Our recent study showed the active participation of Aff3ir in EC differentiation from vascular SPCs induced by laminar shear stress (Zhao et al., 2025), suggesting the potential involvement of this novel protein-encoding nested gene in mediating hemodynamic stimulation. To further elucidate the functional role of Aff3ir and its encoded proteins in disturbed shear stress-induced EC activation, we examined the expression of Aff3 and Aff3ir in the intima of mouse aorta. We found that the mRNA level of Aff3ir, but not its parent gene Aff3, was significantly lower in the intima of the aortic arch, an area exposed to disturbed shear stress, compared to the intima of the thoracic aorta, which was exposed to steady unidirectional shear stress (Li et al., 2019; Figure 1A). Aff3ir transcript variants can generate two proteins (Zhao et al., 2025), therefore, we measured the protein levels of Aff3ir-ORF1 and Aff3ir-ORF2. While Aff3ir-ORF1 and Aff3 showed comparable expression levels in the intima of aortic arch and thoracic aorta of mice, the expression of Aff3ir-ORF2 showed an 87% reduction in the intima of aortic arch compared to the intima of thoracic aorta (Figure 1B and C), suggesting that Aff3ir-ORF2 may be a novel mechanosensitive protein that responds to disturbed shear stress. Enface immunofluorescence staining confirmed a marked reduction in Aff3ir-ORF2 expression in the inner curvature of aortic arch compared to both the outer curvature of aortic arch and the thoracic aorta (Figure 1D and E). Moreover, to demonstrate the change in Aff3ir-ORF2 within the same visual field, we examined its expression in longitudinal sections of the mouse aorta (Li et al., 2019). We found that the expression of Aff3ir-ORF2, but not Aff3, was notably downregulated in athero-prone regions (the inner curvature of the aortic arch and bifurcation of the carotid artery) compared to that in the protective region in the outer curvature of the aortic arch (Figure 1F). Additionally, we found that the expression of Aff3, Aff3ir-ORF1, and Aff3ir-ORF2 in the media and adventitia was comparable between the aortic arch and the thoracic aorta (Figure 1—figure supplement 1A, B).

Next, we explored the impact of disturbed shear stress on Aff3ir-ORF2 expression in vitro. Mouse embryonic fibroblasts (MEFs) exhibit responses consistent with those of ECs (Chen et al., 2010; Wen et al., 2013), therefore, we investigated Aff3ir-ORF2 expression in MEFs from WT mice exposed to static or disturbed flow (0.5±4 dyn/cm², 1 Hz). Consistent with our in vivo findings, while disturbed shear stress increased the expression of Vcam1, a critical inflammatory marker of ECs (Nakashima et al., 1998), it significantly reduced both the protein and mRNA levels of Aff3ir-ORF2 (Figure 1G–I). The expression of Aff3 in response to the disturbed flow was minimally affected at both the mRNA and protein levels (Figure 1G–I). These results collectively demonstrate that disturbed shear stress induces a reduction in Aff3ir-ORF2 expression both in vivo and in vitro.

Disturbed shear stress-induced atherogenic responses are initial events in atherosclerotic plaque formation (Davis et al., 2023). To elucidate the regulatory role of Aff3ir-ORF2 in disturbed shear stress-induced inflammation, we overexpressed Aff3ir-ORF2 in MEFs. Aff3ir-ORF2 overexpression attenuated Icam1 expression induced by disturbed shear stress at both the protein and mRNA levels (Figure 2A, B, Figure 2—figure supplement 1A). To further validate our findings in ECs, we overexpressed Aff3ir-ORF2 in human umbilical vein endothelial cells (HUVECs). Consistent with our previous results, Aff3ir-ORF2 overexpression reduced the protein level of Icam1 induced by disturbed shear stress in HUVECs (Figure 2C–D). Moreover, Aff3ir-ORF2 overexpression attenuated disturbed shear stress-induced expression of several inflammatory genes, including Vcam1, Il6, and Il1b, in both MEFs and ECs. (Figure 2—figure supplement 1A–B). Interestingly, we found that Aff3ir-ORF2 overexpression did not affect the basal expression of these inflammatory genes under ST conditions (Figure 2—figure supplement 1A–B), likely due to the relatively low levels of inflammatory gene expression under ST compared to OSS conditions. Notably, despite the significant anti-inflammatory effects of Aff3ir-ORF2, the sequence of this gene is not conserved in Homo sapiens. Furthermore, we measured the concentrations of inflammatory factors, including Il6 and Il1b, in the culture medium of MEFs. As expected, while Aff3ir-ORF2 overexpression had little effect on the concentrations of Il6 and Il1b under ST condition, it significantly reduced their release induced by disturbed shear stress (Figure 2E).

