The duck embryo fibroblast (DEF) cell line was obtained from the American type culture collection (ATCC) and cultured in Dulbecco’s modified Eagle medium (DMEM, Corning, China) containing 10% fetal bovine serum (FBS) (Sigma, US), followed by incubation in 5% CO₂ at 37°C. Three virus strains, NDRV NP03, DAdV-B2 BG61, and SBDSV M15, previously preserved in our laboratory, were propagated in DEFs, and the virus titers were determined by median tissue culture infective dose (TCID ₅₀) assay using the Reed and Muench calculation.

Mab21, 122aa-423aa domain of du-cGAS, was cloned into pET28a(+) using specific primers shown in Table 1 to create a du-Mab21^-pET28a(+) recombinant plasmid. The recombinant plasmid was then transformed into E. coli BL21 for expressing the recombinant protein cGAS-Mab21 using a previously described method (22). The Balb/c mice were immunized with recombinant protein at an immunization interval of two weeks. Anti-Mab21 sera of du-cGAS protein was collected on day 7 after four immunizations.

The specific single-guide RNAs (sgRNAs) targeting the du-cGAS were designed using the CRISPR guide RNA designing website (http://crispr.mit.edu/). The designed sequences are shown in Table 1 . The synthesized oligonucleotide pairs were annealed and cloned into pX459-v2 CRISPR/Cas9 to construct the recombinant plasmid, as previously described (23). To generate the cGAS-knockout DEF cells, we transfected Du-cGAS-px459 into ATCC DEF using PolyFast Transfection Reagent (MCG, US), according to the manufacturer’s instructions. After 48h of transfection, puromycin (0.5 μg/ml) was added to the culture solution to select the positive DEFs. Thereafter, du-cGAS-knockout was verified by western blotting and DNA sequencing. The Du-cGAS knockout cell line was infected with DAdV-B2, SBDSV, and NDRV at the multiplicity of infection (MOI) of 0.25. The treatment and control cells were collected at three time points (24, 48 and 72 hours post-infection, hpi) for total RNA and protein extraction. Moreover, viral replication and gene mRNA expressions were measured by western blotting and qRT-PCR, respectively, as described previously (24). Supernatants were collected to detect the viral titers using the median tissue culture infectious dose (TCID₅₀) assay.

Different functional domains of Du-cGAS, including cGAS, cGAS with 68 aa deletion at the N terminus (cGAS (69aa-423aa)), Mab21 (122aa-423aa), NTase Core (70aa-238aa), and Zinc-Ribbon (289aa-358aa), were cloned into pFlag-CMV5a vector to generate truncated du-cGAS proteins according to a previous method (24). The primers used for cloning are shown in Table 1 . DEF cells with 60–80% confluence was transfected with 6 µg of plasmid DNA per well using PolyFast Transfection Reagent. After 24 hours post-transfection, the DEF cells were infected with DAdV-B2, SBDSV, and NDRV strains at the MOI of 1 and harvested at 12, 24 and 36 hpi for qRT-PCR and western blotting analysis. The culture supernatants were then used for viral load detection using TCID₅₀ assay. The infected cells were collected for RNA and protein extraction. Viral replication was measured by qRT-PCR and western blotting analysis, as described previously (24).

Cell monolayers with or without DAdV B2 infection were subjected to indirect immunofluorescence assay (IFA) to detect the DAdV B2 hexon protein expression. After incubated with primary and secondary antibody as the method described previously (25, 26), the cells were subsequently treated with 4′,6-diamidino-2-phenylindole (DAPI) for 10 min. The fluorescent signals were captured using a BZ-X800 confocal microscope (KEYENCE, Japan). Cell lysates were separated by 12% SDS-PAGE gel (Yamei, China) and transferred to polyvinylidene difluoride (PVDF) membranes. The PVDF membranes were blocked using 3% bovine serum albumin in phosphate-buffered saline (BSA-PBST) at room temperature for 2 h. The membranes were then incubated overnight with indicated primary antibodies (including monoclonal antibodies against DAdV B2, NDRV, SBDSV, and Flag-tag and the mouse anti-du-cGAS sera) at 4°Cand washed with Tris-buffered saline (TBS), followed by incubation with Daylight 680 Affineur Goat Anti-Mouse IgG (1:20,000) (Odyssey, US) at room temperature for 1h. Thereafter, the membranes were imaged using the Odyssey^® DLx (Bio-Techne, USA).

