WDR77 inhibits prion-like aggregation of MAVS to limit antiviral innate immune response

RIG-I-MAVS signaling pathway plays a crucial role in defending against pathogen infection and maintaining immune balance. Upon detecting viral RNA, RIG-I triggers the formation of prion-like aggregates of the adaptor protein MAVS, which then activates the innate antiviral immune response. However, the mechanisms that regulate the aggregation of MAVS are not yet fully understood. Here, we identified WDR77 as a MAVS-associated protein, which negatively regulates MAVS aggregation. WDR77 binds to MAVS proline-rich region through its WD2-WD3-WD4 domain and inhibits the formation of prion-like filament of recombinant MAVS in vitro. In response to virus infection, WDR77 is recruited to MAVS to prevent the formation of its prion-like aggregates and thus downregulate RIG-I-MAVS signaling in cells. WDR77 deficiency significantly potentiates the induction of antiviral genes upon negative-strand RNA virus infections, and myeloid-specific Wdr77-deficient mice are more resistant to RNA virus infection. Our findings reveal that WDR77 acts as a negative regulator of the RIG-I-MAVS signaling pathway by inhibiting the prion-like aggregation of MAVS to prevent harmful inflammation.

Activation of the RIG-I-MAVS pathway is crucial for providing protection against pathogen invasion.However, improper activation can lead to chronic cytotoxicity and autoimmune diseases.Inappropriate or persistent MAVS aggregation has been linked to increased production of IFN-I and autoimmunity in a significant fraction of Systemic Lupus Erythematosus (SLE) patients 18 .Therefore, precise regulation of MAVS activation is necessary to avoid potentially harmful immunopathology 19 .During later viral infection, MAVS activity is negatively regulated by various mechanisms.UBXN1/ GPATCH3 competes with TRAF3/TRAF6 to bind to MAVS for signaling attenuation 20,21 .TSPAN6 binds to MAVS in a ubiquitination-dependent manner, interfering with the recruitment of its downstream signaling molecules and inhibiting MAVS activity 22 .Gp78 localized on the endoplasmic reticulum can mediate the proteasomal degradation of MAVS through the contact of mitochondria-endoplasmic reticulum interactions 23 .Moreover, posttranslational modifications (PTMs), including phosphorylation, ubiquitination, O-GlcNAcylation, and arginine methylation, have also been reported to modulate MAVS activity 8,24 .Phospholipase A2 (cPLA2) in astrocyte cytoplasm activates downstream NF-κB molecules through direct interaction with MAVS, promoting the transcription of several cytokines 25 .TRAF3IP3 binds to MAVS and facilitates its recruitment of downstream TRAF3 during viral infection, thereby promoting antiviral response 26 .ER-localized ATP13A1 is required specifically for the stability and activation of MAVS 27 .
WDR77, also known as MEP50, is a human WD repeat domain protein and a part of the 20 S methylosome complex that includes PRMT5, which is responsible for the methylation of arginine [28][29][30] .WDR77 is expressed in various human tissues 31 and can form distinct complexes with other proteins for different functions 31,32 .Additionally, WDR77 acts as a coactivator for both androgen and estrogen receptors and plays a role in hormonal effects during prostate and ovarian tumorigenesis 33,34 .However, it is currently unclear how WDR77 may contribute to innate antiviral immunity and if it is involved in the RIG-I-MAVS pathway.
We conducted an affinity purification experiment to investigate the unknown mechanism of MAVS-mediated innate immunity and found that WDR77 binds to MAVS.Through direct protein-protein interaction, WDR77 represses the prion-like aggregation of MAVS, thereby negatively regulating MAVS activity after viral infection.Deficiency of WDR77 enhances the innate antiviral response to negativestrand RNA virus both in cells and in vivo.Collectively, our study reveals that WDR77 is a critical negative regulator of the antiviral innate immune response.