Given the anti-inflammatory effects of Aff3ir-ORF2, we speculated that it may ameliorate disturbed shear stress-induced inflammation and atherosclerosis in vivo. Apoe knockout (Apoe^-/-) mice were subjected to partial ligation surgery to induce disturbed flow in the left carotid arteries (LCAs). The endothelium of the LCAs was intravascularly infected with adenovirus (Ad-Scramble or Ad-Aff3ir-ORF2) prior to surgery (Nam et al., 2009; Zhang et al., 2020). Enface immunofluorescence staining confirmed the successful Aff3ir-ORF2 overexpression in the left carotid artery (Figure 2—figure supplement 1C–D). Mice infected with Ad-Aff3ir-ORF2 exhibited a significant decrease in lesion area in the LCAs compared to those infected with Ad-Scramble (23 ± 17% vs 63 ± 14%) (Figure 2F–H), with no obvious plaque formation observed in the right carotid arteries (Figure 2—figure supplement 1E). Overexpression of Aff3ir-ORF2 also attenuated disturbed flow-induced inflammatory responses, as evidenced by decreased Vcam1 expression in the endothelium of LCAs (Figure 2I–J). These findings suggested that Aff3ir-ORF2 ameliorates shear stress-induced inflammation and atherosclerosis.

To explore the effects of Aff3ir-ORF2 on inflammation and atherosclerosis, we generated Aff3ir-ORF2 global knockout (Aff3ir-ORF2^-/-) mice. Genotyping PCR (Figure 3—figure supplement 1A) and western blot analysis of Aff3ir-ORF2 expression in mouse aortas (Figure 3—figure supplement 1B–C) confirmed the successful knockout. No obvious phenotypic abnormalities were observed in Aff3ir-ORF2^-/- mice up to 20 wk of age and monitoring was discontinued thereafter. Additionally, Aff3ir-ORF2 deficiency did not alter systolic blood pressure, diastolic blood pressure, or mean arterial pressure (Figure 3—figure supplement 1D), suggesting that Aff3ir-ORF2 is dispensable for physiological blood pressure maintenance. We then isolated MEFs from WT and Aff3ir-ORF2^-/- mice. RT-PCR analysis confirmed the deficiency of Aff3ir-ORF2 in Aff3ir-ORF2^-/- MEFs (Figure 3—figure supplement 1E). Interestingly, the Aff3ir-ORF2 knockdown efficiency showed discrepancies between the western blot (Figure 3—figure supplement 1B) and RT-PCR results (Figure 3—figure supplement 1E). In addition to the technical differences between PCR and western blot, the characteristics of Aff3ir-ORF2 may also contribute to this inconsistency. The parent gene, Aff3, is located in a genetically variable region, and it can be excised via intron 5 to form a replicable transposon that translocates to other chromosomes, potentially contributing to leukemia (Chinen et al., 2008; Hiwatari et al., 2003; Miller et al., 2022; Bergh et al., 2002). Aff3ir, located in intron 6, exists within this transposon, which may complicate the measurement of its expression. Furthermore, we found that Aff3ir-ORF2 deficient MEFs displayed higher expression of inflammatory genes, including Icam1, Vcam1, and Il1b, compared to those in WT MEFs, under disturbed flow stimulation (Figure 3—figure supplement 1F).