Total RNA was extracted using the FastPure Cell/Tissue Total RNA Isolation Kit (Vazyme, Nanjing, China) containing genomic DNA Eraser, according to the manufacturer’s instructions. Purified RNA was reverse transcribed into cDNA using EasyScript First-Strand cDNA Synthesis SuperMix (Tranz, Beijing). Thereafter, qRT-PCR was performed using the PerfectStart Green qPCR SuperMix (Tranz, Beijing) on the Roche 96 Light Cycler (Roche, US), as previously described. The primer sequences used in this analysis have been described in a previous study (27). The primer sequences of NDRV σB and DAdV-B2 are shown in Table 1 . The relative expression of each target gene was analyzed using the 2^-ΔΔct method, with duck GAPDH as a housekeeping gene.

All data were presented as mean ± standard deviation (SD). The student’s t-test was used to determine the statistically significant differences using the Prism 8 software. A significance level of P < 0.05 was considered significant, P < 0.01 was considered highly significant, and P < 0.001 was extremely significant.

The complete coding sequence of du-cGAS contains 1296 nucleotides, encoding a protein with 432 aa with a molecular weight of about 48.9 kDa ( Figures 1A, C ). Moreover, as shown in Figure 1A , du-cGAS could be divided into the N-terminus domain (NTase Core, 69aa-238aa) and C-terminus domain (Mab21, 121aa-401aa), with the latter also containing a Zinc-Ribbon domain (Zinc, 289aa-301aa). The CDS sequence of du-cGAS was submitted in the GenBank database (Submission ID: 2645032). There was 77.83% and 93.35% amino acid identity between the du-cGAS protein and the chicken (GenBank ID: XP_040522657) and goose (GenBank ID: XP_040409319) cGAS proteins, respectively. According to the phylogenetic analysis, cGAS homologs could be divided into four groups, including Birds, Primates, Rodentia, and Artiodactyla, whereby du-cGAS clustered in the Birds branch ( Figure 1B ).

To generate cGAS knockout in DEF cells, we utilized CRISPR-Cas9 gene editing technology to disrupt the translation of du-cGAS ( Figure 2A ). Thereafter, antibiotic-Puromycin (0.5ug/μL) incubation and subcloning were conducted to generate a stable cGAS-deficient DEF cell line. The DNA sequencing confirmed the presence of 5-12 bp deletion in cGAS deficient cell lines, which caused a frameshift mutation ( Figures 2B, C ). Additionally, no cGAS protein was detected in the cGAS deficient cell line, compared with wild-type DEF cells, via Western blot analysis ( Figure 2D ). These data showed a successful generation of the cGAS knockout (cGAS-KO) DEF cell line by the CRISPR/Cas9 system.

To explore the function of cGAS in response to viral infections, we infected the cGAS-KO DEF cells (cGAS^-/-) and the normal control DEFs (cells transfected with empty vector pX459-v2, EV) with DAdV B2, SBDSV and NDRV viral strains at the MOI of 0.25 and harvested at the indicated time points. As shown in Figure 3A-a , the mRNA expression of the DAdV B2 hexon gene was significantly higher at 48 and 72 hpi compared to DEF-infected WT cells. The DAdV-B2-infected cGAS^-/- DEF cells had higher viral titers in cell culture supernatant than those of EV cells at 48 and 72 hpi ( Figure 3A-b ). Consistently, the expression of DAdV-B2 hexon protein was significantly increased in the cGAS^-/- DEF cells ( Figure 3A-c ). Consistently, silencing of the endogenous cGAS also increased the SBDSV mRNA expression and viral titer at 24 hpi, and significantly increased the SBDSV VP2 protein expression, as shown by qRT-PCR, TCID50, and Western blot analyses ( Figure 3B-a-c ). However, no significant differences were detected in NDRV sigma-B gene expression and viral titer between cGAS^-/- and EV DEF cells in response to NDRV infection ( Figure 3C-a-c ). These data suggest that cGAS deficiency could strongly promote the replication of duck-origin DNA viruses rather than dsRNA viruses.

