 24 hours, IFNβ (400 U/mL) was added. Following 12 hours of stimulation, cells were infected with NDV-GFP at an MOI of 1. After an additional 16 hours, cells were fixed using 4% formaldehyde, stained with DAPI, and images were captured using the Operetta system to measure the NDV-GFP infection rate. (B) Primary screening results for each cDNA plasmid. Z-scores were calculated as (infection rate − average infection rate)/standard deviation of the infection rate. SOCS1, USP18 and ZC3H10 were highlighted in red. (C) Representative infection outcomes from the primary screening. Infectivity was indicated by GFP expression, and cell nuclei were stained with DAPI (blue). Scale bar, 100 μm. (D) Validation of candidate genes. The 50 top candidate genes from the primary screening were individually co-transfected with the IFIT1 promoter luciferase reporter. After a 24-hour incubation, IFNβ (400 U/mL) was added to the supernatant for stimulation. Twelve hours later, dual luciferase reporter gene expression was measured. IFIT1 promoter activity is shown as firefly fluorescence intensity normalized to Renilla fluorescence intensity. All data are presented as mean values ± SD. *p < 0.05, **p < 0.01, ***p < 0.001. The data underlying this Figure can be found in S1 Table, S1 Data, and S1 Raw Images.

https://doi.org/10.1371/journal.pbio.3003881.g001

To validate these findings, we performed a secondary screen using an IFIT1-promoter luciferase assay. From the 257 candidates, we selected the top 50 genes for further evaluation using the IFIT1-promoter luciferase reporter system. Each candidate plasmid was co‑transfected with the pFL‑IFIT1‑promoter‑Luc reporter and pRL‑SV40 (internal control). After 24 h, cells were treated with IFNβ for 12 h, after which luciferase activity was measured. The results showed that most of the top 50 genes inhibited IFIT1-promoter activation, and ZC3H10 demonstrated the most potent suppression of IFN-induced transcription, outperforming even SOCS1 (Fig 1D). Therefore, we selected ZC3H10 for further investigation.

ZC3H10 overexpression inhibits the expression of ISGs

To further validate the functional role of ZC3H10 in IFN-mediated antiviral activity, we employed immunofluorescence and TCID₅₀ analyses to assess the effect of ZC3H10 on NDV-GFP replication in the presence of IFN. These assays showed that ZC3H10 overexpression significantly antagonized IFNβ antiviral activity, enhancing NDV-GFP infection (Figs 2A and S1A). Given that IFNβ exerts its antiviral effects by inducing ISGs, we next examined the impact of ZC3H10 on ISG promoter activation. Luciferase reporter assays revealed that ZC3H10 suppresses both IFIT1 promoter- and ISRE-driven luciferase expression in a dose-dependent manner (Fig 2B and 2C). Consistent with these results, ZC3H10 also decreased the IFN-induced endogenous mRNA levels of several ISGs, including ISG15, IFIT1, IFIT2, OAS1, and OAS2 (Fig 2D). We further assessed IFIT2 protein expression levels by western blot, which showed that IFNβ induced IFIT2 expression in a time-dependent manner, whereas ZC3H10 overexpression decreased IFIT2 protein levels (Fig 2E). Importantly, this phenomenon was recapitulated in RAW264.7 macrophages, a well-established model for studying innate immune responses. In these cells, ZC3H10 downregulated mRNA levels of ISGs (ISG15, IFIT1/2, OAS1/2) in both the absence and presence of IFNβ stimulation (Fig 2F).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. ZC3H10 inhibits ISG expression and promotes viral replication.

(A) HEK293A cells were transfected with either empty or ZC3H10 plasmids. After 24 hours, cells were stimulated with IFNβ (400 U/mL). Twelve hours later, cells were infected with NDV-GFP. After an additional 16 hours, cells were fixed with 4% formaldehyde and DAPI was used to stain the nuclei. NDV-GFP is shown in green and nuclei in blue. Simultaneously, viral titers were measured by TCID₅₀. Scale bar, 100 μm. (B, C) IFIT1 promoter luciferase (B) and ISRE luciferase (C) reporter plasmids were co-transfected into HEK293A cells with 0, 50, 100, 200, and 400 ng of ZC3H10 plasmids into HEK293A cells. After 24 hours, cells were either stimulated or not with IFNβ (400 U/mL). Twelve hours later, dual luciferase activity was measured, and ZC3H10 expression was assessed by western blot using anti-flag antibody. (D) HEK293A cells were transfected with 500 ng of empty plasmid (negative control), SOCS1 (positive control), and ZC3H10 plasmid. After 24 hours, IFNβ (400 U/mL) was added. At 0, 4, 8, 12, and 24 hours post-stimulation, mRNA expression of ISG15, IFIT1, IFIT2, OAS1, and OAS2 was measured by qRT-PCR. (E) HEK293A cells were transfected with 500 ng of empty plasmid (negative control) and ZC3H10 plasmid. After 24 hours, cells were stimulated with IFNβ (400 U/mL). At 0, 4, 8, 12, and 24 hours post-stimulation, protein expression of IFIT2 was detected. ZC3H10 expression was detected by using anti-Flag antibody. (F) RAW264.7 cells were transfected with 500 ng of empty plasmid (negative control) and ZC3H10 plasmid. After 24 hours, cells were either stimulated or not with IFNβ (100 U/mL). At 4 hours post-stimulation, mRNA expression of ISG15, IFIT1, IFIT2, OAS1, and OAS2 was detected by qRT-PCR, and protein expression of IFIT2 and ISG15 was assessed by western blot. ZC3H10 expression was detected by using anti-flag antibody. (G–H) HEK293A cells were transfected with 500 ng of empty plasmid or ZC3H10 plasmid. After 24 hours, IFNβ (400 U/mL) was added. Twelve hours post-stimulation, cells were infected with EV71 or HRV16 at an MOI of 0.5. Viral replication was assessed after another 16 hours by western blot (G), and viral titers were measured by TCID₅₀ assay (H). All experiments were repeated at least twice, with one representative result shown here. All data are presented as mean values ± SD. *p < 0.05, **p < 0.01, ***p < 0.001. The data underlying this Figure can be found in S1 Data and S1 Raw Images.