WDR77 is a MAVS-associated protein that dampens IFN-β signaling
To uncover potential MAVS-associated factors that regulate its activity, we generated stable cells expressing Flag-MAVS based on parental HEK293T cells lacking endogenous MAVS, i.e.HEK293T-Flag-MAVS (Fig. 1a) and performed Flag-MAVS pull-down assays using M2 beads (Fig. 1b).Silver staining following SDS-PAGE of the pulled-down proteins revealed two distinct bands with apparent molecular weights of ~70 kDa and 40 kDa (Fig. 1c).Mass spectrometry analysis revealed the presence of MAVS, PRMT5, WDR77 and other proteins, indicating the association of MAVS with the PRMT5-WDR77 complex (Fig. 1d).Coimmunoprecipitation assays confirmed the binding of MAVS with both PRMT5 and WDR77, in addition to the known binding partner TRAF2 (Fig. 1e).
We proceeded to explore the involvement of PRMT5 and WDR77 in MAVS signaling.HEK293T cells were transiently transfected with Flag-PRMT5 and Flag-WDR77 with or without infection by Sendai virus (SeV) (Fig. 1f-h).We found that WDR77, but not PRMT5, significantly reduced the activation of IFNB1 promoter in response to SeV infection in a dose-dependent manner (Fig. 1f-h).This was further supported by the observation that Flag-WDR77 also inhibited the activation of ISRE and NFKB1 promoters following SeV infection (Supplementary Fig. 1a-c).Notably, co-expression of Flag-PRMT5 did not enhance the inhibitory effect of Flag-WDR77 on IFNB1 promoter activation (Fig. 1i and Supplementary Fig. 1d).Furthermore, the expression of WDR77 attenuated IFNB1 promoter activation induced by various stimuli, such as SeV, vesicular stomatitis virus (VSV), and poly(I:C) (Fig. 1j, k).Collectively, these findings suggested that WDR77 alone, but not PRMT5, has an inhibitory role in RIG-I-MAVS-mediated IFNB1 induction.

WDR77 deficiency potentiates antiviral immune response and restricts virus replication
To investigate how WDR77 functions in RIG-I-MAVS signaling, short hairpin RNAs (shRNAs) were designed to target endogenous WDR77 and were introduced into HEK293T cells.The knockdown efficiency of WDR77 was confirmed at both mRNA and protein levels (Supplementary Fig. 2a, b).Notably, shRNAs targeting WDR77 or PRMT5 resulted in concurrent reduction of both proteins, as reported previously due to their mutual dependence on each other 29 .Knockdown of either WDR77 or PRMT5 resulted in a significant increase in IFNB1 induction upon SeV or VSV infection (Supplementary Fig. 2c).Furthermore, cells stably expressing these shRNAs were obtained, which also showed enhanced IFNB1 induction in the absence of WDR77 (Supplementary Fig. 2d-g).
Additionally, we generated wdr77-deficient mouse embryonic fibroblast (MEF) cells (Supplementary Fig. 3f) and observed a significant increase in Ifnb1 induction upon SeV or VSV infections (Supplementary Fig. 3g).We also observed the inhibitory effect of WDR77 on IFNB1 induction in HeLa cells (Supplementary Fig. 3h, i).To investigate whether WDR77 plays a role in the DNA-sensing pathway such as cGAS-STING signaling, we generated WDR77 -/- HEK293 cell lines (Supplementary Fig. 3j).Consistent with abovementioned data, we observed a significant upregulation of IFNB1 induction in WDR77 -/-HEK293 cells in response to SeV, VSV, or poly(I:C) stimulation (Fig. 2i, j).In contrast, the induction of IFNB1 by Herring testis DNA (HT-DNA) was not affected in the absence of WDR77 (Fig. 2i, j).Furthermore, IRF3 dimer formation, a hallmark of activation of both RIG-I-MAVS and cGAS-STING pathways, was significantly increased in WDR77-deficient HEK293 cells, correlating with enhanced IFNB1 induction (Fig. 2k).Notably, transient expression of wild-type WDR77 suppressed the enhancement of IFNB1 induction in a dose-dependent manner in WDR77 -/-HEK293T cells (Fig. 2l and Supplementary Fig. 3k).These findings collectively demonstrate that WDR77 negatively regulates IFN-β signaling mediated by the RLRs pathway.