Next, we crossed Aff3ir-ORF2^-/- mice with Apoe^-/- mice to generate double-knockout (Apoe^-/-Aff3ir-ORF2^-/-) mice. Eight-wk-old Apoe^-/- and Apoe^-/-Aff3ir-ORF2^-/- mice were fed a high-fat diet for 12 wk to induce atherosclerosis (Figure 3A). En face Oil-Red O staining indicated that Aff3ir-ORF2 deficiency accelerated the development of atherosclerosis in the entire aorta, AA, and TA. (Figure 3B and C). Furthermore, Aff3ir-ORF2 deletion increased the lesion area and lipid deposition in the aortic roots of Apoe^-/- mice without altering the collagen fiber content (Figure 3D and E). Similar results were observed in distributing arteries (LCAs) (Figure 3F and G). Given that the expression of adhesion proteins, such as Vcam1 in ECs is crucial for monocyte infiltration into plaques (Kobiyama and Ley, 2018), we assessed Vcam1 expression in the aortic roots of these mice. We found that Aff3ir-ORF2 deletion increased Vcam1 expression in the aortic roots of Apoe^-/- mice (Figure 3H, I), indicating that the atherogenic effects of Aff3ir-ORF2 deletion may result from endothelial inflammation. Additionally, there were no significant differences between the two groups in body weight or triglyceride, total cholesterol, LDL cholesterol, and HDL cholesterol levels (Figure 3—figure supplement 2A, B), indicating that the atherogenic effect of Aff3ir-ORF2 silencing is unlikely to be related to lipid metabolism. Taken together, these results indicate that Aff3ir-ORF2 deficiency aggravates inflammation and atherosclerotic lesions in mice.

To explore the mechanism by which Aff3ir-ORF2 mitigates atherogenesis, we performed RNA sequencing (RNA-seq) on MEFs from WT and Aff3ir-ORF2^-/- mice. Expression of all differentially expressed genes in the Supplementary file 3. The Principal component analysis plot depicted a clear clustering of WT versus Aff3ir-ORF2^-/- samples (Figure 4A). We identified 1167 upregulated and 310 downregulated genes in the Aff3ir-ORF2^-/- group, with a criterion of 1.5-fold change and p<0.05 (Figure 4B, Figure 4—figure supplement 1A). All the differentially expressed genes were subjected to bioinformatics enrichment analysis using Gene Ontology (GO) databases. GO analysis showed that these genes were mainly enriched in processes, including leukocyte cell-cell adhesion, regulation of cell−cell adhesion, and leukocyte activation involved in immune response (Figure 4C), which is highly consistent with the phenotypes observed in Aff3ir-ORF2^-/- mice. To further investigate the functional features of these differentially expressed genes in the context of the atherosclerotic microenvironment, we mapped the differential gene list onto the atherosclerosis-related gene dataset (Rouillard et al., 2016), resulting in 363 overlapping genes. GO analysis of these genes revealed enrichment in processes related to cell−cell adhesion and leukocyte activation involved in immune response (Figure 4—figure supplement 1B), which is highly consistent with the observed effects of Aff3ir-ORF2 on Vcam1 expression.