https://doi.org/10.1371/journal.pbio.3003881.g002

To further demonstrate the specificity of ZC3H10-mediated repression, we detected its effect on non-ISG genes, such as phosphoglucomutase 2 (PGM2) and euchromatic histone lysine methyltransferase 2 (EHMT2). The data showed that ZC3H10 overexpression did not alter the transcript levels of these two genes (S1B Fig). Another important remaining question was whether ZC3H10 modulates the expression of interferon regulatory factor (IRF)- and nuclear factor kappa B (NF-κB)-driven genes, which represent upstream transcriptional regulators of the antiviral response. Our findings demonstrated that ZC3H10 did not affect these key immune signaling axes (S1C and S1D Fig), indicating that its inhibitory function is specific to ISG transcription. Collectively, these results demonstrate that ZC3H10 specifically suppresses ISGs transcription.

To further validate the biological significance of these findings, we examined the effect of ZC3H10 on the IFN-sensitive viruses enterovirus 71 (EV71) and rhinovirus 16 (HRV16). Remarkably, ZC3H10 overexpression significantly enhanced the infectivity of both viruses, as measured by viral protein levels and viral titers (Fig 2G and 2H). Collectively, our results demonstrate that ZC3H10 functions as a potent negative regulator of the antiviral response by suppressing ISG expression, thereby supporting viral replication.

Type I IFN signaling is upregulated in ZC3H10 knockout cells

To further investigate the functional role of ZC3H10 in IFN responses, we generated a ZC3H10 knockout (KO) cell line using CRISPR-Cas9 technology. Successful disruption of the ZC3H10 locus was confirmed by genome sequencing, and the absence of ZC3H10 protein expression was verified by western blot (Fig 3A). Having established these KO cells, we next examined whether ZC3H10 depletion affects the expression of ISGs. Quantitative real-time PCR (qRT-PCR) analysis revealed that ZC3H10 KO significantly upregulated the IFNβ-mediated expression of ISGs, including ISG15, IFIT1, IFIT2, OAS1, and OAS2 (Fig 3B). Notably, even in the absence of IFNβ stimulation, the basal expression levels of ISG15, IFIT1, IFIT2, OAS2 were also elevated in ZC3H10 knockout cells, suggesting that ZC3H10 exerts a repressive effect on ISG transcription under steady-state conditions. Consistent with the mRNA data, western blot analysis demonstrated that IFIT2 expression was markedly increased in ZC3H10 knockout cells relative to wild-type (WT) controls (Fig 3C).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 3. Enhanced ISG expression and reduced viral replication in ZC3H10 knockout cells.