WDR77 targets MAVS to inhibit IFN-β signaling
In order to investigate the mechanism by which WDR77 suppresses RIG-I-MAVS signaling, an epistasis analysis was performed.HEK293T cells were transiently transfected with WDR77, together with RIG-I(N), MAVS, TBK1, IRF3(S396D), and IKKβ, and the activation of IFNB1 was measured.The results showed that overexpression of WDR77 crippled IFNB1 activation by RIG-I(N) and MAVS, but had no effect on IFNB1 activation by TBK1, IRF3(S396D), or IKKβ (Fig. 3a-c and Supplementary Fig. 4a).Furthermore, WDR77 depletion potentiated the activation of ISRE reporter induced by RIG-I(N) or MAVS, but had no effect on ISRE activation induced by TBK1 or IRF3(S396D) (Supplementary Fig. 4b, c).These findings suggest that WDR77 functions upstream of TBK1, likely at or downstream of MAVS, in the RIG-I pathway.To determine whether WDR77 inhibits IFNB1 induction through MAVS, WDR77 and MAVS double knockout cells (DKO) were generated (Supplementary Fig. 4d).The results demonstrated that the enhancement of interferon induction in the absence of WDR77 is indeed dependent on MAVS, as there was a complete loss of IFNB1 induction upon SeV stimulation in both MAVS -/-and DKO cells (Supplementary Fig. 4e, f).
Consistent with the aforementioned results showing that MAVS binds to WDR77, co-immunoprecipitation experiments demonstrated that WDR77 interacts specifically with MAVS but not with RIG-I, TBK1, or IRF3 (Fig. 3d).Interestingly, WDR77 also co-immunoprecipitated with TRAF3, but not TRAF2, TRAF5, or TRAF6 (Fig. 3d).Furthermore, immunofluorescence analysis showed that WDR77 colocalizes with MAVS in cells, and the interaction between WDR77 and MAVS was enhanced upon virus infection (Fig. 3e).These findings suggest that WDR77 may act on MAVS to negatively regulate IFN-β induction.

WDR77 inhibits MAVS aggregation and activation
We then went to examine how the interaction between MAVS and WDR77 is modulated during viral infection.In the absence of SeV infection, Flag-WDR77 bound to HA-MAVS, and upon SeV infection, the association between Flag-WDR77 and HA-MAVS was substantially increased (Fig. 4a).Furthermore, enhanced interaction was observed between endogenous WDR77 and MAVS upon stimulation (Fig. 4b, c).On the other hand, the interaction between WDR77 and TRAF3 was not influenced by virus infection (Supplementary Fig. 5a, b).Additionally, WDR77 was not required for TRAF3 recruitment to MAVS, suggesting that TRAF3 might not be involved in the interaction between WDR77 and MAVS (Fig. 4b).Furthermore, we examined the interaction between MAVS and WDR77 at different time points post virus infection.Our findings indicated that the MAVS-WDR77 interaction was augmented at the early stage from zero to 16 h, and then declined at the later stage from 16 to 32 h (Fig. 4c).Notably, the association between WDR77 and RIG-I or MDA5 was not detected throughout the entire time course examined.Based on these observations, we speculate that WDR77 may specifically participate in dampening MAVS activity once an effective innate immune response has been reached to prevent deleterious inflammation.
MAVS is known to form prion-like aggregates to mediate antiviral signaling 14 .Overexpression of HA-MAVS leads to its activation and prion-like aggregation independent of stimulation, which can be detected through semi-denaturing detergent agarose-gel electrophoresis (SDD-AGE) (Fig. 4d and Supplementary Fig. 5c, d).Our results showed that HA-MAVS aggregation was suppressed by Flag-WDR77 in a dose-dependent manner.Similarly, prion-like aggregation of endogenous MAVS induced by SeV infection was dampened by WDR77 expression (Fig. 4e and Supplementary Fig. 5e).As a result, MAVS downstream signalings, including IRF3 phosphorylation and dimerization, were all attenuated (Fig. 4f and Supplementary Fig. 5f).Furthermore, MEF cells deficient in Wdr77 displayed increased MAVS aggregation upon viral infection, indicating a conserved inhibitory mechanism of WDR77 in antiviral signaling from mouse to human (Fig. 4g).These findings suggest that WDR77 acts as a negative regulator of RLR signaling by repressing MAVS aggregation.
We next investigated the role of WDR77 in the regulation of MAVS aggregation over time, and we performed a time course analysis of MAVS aggregation following virus infection.In wild type cells, MAVS aggregation increased from zero to 16 h and then decreased (Fig. 4h).Notably, in WDR77-deficient cells, MAVS aggregation displayed a similar pattern, but decreased at a slower rate after 16 h.We also observed a correlation between IFNB1 induction and MAVS aggregation in all samples examined (Fig. 4i and Supplementary Fig. 5g).These findings provide further support for our earlier hypothesis that WDR77 plays a role in downregulating MAVS activity following an effective innate immune response.