To further identify the upstream transcriptional regulators of these genes, we used the list of differentially expressed genes from the RNA-seq data to predict upstream transcription factors using the ChEA3 database (Keenan et al., 2019). Then, the top 20 transcription factors obtained from the ChEA3 database were mapped to the atherosclerotic disease-related gene list in the Disgenet database (Piñero et al., 2017). Interferon regulatory factor 5 (Irf5) and Irf8 were identified as key upstream regulators (Figure 4D). Irf5 and Irf8, which are members of the same family of transcription factors originally implicated in interferon production, have been identified as critical regulators of the inflammatory response and contribute to the pathogenesis of various inflammatory diseases (Almuttaqi and Udalova, 2019; Salem et al., 2020). However, their potential roles in disturbed shear stress-induced inflammation remain unclear. We speculated that Aff3ir-ORF2 interacts with Irf5 and/or Irf8. Coimmunoprecipitation assays indicated that endogenous Aff3ir-ORF2 could bind to both Irf5 and Irf8 (Figure 4E). To determine which transcription factor mediates the inflammatory effects of Aff3ir-ORF2 deficiency, we silenced Irf5 and Irf8 in WT and Aff3ir-ORF2^-/- MEFs exposed to disturbed flow. Notably, silencing Irf5, but not Irf8, blunted the upregulation of inflammatory genes, including Icam1, Vcam1, Il6, and Il1b (Figure 4F), suggesting that Irf5 was the predominant factor mediating the anti-inflammatory effects of Aff3ir-ORF2 in the context of disturbed shear stress. Consistently, we found that Irf5 silencing significantly inhibited the upregulation of Icam1 protein levels induced by Aff3ir-ORF2 deficiency under disturbed shear stress (Figure 4G and H). In addition, neither Irf5 nor Irf8 expression levels were affected by Aff3ir-ORF2 deficiency (Figure 4—figure supplement 1C). However, we found that Aff3ir-ORF2 deficiency significantly increased the expression of Irf5-targeted genes (predicted by the ChEA3 database), including Icam1, Ccl5, and Cxcl10 (Figure 4—figure supplement 1D). Notably, the protein level of Irf5 was not significantly affected by disturbed shear stress (Figure 4G and H). Consistently, the mRNA levels of Irf5 were previously reported to be barely changed in the context of disturbed shear stress (Deng et al., 2025) (GSE276195, Figure 4—figure supplement 1E) or the atherosclerotic environment (Li et al., 2025) (GSE222583, Figure 4—figure supplement 1F). Given that the transcriptional activity of Irf5 depends on its nuclear translocation (Lv et al., 2024), we next explored whether Aff3ir-ORF2 affects the subcellular localization of Irf5. Subcellular fractionation assays indicated that Irf5 was predominantly localized in the cytoplasm under static conditions, but exhibited obvious nuclear localization when exposed to disturbed shear stress (Figure 4I and J). While the total expression of Irf5 was barely affected by Aff3ir-ORF2 deficiency or overexpression, nuclear localization of Irf5 increased with Aff3ir-ORF2 deficiency (Figure 4I and J). To further ascertain the role of Aff3ir-ORF2 in regulating the transcriptional activity of Irf5, we performed a luciferase reporter assay (Qiao et al., 2022). Aff3ir-ORF2 overexpression significantly decreased the transcriptional activity of Irf5 (Figure 4K). In summary, these results suggested that Aff3ir-ORF2 acts as an endogenous inhibitor of Irf5 and exerts anti-inflammatory effects by retaining Irf5 in the cytosol.

Next, we investigated the role of Irf5 in disturbed flow-induced atherosclerosis in vivo and whether it mediates the atherogenic phenotype associated with Aff3ir-ORF2 deficiency. Apoe^-/- and Apoe^-/-Aff3ir-ORF2^-/- mice were subjected to partial ligation surgery in the LCAs and intravascularly infected with lentiviruses expressing either Irf5-specific shRNA (lenti-shIrf5) or Scramble shRNA (lenti-shScramble). En-face immunofluorescence staining confirmed successful Irf5 deletion in the left carotid artery (Figure 5—figure supplement 1A). After a 4 wk high-fat diet challenge, Irf5 deletion resulted in an approximately 60% reduction in plaque area in the LCAs of Apoe^-/- mice (Figure 5A and B). In addition, Irf5 deletion attenuated endothelial activation, as evidenced by reduced Vcam1 expression in the endothelium of LCAs (Figure 5C and D). Notably, although Apoe^-/-Aff3ir-ORF2^-/- mice exhibited an increased plaque area in the LCAs compared to Apoe^-/- mice, Irf5 deletion almost completely abolished these differences, reducing both the plaque area and Vcam1 expression in the endothelium of the LCAs (Figure 5A–D). These findings provide in vivo evidence that Aff3ir-ORF2 deficiency-induced atherosclerosis is mediated by endothelial Irf5.