(A) HEK293A ZC3H10-knockout cells were generated by CRISPR-Cas9 genome editing. The left panel displays the sequences of the targeted region. The right panel shows ZC3H10 expression. (B) WT and KO cells were stimulated with or without IFNβ (400 U/mL). mRNA expression of ISG15, IFIT1, IFIT2, OAS1 and OAS2 was detected 12 hours post-stimulation by RT-PCR. (C) IFIT2 protein expression was detected by western blot at 0, 4, 8, 12, and 24 hours post-stimulation. ZC3H10 expression was detected by using anti-ZC3H10 antibody. (D) Volcano plot of differentially expressed genes (DEGs) comparing ZC3H10 knockout and wild-type cells. Each group consisted of three biological replicates. Red indicates upregulated DEGs, blue indicated downregulated DEGs. (E) Gene Ontology (GO) analysis of upregulated DEGs. (F) Heatmap of differentially expressed interferon-stimulated genes (DE-ISGs). (G) RAW264.7 cells were transfected with scramble siRNA and ZC3H10-siRNA. After 48 hours, cells were stimulated or not with IFNβ (100 U/mL). At 4 hours post-stimulation, mRNA expression of ISG15, IFIT1, IFIT2, OAS1, and OAS2 was detected by RT-PCR, and protein expression of IFIT2 and ISG15 was assessed by western blot. ZC3H10 was detected using anti-ZC3H10 antibody. (H) After stimulation with IFNβ (400 U/mL) for 12 hours, WT and KO cells were infected with EV71 (MOI = 0.5), HRV16 (MOI = 0.5), NDV-GFP (MOI = 1). Viral replication was assessed by western blot analysis after an additional 16 hours. (I) Cells seeded in 24-well plates were pre-treated with or without IFNβ (400 U/mL) for 12 hours, followed by infection with NDV-GFP. At 16 hours post-infection, cells were fixed with 4% paraformaldehyde and stained with DAPI. Images were then captured using an Operetta high-content imaging system. The percentage of infected cells were quantified by computing the ratio of virus protein-positive cells relative to the total number of nuclei (DAPI-stained). For SARS-CoV-2, cells were first transduced with a lentivirus expressing human ACE2. Cells were then stained using an antibody against the viral nucleocapsid (N) protein, followed by a 488-conjugated secondary antibody, and the percentage of N-positive cells were calculated using the same method. (J) Viral titers of EV71, HRV16, NDV, and SARS-CoV-2 in WT and ZC3H10 knockout (KO) cells were measured using a TCID₅₀ assay. All data are presented as mean values ± SD. *p < 0.05, **p < 0.01, ***p < 0.001. The data underlying this Figure can be found in S2–S3 Tables, S1 Data, and S1 Raw Images.

https://doi.org/10.1371/journal.pbio.3003881.g003

To comprehensively investigate the transcriptional influence of ZC3H10, we performed RNA sequencing (RNA-Seq) comparing WT and ZC3H10 KO cells under basal conditions (without IFNβ stimulation). Differential expression analysis identified 259 upregulated and 241 downregulated genes in ZC3H10 KO cells compared to WT controls (Fig 3D and S2 Table). Gene Ontology (GO) analysis revealed that the upregulated genes were strikingly enriched in pathways associated with “type I interferon signaling” and “interferon-alpha response” (Fig 3E), indicating that ZC3H10 primarily represses genes involved in antiviral immunity. A heatmap depicting 18 differentially expressed ISGs further confirmed increased expression of IFITM family, GBP family, IFIT family, and ISG15 in ZC3H10 KO cells (Fig 3F), consistent with our qRT-PCR findings. To validate these observations in a physiologically relevant immune cell context, we performed siRNA-mediated knockdown of ZC3H10 in RAW264.7 macrophages. Similarly to the KO cell phenotype observed in HEK293A cells, ZC3H10 knockdown in RAW264.7 significantly enhanced ISG expression (Fig 3G), confirming that the repressive function of ZC3H10 is conserved across cell types.

Having established that ZC3H10 deficiency enhanced ISG expression, we next assessed the functional consequence for viral replication. Functional assays demonstrated that ZC3H10 KO cells exhibited markedly reduced replication of several IFN-sensitive viruses, as measured by viral protein expression levels, including EV71, HRV16, and NDV-GFP (Fig 3H). Furthermore, infections with NDV-GFP and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) were substantially impaired in ZC3H10-deficient cells, and this effect was observed irrespective of the presence or absence of exogenous IFNβ (Fig 3I). Consistent results were obtained from viral titer assays (Fig 3J). Collectively, these results establish ZC3H10 as a critical negative regulator of the type I IFN response that promotes viral replication through broad suppression of ISG expression.

The nuclear localization signal and zinc finger domains are crucial for ZC3H10-mediated ISG downregulation

ZC3H10 is a CCCH-type zinc finger protein known to bind RNA and contribute to microRNA maturation [38], as well as to regulate transcription by interacting with the distal promoter of UCP-1 [39]. Structurally, ZC3H10 contains three zinc finger domains (ZF1–ZF3), a nuclear localization signal (NLS), a coiled-coil region, an N-terminal glycine-rich region, and a C-terminal proline-rich region (Fig 4A). To identify the domains required for ZC3H10-mediated suppression of type I IFN signaling, we generated a series of truncation mutants and assessed their effects on NDV-GFP infection (Fig 4A). Overexpression of wild-type ZC3H10 significantly enhanced viral replication as expected. However, deletion of the NLS (ΔNLS, 205–213 aa), any of the three ZF domains (ΔZF1: 35–63 aa; ΔZF2: 73–99 aa; ΔZF3: 134–161 aa), or the coiled-coil domain (Δcoiled-coil, 234–280) abolished this enhancing effect, indicating that nuclear localization, intact zinc finger motifs and coiled-coil domain are critical for ZC3H10 function (Fig 4B). Subcellular localization studies confirmed that only the ΔNLS mutant was retained exclusively in the cytoplasm, whereas wild-type ZC3H10 and all other mutants localized to the nucleus (Fig 4C). These findings suggest that the NLS is necessary for nuclear import, while the zinc finger and coiled-coil domains, despite not affecting subcellular distribution, are indispensable for downstream functional activity.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 4. The nuclear localization signal (NLS) and zinc finger (ZF) domains of ZC3H10 are crucial for its function.