Since cGAS deficiency greatly promotes the replication of DNA viruses rather than dsRNA viruses, we sought to explore the molecular mechanism underlying this replication enhancement. We evaluated the relative mRNA expression level of the vital genes associated with the cGAS-dependent signaling pathway via qRT-PCR analysis. As shown in Figure 4A , the mRNA expression levels of IFN-α and IFN-β were significantly lower in cGAS^-/- DEF cells than in EV DEF cells at 24 hpi in response to DAdV-B2 infection. Moreover, the mRNA expression of STING was significantly reduced in DAdV-B2 infected cGAS^-/- DEF cells at 48 hpi. The mRNA expression of vital ISGs, including Mx1, OAS, and IFITM1, were also greatly impaired in DAdV-B2 infected cGAS^-/- DEF cells at 24 and 48 hpi. Similarly, the mRNA expressions of STING, IFN-β, and ISGs (Mx1, OAS, IFITM1, and IFITM2) were highly inhibited in SBDSV-infected cGAS^-/- DEF cells at 24 and 48 hpi ( Figure 4B ). However, NDRV infection had no significant effect on the expression of IFNs (IFN-α and IFN-β), STING and the key ISGs (Mx1, OAS, IFITM1, IFITM2, IFITM3, and IFIT5) ( Figure 4C ). These data suggest that the increased viral replication in cGAS^-/- DEF cells was due to the destruction of cGAS/STING-dependent pathways, which resulted in the impaired expression of IFNs and ISGs.

To further investigate the antiviral function of du-cGAS, we overexpressed the du-cGAS in DEF cells and infected the cells with DNA and RNA viruses. As shown in Figure 5A-a , viral copy numbers of DAdV B2 significantly decreased at 12 and 36 hpi compared to control DEFs cells (cells transfected with empty vector pFlag-CMV5a, EV). Overexpressing du-cGAS also suppressed DAdV-B2 replication, lowering the TCID₅₀ titers in culture supernatants ( Figure 5A-b ). Consistently, du-cGAS overexpression in the DEFs significantly reduced DAdV-B2 replication, decreasing the expression of hexon protein, as determined by western blotting analysis ( Figure 5A-c ). SBDSV replication was also inhibited by du-cGAS overexpression, significantly lowering its viral mRNA expression level and viral titer ( Figure 5B-a, b ). Consistent with the results of TCID₅₀ assay and qRT-PCR, the western blotting analysis showed that the expression of SBDSV VP2 protein was decreased in du-cGAS-overexpression cells compared to EV cells ( Figure 5B-c ). Overexpressing du-cGAS also significantly inhibited NDRV replication, as determined by qRT-PCR, TCID₅₀ assay, and western blotting results ( Figure 5C-a, b, c ).

To clarify the possible antiviral mechanism of du-cGAS, qRT-PCR was used to detect the the relative mRNA expression levels of IFNs, STING, and key ISGs in DAdV-B2 infected cGAS-overexpression and EV DEFs. As expected, the mRNA expression levels of IFN-α, IFN-β, STING, OAS, Mx1, IFITM1, IFITM2, IFITM3, and IFIT5 were greatly enhanced in cGAS-overexpression DEFs than those in EV DEFs at 24 and 48 hpi ( Figures 6A–I ). These results indicate that du-cGAS is a vital antiviral component of the innate immune system of ducks.

Since the preceding findings have shown that du-cGAS plays an important antiviral role in viral infections, we determined the effects of the NTase domain and Mab21 domains of du-cGAS on virus replication by constructing eukaryotic expression plasmids of truncated du-cGAS proteins. As shown in Figure 7A b, c , induced expression of full-length du-cGAS and du-cGAS (69aa-423aa) could strongly inhibit the expression of DAdV-B2 hexon protein as detected by IFA. However, induced sole expression of NTase, Mab21 and Zinc-Ribbon proteins of du-cGAS had no significant effects on the the expression of DAdV-B2 hexon protein. In order to further confirm this result, the viral loads were detected by TCID₅₀ assay in the cell culture supernatant and the expression of DAdV-B2 hexon was also detected by western blotting. Consistently, forced expression of full-length du-cGAS and du-cGAS (69aa-423aa) could significantly reduce the viral loads with lower TCID₅₀ titers, as compared to those in NTase, Mab21, Zinc-Ribbon overexpression and EV DEFs ( Figure 7B ). Similarly, the expression of DAdV-B2 hexon protein was obviously reduced in full-length du-cGAS and du-cGAS (69aa-423aa) overexpression DEFs as detected by western blotting ( Figure 7C ). To further address the correlation of these truncated du-cGAS proteins with the antiviral function, qRT-PCR was performed to detect the relative mRNA expression of IFNs, STING and key ISGs in DEFs transfected with these truncated du-cGAS plasmids. As shown in Figure 7D , the relative mRNA expression levels of IFNs-α, STING and key ISGs were greatly induced in du-cGAS and du-cGAS (69aa-423aa) overexpression DEFs. No significant up-regulation was observed in NTase, Mab21, Zinc-Ribbon overexpression DEFs. These results indicate that a relatively intact cGAS domain including NTase and Mab21 domains are required for the antiviral function of du-cGAS.