Given the significant anti-inflammatory effects of Aff3ir-ORF2 on endothelial activation and atherosclerosis, we explored the potential use of gene therapy targeting Aff3ir-ORF2 to treat atherosclerosis. Endothelial-specific Aff3ir-ORF2 overexpression was achieved using an EC-enhanced AAV-mediated CRISPR/Cas9 genome-editing system controlled by an EC-specific ICAM2 promoter as we previously reported (Li et al., 2024; Swiech et al., 2015). Apoe^-/- mice infected with AAV-ICAM2-Control or AAV-ICAM2- Aff3ir-ORF2 were fed a high-fat diet for 12 wk (Figure 6A). En-face immunofluorescence staining confirmed successful Aff3ir-ORF2 overexpression in ECs (Figure 6—figure supplement 1A–B). Endothelial-specific Aff3ir-ORF2 overexpression had a minimal effect on triglycerides, total cholesterol, LDL cholesterol, and HDL cholesterol levels in the plasma of mice (Figure 6—figure supplement 1C). However, compared to the negative control, endothelial-specific Aff3ir-ORF2 overexpression significantly reduced the Oil-red O-positive lesion area in the whole aortas of Apoe^-/- mice (19 ± 5% vs 54 ± 8%) (Figure 6B and C). Moreover, endothelial-specific Aff3ir-ORF2 overexpression reduced the lesion area and lipid deposition in the aortic roots of Apoe^-/- mice without altering the collagen fiber content (Figure 6D and E). In addition, Aff3ir-ORF2 overexpression effectively suppressed Vcam1 expression in the endothelium of the aortic roots of Apoe^-/- mice (Figure 6F and G). Collectively, these results suggest that supplementation with Aff3ir-ORF2 was effective in preventing atherosclerosis development.

In conclusion, this study provides novel evidence that the disruption of Aff3ir-ORF2 expression under disturbed flow promotes endothelial inflammatory responses and atherosclerosis. Aff3ir-ORF2 serves as an endogenous inhibitor of Irf5 by binding to Irf5 and preventing its nuclear translocation. Supplementation with endothelial Aff3ir-ORF2 may be a promising therapeutic strategy for treating atherosclerosis.

C57BL/6 and Apolipoprotein E-null (Apoe^-/-) mice were purchased from the Experimental Animal Centre of the Military Medical Science Academy (Beijing, China). Aff3ir-ORF2-heterozygote (Aff3ir-ORF2^+/-) mice were acquired from Dr. Lingfang Zeng’s Laboratory at King’s College London. Briefly, gRNAs targeting the mouse Aff3ir-ORF2 locus (gRNA1: GCAACCCACGGAGTTGCAGTTGG; gRNA2: GTCATTAACTCCTTTAATATAGG; gRNA3: TGCAACTCCGTGGGTTGCTGTGG; gRNA4: GACCACACATAACAGTGAATAGG) and Cas9 mRNA were co-injected into fertilized mouse eggs to generate targeted knockout offspring. F0 founder animals were identified by PCR followed by sequence analysis, and then bred to WT mice to test germline transmission and produce F1 animals. Heterozygous targeted mice were intercrossed to generate homozygous targeted mice. The genotyping primers used were: 5’-GGAAAGACCACAGAATCAATGACA-3’, 5’-AACATTGCTATACCCCACTATA-3’. To generate Apoe and Aff3ir-ORF2 double knockout mice (Apoe^-/-Aff3ir-ORF2^-/-), Apoe^-/- mice were crossed with Aff3ir-ORF2^-/- mice. The animals were maintained at 21 ± 1°C under a 12 hr light/dark cycle (lights on at 07:00, lights off at 19:00) with ad libitum access to water and standard chow unless specified otherwise. This study adhered to the Guide for the Care and Use of Laboratory Animals of the US National Institutes of Health (NIH Publication No. 85–23, revised 2011). All study protocols were approved by the Institutional Animal Care and Use Committee of Tianjin Medical University.