(A) Schematic representation of ZC3H10 domains and its truncated variants. (B) HEK293A cells were transfected with either empty or plasmids encoding ZC3H10 variants. After 24 hours, cells were stimulated with IFNβ (400 U/mL). After an additional 12 hours, cells were infected with NDV-GFP at an MOI of 1. Sixteen hours later, viral titers were determined by TCID₅₀ assay. (C) HEK293A cells were transfected with plasmids encoding wild-type ZC3H10 or its truncated variants. After 24 hours, cells were fixed and stained with Flag antibodies. Green represents the variants and blue indicates nuclei. Scale bar, 5 μm. (D, E) HEK293A cells were transfected with either empty or plasmids encoding ZC3H10 variants. After 24 hours, cells were stimulated with IFNβ (400 U/mL). After an additional 12 hours, mRNA expression of ISG15, IFIT1, IFIT2, OAS1, and OAS2 (D) were measured by RT-PCR, as well as protein expression of STAT1 and STAT2 (E) were detected by western blot analysis. All experiments were repeated at least twice, with one representative result shown here. All data are presented as mean values ± SD. *p < 0.05, **p < 0.01, ***p < 0.001. The data underlying this Figure can be found in S1 Data and S1 Raw Images.

https://doi.org/10.1371/journal.pbio.3003881.g004

Consistent with the viral replication data in Fig 4B, mutants lacking the NLS, any ZF domain, or coiled-coil domain lost the ability to suppress transcription of ISGs as assessed by qRT-PCR (Fig 4D). Furthermore, in contrast to wild-type ZC3H10, which suppressed IFNβ-induced expression of STAT1 and STAT2, the ΔNLS, ZF, and coiled-coil domain deletion mutants failed to reduce STAT1 and STAT2 protein levels (Fig 4E). Collectively, these results demonstrate that the NLS, the three ZF domains, and coiled-coil domain are essential for ZC3H10 to inhibit the IFN signaling pathway and promote viral replication.

To further validate the role of ZC3H10, we performed reconstitution experiments. Reintroduction of wild-type ZC3H10 suppressed ISG expression in WT cells, consistent with the above results. As expected, expression of ZC3H10 in KO cells restored its inhibitory effect on ISG expression to levels comparable to those in WT cells (Fig 5A and 5C). However, reintroduction of the NLS-deficient mutant failed to inhibit ISG expression in either WT or KO cells (Fig 5B and 5C). Consistently, in viral replication assays, ZC3H10 knockout reduced HRV16 and NDV-GFP viral protein expression, and reintroduction of wild-type ZC3H10 partially restored viral protein expression in the knockout background (Fig 5D). Viral titer assays yielded similar results (Fig 5E). Together, our data suggest that the nuclear localization, ZF domains and coiled-coil domain of ZC3H10 contribute to its regulatory function, likely via DNA interactions, enabling suppression of IFNβ signaling.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 5. Re-expression of ZC3H10 in ZC3H10-deficient cells restores its inhibitory effect on ISG expression.

(A) WT and KO cells were transfected with empty vector or ZC3H10 plasmid. After 24 hours, IFNβ (400 U/mL) or control medium was added. Following 12 hours of stimulation, mRNA expression of ISG15, IFIT1, and IFIT2 was measured by RT-PCR. (B) WT and KO cells were transfected with empty vector or NLS deletion mutant. Then, the cells were treated as described in panel A. (C) The expression of ZC3H10 and NLS deletion mutant was detected by western blot. (D, E) WT and KO cells were transfected with empty vector, ZC3H10 plasmid or NLS deletion mutant. After 24 hours, IFNβ (400 U/mL) was added. Following 12 hours of stimulation, cells were infected with HRV16 (MOI = 0.5) or NDV-GFP (MOI = 1). After another 16 hours, viral replication was assessed by western blot analysis (D), and viral titers were measured by TCID₅₀ assay (E). Grayscale intensity of viral protein was quantified using Odyssey Image Studio software. All experiments were repeated at least twice, with one representative result shown here. All data are presented as mean values ± SD. *p < 0.05, **p < 0.01, ***p < 0.001. The data underlying this figure can be found in S1 Data and S1 Raw Images.

https://doi.org/10.1371/journal.pbio.3003881.g005

ZC3H10 inhibits ISG expression by binding to the TTTC motif of ISG promoters

To investigate the mechanisms by which ZC3H10 inhibits ISG expression, we first examined whether its expression and localization are influenced by interferon stimulation. Cells were treated with IFNβ, and immunofluorescence analysis revealed that ZC3H10 remained constitutively nuclear, with no discernible changes in expression or nuclear localization following IFNβ treatment (Fig 6A and 6B). These observations suggest that ZC3H10 functions independently of IFN-induced signaling events in the cytoplasm. Consistent with this notion, we found that overexpression of ZC3H10 did not impair STAT1 nuclear translocation upon IFNβ stimulation (Fig 6A).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 6. ZC3H10 can bind to the TTTC motif within the promoters of ISGs.

(A) A549 cells were transfected with ZC3H10 plasmid and stimulated with or without IFNβ. After 6 hours, immunofluorescence was used to detect the expression of ZC3H10 (green) and STAT1 (red). Scale bar, 5 μm. (B) Following stimulation with IFNβ for the indicated time periods, HEK293A cells were harvested. Nuclear and cytoplasmic fractions were separated using a nuclear-cytoplasmic extraction kit. Protein expression of ZC3H10 and phosphorylated STAT1 was analyzed by western blot using anti-ZC3H10, and anti-p-STAT1 antibodies. Lamin A served as a nuclear loading control, and β-actin served as a cytoplasmic loading control. (C) Enriched motifs from ChIP-Seq analysis. (D) Gene expression of cluster 3 ISGs compared with other ISGs from RNA-Seq results in Fig 3D. (E) The number of TTTC/GAAA motifs in the promoter regions of DE-ISGs compared with other ISGs from RNA-Seq results in Fig 3D. (F) HEK293T cells were transfected with ZC3H10. Nuclear extracts were obtained by nuclear-cytoplasmic separation. Double-stranded DNA and sense/antisense single-stranded IFIT1 promoter oligos were used to pull down ZC3H10 protein. Five percent of the input and DNA-bound ZC3H10 protein were analyzed by western blot. (G) The binding sites of ZC3H10 protein and the ISRE sequence were predicted by AlphaFold3. (H) HEK293A cells were transfected with empty vector, ZC3H10, or R89A mutant plasmid. After 12 hours of stimulation with IFNβ, mRNA expression of ISG15, IFIT1, IFIT2, OAS1 and OAS2 was measured. ZC3H10 and R89A proteins were analyzed by western blot. (I) Sense single-stranded DNA oligos of the IFIT1 promoter were used to pull down ZC3H10 and R89A mutant proteins. Five percent of the input and DNA-bound ZC3H10 proteins were analyzed by western blot. (J) HEK293A cells were transfected with plasmids encoding wild-type ZC3H10 or R89A, with an empty vector serving as a control. At 24 h post-transfection, cells were stimulated with IFNβ (400 U/mL) for 12 hours. Then cells were infected with HRV16 (MOI = 0.5) or NDV-GFP (MOI = 1). At 16 hours post-infection, total cultures were harvested. After freeze-thaw cycles, viral titers were determined by TCID₅₀ assay. All experiments were repeated at least twice, with one representative result shown here. All data are presented as mean values ± SD. *p < 0.05, **p < 0.01, ***p < 0.001. The data underlying this figure can be found in S1 Data and S1 Raw Images.

https://doi.org/10.1371/journal.pbio.3003881.g006

We next assessed whether ZC3H10 negatively regulates IFNβ signaling at the receptor level. qRT-PCR analysis revealed that ZC3H10 overexpression did not alter the transcriptional expression levels of IFNAR1/IFNAR2 (S2A Fig). Given that ZC3H10 did not affect upstream IFNβ signaling components, we hypothesized that ZC3H10 directly regulates ISG transcription. To test this hypothesis, we examined whether ZC3H10 influences ISG mRNA stability. HEK293A cells overexpressing ZC3H10 were treated with actinomycin D to block transcription, and the decay rates of representative ISG transcripts were monitored. As shown in S2B Fig, ZC3H10 did not significantly alter the degradation rates of ISG mRNAs.

We therefore investigated whether ZC3H10 directly associates with ISG gene promoters to regulate transcription. To this end, we performed ChIP-Seq using a flag-ZC3H10 overexpression system. Chromatin fragments were immunoprecipitated with an anti-flag antibody, and the enriched DNA was subjected to high-throughput sequencing. Motif enrichment analysis showed that ZC3H10 preferentially bound the 5′-TTTC-3′ sequence and its complementary 5′-GAAA-3′ motif (Fig 6C). Notably, both of which reside within the canonical ISRE sequence (5′-A/GGTTTCN(1-2)TTTCC/T-3′), a well-characterized regulatory element present in the promoters of numerous ISGs. Previous studies have categorized ISGs into five clusters based on their expression patterns, with cluster 3 predominantly featuring antiviral effectors such as the OAS and IFIT families, alongside critical regulators including USP18, STAT1, and IRF9 [22]. This cluster showed strong enrichment for the ISRE motif, which is recognized by the interferon-stimulated gene factor 3 (ISGF3) complex. Our RNA-Seq analysis demonstrated that the expression of cluster 3 genes was markedly increased in ZC3H10 knockout cells compared to WT controls (Fig 6D). Within this cluster, we identified 18 ISGs with significantly elevated expression levels in ZC3H10-deficient cells, which we designated as differentially expressed ISGs (DE-ISGs). To assess whether DE-ISGs are enriched for ZC3H10 binding sites, we quantified the presence of the TTTC and GAAA motifs in their promoter regions (Figs 6E and S3A). Notably, we found that DE-ISGs harbored significantly more of these motifs compared to other ISGs (Fig 6E). This finding supports the model that ZC3H10 suppresses ISG expression by directly binding to these motifs within their promoter regions.