The surgery was performed as we previously described (Yang et al., 2021). Briefly, mice were anesthetized with isoflurane (2–3%). A ventral midline incision (4–5 mm) was made in the neck and the left carotid artery was exposed through blunt dissection of subcutaneous fat and muscle tissue. The left external carotid, internal carotid, and occipital arteries were ligated with a 6–0 silk suture, leaving the superior thyroid artery intact. For adenovirus and lentivirus infection studies, adenovirus (Ad-ORF2 or Ad-Scramble) or lentivirus (lenti-shRNA-Irf5 or lenti-shRNA-Scramble) was introduced into the lumen of the left carotid artery and kept inside for 40 min. After infection, the adenovirus or lentivirus was released, and blood flow to the common carotid artery was restored. Mice were fed with a high-fat diet (TD88137, ENVIGO, USA) immediately after surgery and continued for 4 wk.

Endothelial-specific adeno-associated virus (AAV)-mediated CRISPR/Cas9 shuttle plasmid was constructed by Cell & Gene Therapy (Shanghai, China) as previously reported (Wang et al., 2016). Apoe^-/- mice received a single tail vein injection of recombinant AAV containing an endothelial-specific human ICAM-2 promoter driving Aff3ir-ORF2 overexpression (AAV-Aff3ir-ORF2) or a control empty vector (AAV-Scramble), with a dose of 1×10¹¹ viral genomes in a 200 μL vol of sterile PBS. Subsequently, the mice were fed with a high-fat diet (TD88137, ENVIGO, USA) for 3 mo.

The Apoe^-/-, Apoe^-/-Aff3ir-ORF2^-/-, and EC-specific Aff3ir-ORF2 overexpression mice were anesthetized by inhalation of 2% isoflurane and euthanized by cervical dislocation. The aortas were dissected in 1x PBS and opened to expose the atherosclerotic plaques. After fixation in 4% formaldehyde for 1 hr at 4 °C, the tissues were rinsed in water for 10 min, followed by 60% isopropanol. The aortas were then stained with Oil-Red O for 30 min with gentle shaking, rinsed again in 60% isopropanol, and subsequently rinsed in water three times. The samples were mounted on wax with the endothelial surface facing upwards. Images were captured using an HP Scanjet G4050. Plaque areas were quantified using NIH ImageJ software by calculating the plaque area relative to the total vascular area.

MEFs slides or frozen sections were fixed in 4% paraformaldehyde for 30 min, then permeabilized in 0.1% Triton X-100 (in PBS) and blocked with 1% bovine serum albumin for 30 min at room temperature. Sections were incubated overnight at 4 °C with primary antibodies (1:100). Aff3ir-ORF2 (Cat. No. C0302HL300-4) antibody was from GenScript (Piscataway, NJ, USA). The vWF (Cat. No. ab11713) and VE-Cadherin (Cat. No. ab33168) antibodies were obtained from Abcam (Cambridge, UK). Vcam1 (Cat. No. sc-13160) antibody was from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Following primary antibody incubation, sections were treated with Alexa Fluor 488- or Alexa Fluor 594-conjugated secondary antibodies (1:200, Thermo Fisher Scientific, Grand Island, NY, USA) at room temperature for 1 hr. Slides were then mounted with DAPI-containing mounting medium. Antibody specificity and target staining authenticity were verified using negative controls. Immunofluorescence micrographs were acquired using a Leica confocal laser scanning microscope. Representative images were randomly selected from each group.

Harvested carotid arteries and cross-sections of aortic roots were fixed in 4% paraformaldehyde and embedded in optimal cutting temperature compound (OCT). OCT-embedded tissues were sectioned at a thickness of 7 μm. Slides were immersed in 1x PBS for 5 min to remove OCT, and subsequently stained with Oil-Red O, hematoxylin and eosin (HE), and Masson’s trichrome stain to assess lipid accumulation, lesion area, and collagen deposition, respectively (Li et al., 2019). Images were acquired using microscopy.

Blood samples were obtained via cardiac puncture, rinsed with heparin, and collected in 1.5 mL Eppendorf tubes. Total plasma cholesterol, triglycerides, LDL cholesterol, and HDL cholesterol levels were measured enzymatically using an automated clinical chemistry analyzer kit (Biosino Biotech, Beijing, China).