To further validate the ChIP-seq results demonstrating ZC3H10 binding to the TTTC motif, we performed a DNA pull-down assay using biotinylated oligonucleotides. Nuclear extracts containing Flag-ZC3H10 were incubated with DNA oligonucleotides containing the wild-type TTTC motif within the ISRE sequence of the IFIT2 promoter. This assay revealed that ZC3H10 specifically interacted with oligonucleotides encompassing the wild-type ISRE sequence (Fig 6F). As a control, we synthesized oligonucleotides harboring a mutant ISRE sequence and found that the binding of ZC3H10 to the mutant ISRE was markedly diminished (S3B Fig). These results demonstrate that ZC3H10 preferentially binds the ssDNA form of the ISRE containing the TTTC motif.

Previous studies have demonstrated that ZC3H10 requires phosphorylation at serine 126 (S126) to bind the distal promoter region of UCP1, thereby activating its transcription and promoting thermogenesis [39]. To determine whether this phosphorylation event is also required for ZC3H10-mediated repression of ISGs, we assessed the functional consequences of S126A and S126D phosphor-mutants. Notably, neither mutation affected the ability of ZC3H10 to repress ISG expression (S3C Fig), indicating that S126 phosphorylation is dispensable for this function.

To gain structural insight into the molecular basis of ZC3H10-mediated ISG repression, we employed AlphaFold3 to model the binding of ZC3H10 to the ISRE sequence (5′-AGTTTC-3′). The predicted model identified arginine 89 (R89) as a critical residue for DNA binding (Fig 6G). To experimentally validate this prediction, we generated a ZC3H10 mutant in which arginine 89 was substituted with alanine (ZC3H10-R89A). Overexpression of the R89A mutant in HEK293A cells failed to suppress ISG expression, in contrast to wild-type ZC3H10 (Fig 6H). Furthermore, DNA pull-down assays demonstrated that the R89A mutant exhibited severely impaired binding to the ISRE-containing oligonucleotide (Fig 6I). Consistent with this functional requirement, the R89A mutant also lost the ability to antagonize type I interferon signaling and promote viral replication (Fig 6J), further underscoring the physiological relevance of this residue. Collectively, these results demonstrate that ZC3H10 binds directly to ISRE sequences through a mechanism dependent on Arg89 and that this DNA-binding activity is essential for its inhibitory function in type I IFN signaling.

Cell lines and viruses

A549 (CCL-185), HEK293A (CRL-1573), RAW264.7 (TIB-71), and 293T (CRL-3216) cells were purchased from the American Type Culture Collection (ATCC) and maintained in Dulbecco’s modified Eagle’s medium (Invitrogen, Carlsbad, CA), supplemented with 10% fetal bovine serum (HyClone, Logan, UT) and 100 U/ml penicillin and streptomycin at 37°C, 5% CO₂. Newcastle disease virus-GFP (NDV-GFP) and human rhinovirus 16 (HRV16) were obtained from ATCC. Enterovirus A71 (EV71) is the Anhui Fuyang-0805 strain (GenBank accession no. FJ439769.1). The IPBCAMS-WH-01/2019 strain of SARS-CoV-2 was isolated and identified by our laboratory (GISAID: no. EPI_ISL_402123). Viral infection was carried out as described previously [48].

Antibodies

Primary antibodies used were as follows: mouse anti-SARS-CoV/SARS-CoV-2 Nucleocapsid mAb from Sino Biological (Cat# 40143-MM08); mouse anti-Flag mAb from Sigma (Cat# F3165); mouse anti-β-actin mAb from Sigma (Cat# A5441); rabbit anti-ZC3H10 mAb from BGI Genomics Co., Ltd; rabbit anti-STAT1 mAb from Cell Signaling Technology (Cat#9172); mouse anti-Phospho-Stat1 (Tyr701) mAb from Thermo (Cat#QH219604); rabbit anti-STAT2 mAb from Cell Signaling Technology (Cat#72604); mouse anti-EV71 mAb from Millipore (Cat# 60062751); rabbit anti-GFP mAb from Sigma (Cat#G1544); mouse anti-HRV16 mAb from QED bioscience inc. (Cat#18758); mouse anti-IFIT2 mAb from Santa Cruz Biotechnology (Cat#Sc-390724).

Plasmid

A cDNA library was purchased from OriGene Technologies Inc (plasmid backbone: pCMV6-xl4/xl5), as shown in S1 Table. The following plasmids were utilized and stored at the Christophe Merieux Laboratory: pCMV6-entry, pLenti-CRISPRv2, psPAX2, pVSVG. The ZC3H10 R89A mutant constructed using a Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA) according to the manufacturer’s instructions. The primers used for the R89A mutant were Forward: 5′- TGATCTCTATGATATCGCTGACCTTCCTGACAGGGGCTTTGA-3′, Reverse: 5′- CCCTGTCAGGAAGGTCAGCGATATCATAGAGATCATGACGTC-3′.

Primary screening

HEK293A cells were seeded in 96-well plates. The following day, cells were transfected with cDNA plasmids using Lipofectamine 2000 (Invitrogen). At 24 hours post-transfection, the culture medium was replaced with Opti-MEM containing 400 U/mL of IFN-β. Cells were incubated with IFN-β for 12 hours. Then, the medium was removed, and cells were infected with NDV-GFP at an MOI of 1. After 16 hours, the supernatant was discarded, and 4% paraformaldehyde was added for fixation for 20 min. Following DAPI staining, fluorescence images were collected using the Operetta High Content Analysis System (PerkinElmer). The infection rate was calculated as: Infection rate = (GFP⁺ cell number/total cell number) × 100%. The Z-score was calculated as (infection rate − mean infection rate)/standard deviation.

Reporter assays

293T cells were seeded in 24-well plates at a density of 2.5 × 10⁵ cells per well. The following day, cells were transfected using Lipofectamine 2000 (Invitrogen) with the following plasmids, 100 ng of pFL-ISRE-luciferase or pFL-IFIT1 promoter-luciferase, 5 ng of pRL-SV40, and 400 ng of the target gene plasmid. At 24 hours post-transfection, the culture medium was replaced with Opti-MEM containing 400 U/mL of IFN-β. Cell were incubated for 12 hours. Cells were harvested, and cell lysates were subjected to luciferase activity assays which were performed using the Dual-Luciferase Reporter Assay System (Promega). Activation degree was calculated as: Relative luciferase activity = Firefly fluorescence/Renilla fluorescence.

SiRNA transfection for ZC3H10 knockdown

RAW 264.7 cells were seeded in 24-well plates at a density of 100,000 cells. After 24 hours, the cells were transfected with siRNA using the Lipofectamine RNAiMAX (Invitrogen), with a final siRNA concentration of 50 nM. Following 48 hours of incubation, the cells were harvested for subsequent RNA and protein expression analysis. The siRNA sequences used were as follows: GCCGUCAUGAUCUCUAUGATT and UCAUAGAGAUCAUGACGGCTT.

Immunoblot analysis

Cells were lysed in RIPA buffer on ice for 30 min. Lysates were centrifuged at 12,000g for 10 min at 4 °C, and protein concentrations were determined using the BCA assay (Thermo Fisher Scientific, #23227). Proteins were denatured in loading buffer by boiling, separated by SDS‑PAGE, and transferred onto nitrocellulose membranes (Pall, #66485). Membranes were blocked for 1 h at room temperature with 5% non‑fat milk in TBST (Tris‑buffered saline with 0.1% Tween‑20), followed by incubation with primary antibodies overnight at 4 °C and then with appropriate IRDye‑conjugated secondary antibodies (Li-COR) for 30 min at room temperature. Signals were detected using an Odyssey infrared imaging system (Li‑COR) and analyzed with Odyssey Image Studio software. Source data are available in S1 Raw Images.

Generation of HEK293A-ZC3H10 Knockout cell lines Using CRISPR-Cas9

The sequence of single-guide RNA (sgRNA) targeting ZC3H10 was 5′-ACAGCTATGCCAACGGTACC-3′. To package the ZC3H10-KO lentivirus, 293T cells were transfected with the following plasmids using Lipofectamine 2000: 1.2 μg of pLentiCRISPRv2-sgRNA, 0.9 μg of psPAX2, and 0.6 μg of pVSVG. After 72 h, lentivirus-containing supernatant was collected. HEK293A cells were infected with the packaged ZC3H10-KO lentivirus. After 12 hours, infected cells were cultured in DMEM, supplemented with 10% FBS and puromycin. After 48 hours, cells were separated into monoclonal populations and expanded to obtain cell lines. Genomic sequencing and western blot were performed to confirm ZC3H10 knockout efficiency.

Real-time fluorescence quantitative PCR

Total RNA was extracted from cells using TRIzol reagent (Invitrogen) following the manufacturer’s instructions. Reverse transcription was performed using an MLV reverse transcription kit (Promega) according to the manufacturer’s protocol. The synthesized cDNA was used as a template for PCR amplification to detect the expression of target genes, including OAS1, OAS2, ISG15, IFIT2, and IFIT1. Primers used were as follows: human OAS1, TGTCCAAGGTGGTAAAGGGTG and CCGGCGATTTAACTGATCCTG; human OAS2, CTCAGAAGCTGGGTTGGTTTAT and ACCATCTCGTCGATCAGTGTC; human ISG15, CGCAGATCACCCAGAAGATCG and TTCGTCGCATTTGTCCACCA; human IFIT2, CTGCAACCATGAGTGAGAA and CCTTTGAGGTGCTTTAGATAG; human IFIT1, TACAGCAACCATGAGTACAA and TCAGGTGTTTCACATAGGC; human GAPDH, CGGAGTCAACGGATTTGGTCGTA and AGCCTTCTCCATGGTGGTGAAGAC.