Mouse Embryonic Fibroblasts (MEFs) were obtained and cultured as previously described (Ferreira and Hein, 2023). Cell passages 4–7 were used in all experiments. MEFs were cultured in the Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 10% FBS, penicillin (100 U/mL), and streptomycin (100 μg/mL). Cells were incubated at 37 °C in a humidified environment containing 5% CO₂ and grown to 70–80% confluence before treatment.

Small interfering RNA against Irf5 or Irf8 were synthesized from General Biosystems (Hefei, China). The sequences of siRNAs are shown in the Supplementary file 1. The MEFs were passaged to six-well plates and transfected with 20 nmol/L siRNA per well using the Lipofectamine RNA iMAX Reagent (Invitrogen, Carlsbad, CA, USA).

For flow experiments, confluent monolayers of MEFs were seeded onto glass slides, and a parallel plate flow system was used to launch oscillatory flow (0.5±4 dyn/cm²). The flow system was enclosed in a chamber (Frangos et al., 1985; Fu et al., 2011).

HEK293T (Cat NO. CRL-11268, ATCC) cells were cultured in DMEM medium with 10% FBS, penicillin (100 U/mL), and streptomycin (100 μg/mL). Cells were incubated at 37 °C in a humidified environment containing 5% CO₂ and grown to 70–80% confluence prior to treatment. Their identity has been authenticated by the supplier. HEK293T cells used here were negative for mycoplasma contamination test.

Aff3ir-ORF2 sequences were inserted into the GV138 vector (CMV-MCS-3FLAG) to generate recombinant adenovirus (Ad-Aff3ir-ORF2). The short hairpin RNA (shRNA) sequences targeting mouse Irf5 were 5’-GGGACAACACCATCTTCAAGG-3’, 5’GGTTGCTGCTGGAGATGTTCT-3’, and 5’-GCCTAGAGCAGTTTCTCAATG-3’. The control shRNA was 5’-GCGTGATCTTCACCGACAAGA-3’. These shRNAs were constructed and cloned into pLV-U6-shRNA-CMV-EGFP to generate recombinant lentivirus (lenti-shRNA- Irf5 or lenti-shCtrl). MEFs were infected with adenovirus or lentivirus at a multiplicity of infection (MOI) of 10, with no detectable cellular toxicity observed.

Whole-cell lysates were prepared in a lysis buffer containing a complete protease inhibitor cocktail, PhosSTOP, and PMSF (Roche, Mannheim, Germany). Cytoplasmic and nuclear proteins were extracted from wild-type and Aff3ir-ORF2^-/- MEFs using a protein extraction kits (Invent Biotechnologies, SC-003, Beijing, China). Protein were separated by SDS-PAGE and transferred to nitrocellulose membranes (Cat. No. 10600001; GE Healthcare; Chicago, IL, USA). The membranes were incubated with primary antibodies. Irf5 (Cat. No. 96527), Irf8 (Cat. No. 98344), and Flag (Cat. No. 14793) antibodies were from Cell Signaling Technology (Danvers, MA, USA). Icam1 (Cat. No. ab222736) antibodies were from Abcam (Cambridge, UK). Aff3ir-ORF2 (Cat. No. C0302HL300-4) and Aff3ir-ORF1 (Cat. No. C0302HL300) antibodies were from GenScript (Piscataway, NJ, USA). Gapdh (Cat. No. 60004–1-Ig) antibody was from Proteintech (Wuhan, China). Aff3 (Cat. No. PA5-68961) antibody was from Thermo Fisher Scientific (Waltham, MA, USA).

After incubation with horseradish peroxidase-conjugated secondary antibodies, the proteins were visualized using enhanced chemiluminescence reagents in a ChemiScope3600 Mini chemiluminescence imaging system (Clinx Science Instruments; Shanghai, China). Protein levels were quantified by measuring integrated density with NIH Image J software (https://imagej.nih.gov/ij/), using Gapdh as a loading control for normalization.