RNA sequencing and data analysis

Total RNA was extracted from three biological replicates each of wild-type and ZC3H10-knockout cells using TRIzol reagent (Invitrogen, USA). One microgram of total RNA from each sample was used as input for RNA-seq library preparation. Libraries were constructed using the Hieff NGS MaxUp mRNA Library Prep Kit for Illumina (Yeasen, China). Briefly, mRNA was enriched from total RNA using poly-T oligo-attached magnetic beads, fragmented, and reverse transcribed into first-strand cDNA, followed by second-strand cDNA synthesis. After end repair, adaptor ligation, USER enzyme treatment, and PCR amplification, the libraries were purified and quality assessed using an Agilent 2100 Bioanalyzer (Agilent Technologies, USA). Sequencing was performed on an Illumina HiSeq 2500 platform (Illumina, UK).

Raw reads were processed with Fastp for adapter trimming and quality filtering. Clean reads were aligned to the human reference genome GRCh38 using HISAT2, and gene-level counts were generated with featureCounts. Low-expression genes (average CPM < 5 in both groups and detected in < 50% of samples) were filtered out. Differential expression analysis was conducted in R using Bioconductor DESeq2 package (FDR < 0.05, |log₂FC| > 1). Results were visualized with a volcano plot (Bioconductor ggplot2 package. GO enrichment analysis for up-regulated genes was performed using the Bioconductor clusterProfiler package, and results were displayed as bubble charts (ggplot2). A heatmap of upregulated interferon-stimulated genes (ISGs) was generated with Bioconductor heatmap package.

DNA pull-down assay

293T cells were transfected with ZC3H10-Flag plasmids. After 16 hours, cells were scraped off, and nucleoproteins were extracted. Double-stranded DNA (dsDNA) was prepared by annealing non-biotinylated or biotinylated single-stranded DNA (ssDNA). The annealing process was done by heating the ssDNA at 95 °C for 10 min, followed by slow cooling to room temperature. Ten µg dsDNA or ssDNA was mixed with 75 µL of MyOne C1 in 400 µL DB buffer (20mM Tris-HCl (pH8.0), 2M NaCl, 0.5mM EDTA, 0.03% NP40). The mixture was incubated at room temperature for 1 hour to allow DNA binding to the beads. The DNA-linked beads were washed with 1 mL of DB buffer and 1 mL of PB buffer (50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 10 mM MgCl₂, 0.5% NP-40, and a proteinase inhibitor mixture). The washed DNA-linked beads were incubated with 400 µg of nucleoprotein in PB buffer for 2 hours at room temperature, resulting in a total volume of 600 µL. The beads were washed three times with 1 mL of PB buffer. Then, the beads were eluted with 25 μL of 0.1% SDS solution and subjected to immunoblotting.

Immunofluorescence

Cells were fixed with 4% paraformaldehyde at room temperature for 20 min. After fixation, the cells were washed with 1 mL of PBS. 1 mL of 0.5% Triton X-100 (prepared with PBS) was added to the cells and incubated at room temperature for 10 min. After washing three times with PBS, 1 mL of 5% BSA (prepared with PBST) was added and incubated at room temperature for 90 min to block non-specific binding. The primary antibody was diluted with 5% BSA and incubated with the cells overnight at 4 °C. The cells were washed three times with PBST and incubated with the diluted secondary antibody for 1 hour at room temperature. Lastly, the cells were washed three times with PBST and incubated with DAPI at room temperature for 10 min.

Chromatin Immunoprecipitation Sequencing (ChIP-Seq) assay

A549 cells were seeded in 10 cm dishes at a density of 5 × 10⁶ cells per dish. The following day, cells were transfected with 10 µg of either empty vectors or vectors expressing Flag-tagged ZC3H10. ChIP-seq experiments were performed by Biomarker Technologies Co. (Beijing, China) according to a standard crosslinking chromatin immunoprecipitation protocol with minor modifications. Cells were crosslinked with 1% formaldehyde for 10 min at room temperature. Chromatin was isolated and fragmented by sonication. Immunoprecipitation was performed using anti-Flag gels to enrich ZC3H10-associated chromatin fragments. Following extensive washing, DNA-protein complexes were eluted, and DNA was purified using the MinElute PCR Purification Kit (QIAGEN, #28006). ChIP-seq libraries were prepared using the TruePrep DNA Library Prep Kit V2 for Illumina (Vazyme, #TD501) according to the manufacturer’s instructions. Purified DNA fragments were ligated to sequencing adaptors and amplified for library construction. Libraries were subjected to paired-end 150-bp sequencing on an Illumina NovaSeq 6000 platform.

Raw sequencing reads were processed for adapter trimming and quality filtering. Clean reads were aligned to the human reference genome (GRCh38), and enriched binding regions (peaks) were identified using standard ChIP-seq analysis pipelines. Differential binding analysis was performed in R using the Bioconductor DiffBind package to identify peaks with significant changes in enrichment between experimental conditions.

Statistics

Statistical significance was determined using Student t test or one-way ANOVA, calculated using GraphPad Prism software. Data are presented as mean ± SD. Statistical significance was determined as *P < 0.05, **P < 0.01, ***P < 0.001, or ns, P > 0.05.