Whole-cell lysates were prepared by lysing cells in a 1% NP-40 lysis buffer containing 50 mM Tris-HCl, 1% Nonidet-P40, 0.1% SDS, and 150 mM NaCl, supplemented with a complete protease inhibitor cocktail (Cat. No. 04693132001; Roche, Indianapolis, IN, USA), a phosphatase inhibitor (PhosSTOP; Cat. No. 04906845001; Roche), and PMSF (Cat. No. IP0280; Solarbio Life Sciences; Beijing, China). Samples were incubated on ice for 30 min, then centrifuged at 12,000 g for 10 min, and the supernatant was transferred to a new tube. Protein concentrations were determined using the BCA Protein Assay Kit (Thermo Fisher Scientific, Grand Island, NY, USA).

For immunoprecipitation, 1000 μg of protein was incubated with specific antibodies at 4 °C for 12 hr with constant rotation. Subsequently, 50 μL of 50% Protein A/G PLUS-Agarose beads was added, and the incubation continued for an additional 2 hr. Beads were washed five times with the lysis buffer and collected by centrifugation at 12,000 g for 2 min at 4 °C. After the final wash, the supernatant was removed and discarded. Precipitated proteins were eluted by resuspending the beads in 2x SDS PAGE loading buffer and boiling for 5 min. The eluates from immunoprecipitation were subjected to Western blot analysis.

The concentrations of Il6 (EM0121) and Il1b (EM0109) in cell culture supernatant were measured using ELISA kit (FineTest, Wuhan, China). The experiments were conducted according to the protocols provided by the manufacturer.

Total RNA was extracted from cells using RNA extraction kits (Transgen Biotech, ER501-01, Beijing, China). Reverse transcription was performed with a reverse transcription kit (Thermo Fisher Scientific, Grand Island, NY, USA). Quantitative PCR was conducted using SYBR Select (Thermo Fisher Scientific) according to the manufacturer’s protocol, with Gapdh serving as the internal control. The primers for quantitative real-time PCR are listed in Supplementary file 2.

The Irf5-binding motif and the full-length ZNF217 promoter were ligated into pGl3-based plasmids (Genechem, Shanghai, China), as previously described (Qiao et al., 2022). HEK293T cells were seeded into 24-well plates and grown to 70–80% confluency. Cells were transfected with the firefly luciferase reporter plasmid containing the Irf5-responsive ZNF217 promoter along with a β-galactosidase reporter plasmid (Promega, Madison, WI, USA) for 24 hr. Subsequently, cells were then infected with adenovirus (Ad-ORF2 or Ad-Scramble) for an additional 24 hr. Relative luciferase activity was measured using a luciferase assay and normalized to β-galactosidase activity as determined by the β-Galactosidase Enzyme Assay System (Promega, Madison, WI, USA).

RNA-seq was performed as we previously described (Li et al., 2024). Wild-type (WT) and Aff3ir-ORF2^-/- MEFs were harvested, and RNA was extracted using the MagicPure Total RNA Kit (TransGen, Beijing, China). Whole transcriptome RNA-seq analysis were conducted by the Beijing Genomics Institute (BGI). Paired-end sequencing in 150 bp length was performed using the DNBSEQ-G400 platform. Raw data was filtered using SOAPnuke (v1.5.6). Differential gene expression analysis, with thresholds set at p<0.05 and fold change ≥1.5, was performed via the BGI website (http://omiscribe.bgi.com). Pathway enrichment analysis was carried out using DAVID tools.

Statistics analyses were performed using GraphPad Prism 8.0. No sample outliers were excluded. At least six independent experiments were performed for all biochemical experiments and the representative images were shown. Unpaired Student’s t-test (two-tailed), one-way ANOVA, or two-way ANOVA with Bonferroni multiple comparison post hoc test were used for analyses, as appropriate. Sample size, statistical method, and statistical significance are specified in Figures and Figure Legends. Levels of probabilities less than 0.05 were regarded as significant.