WDR77 impedes MAVS to form prion-like filaments
The aggregation and activation of MAVS in cells have been reported to depend on its ubiquitination 36 .However, our data showed that overexpression of WDR77 did not have an effect on MAVS ubiquitination (Supplementary Fig. 6a).This suggests that WDR77 inhibits MAVS aggregation through a mechanism that is independent of its ubiquitination.To further understand how WDR77 may inhibit MAVS aggregation, we performed an in vitro reconstitution assay to examine its inhibitory effect on MAVS activation.HEK293T cells were separated into to S5 and P5 fractions as previously reported 37 , where the S5 fraction contains cytoplasmic components such as RIG-I, TBK1, IRF3 and WDR77, while the P5 fraction contains organelles like mitochondria, where MAVS resides (Supplementary Fig. 6b).The assay was then performed in two steps: in step 1, S5 fraction from SeV-infected cells was incubated with P5 fraction from uninfected cells, and in step 2, P5 fraction isolated from step 1 was further incubated with S5 from uninfected cells to examine MAVS aggregation and IRF3 dimerization (Supplementary Fig. 6c).P5 fractions from wild type or WDR77deficient cells showed no difference in MAVS activation (Fig. 5a).However, the S5 fraction from WDR77 -/-cells infected with SeV was more effective in inducing MAVS aggregation in P5 fractions (Fig. 5b), resulting in more pronounced IRF3 dimerization.These results suggest that WDR77 in the S5 fraction has an inhibitory role in triggering the formation of MAVS aggregates.e Plasmids as indicated were co-transfected into HEK293T cells.36 h after transfection, cells were harvested for immunoprecipitation and immunoblotting.Asterisk indicated nonspecific bands.f-h HEK293T cells were transfected with luciferase reporters and increasing amounts of plasmids as indicated for 24 h, and then stimulated with or without SeV for 12 h.Cells were collected and IFNB1 promotor activation was detected by luciferase assay (f).IFNB1 induction was measured by quantitative PCR (qPCR) (g).Culture medium was collected and IFN-β was detected by ELISA (h) (For f, IFNB1: ****p < 0.0001, ****p < 0.0001, ns p = 0.9540, ns p = 0.4002 in sequence; For g, IFNB1: ****p < 0.0001, ****p < 0.0001, ns p = 0.6511, ns p = 0.1233 in sequence; For h, IFNB1: ***p = 0.0002, ****p < 0.0001, ns p = 0.6413, ns p = 0.0503 in sequence).i HEK293T cells were transfected with luciferase reporters and plasmids as indicated for 24 h, and then stimulated with or without SeV for 12 h.IFNB1 promotor activation was analyzed by luciferase assay (For i, IFNB1: ****p < 0.0001, ns p = 0.0861, ****p < 0.0001, ns p = 0.1511 in sequence).j-k HEK293T cells were transfected with luciferase reporters and Flag-WDR77 plasmids for 24 h, and then stimulated with or without SeV, VSV or poly(I:C) for 12 h.IFNB1 promotor activation was analyzed by luciferase assay (j).IFN-β was detected by ELISA in (k) (For j, IFNB1: ns p = 0.9994; ****p < 0.0001, ****p < 0.0001, ***p = 0.0001 in sequence; For k, IFN-β: ns p > 0.9999, *p = 0.0377, **p = 0.0059, *p = 0.0312 in sequence).Data are representative of one experiment (c and d), two independent experiments with similar results (e), or three independent experiments (f-k).P-values were determined by two-way ANOVA (Dunnett's test) (f) or two-way ANOVA (Šídák's test) (g, h, i, j and k).Source data are provided as a Source Data file.
Furthermore, we purified recombinant MAVS as described previously, except that the proline-rich region is included in this study (Supplementary Fig. 6d) 14 .The recombinant MAVS was separated into two fractions, Peak-I and Peak-II, by size-exclusion chromatography (Supplementary Fig. 6f).According to the previous report, Peak-I contained high molecular weight MAVS filaments, while Peak-II consisted of low molecular weight MAVS particles.MAVS in Peak-II can form prion-like filament spontaneously over time 14 .We next incubated each MAVS fraction with recombinant MBP-WDR77 (Supplementary Fig. 6e) and observed the filament formation under electron microscopy.Consistent with previous reports, the spontaneous aggregation of MAVS in Peak-II fraction was observed after incubation for over 24 h (Fig. 5c) 14,38 .However, recombinant MBP-WDR77 protein effectively prevented the spontaneous aggregation of MAVS in Peak-II fraction.Furthermore, the incubation of MBP-WDR77 with Peak-I promoted disassembly of MAVS filament (Fig. 5d).These results suggested that WDR77 prevents MAVS filament formation through direct proteinprotein interaction.

WDR77 deficiency enhances antiviral immune response in mouse primary cells
To investigate the regulatory function of WDR77 in primary cells, we generated Wdr77 fl//fl mice by CRISPR-Cas9-mediated gene editing, where two loxP sites flank exons 5 and 9 of the Wdr77 gene (Supplementary Fig. 7a).WDR77 is highly expressed in the spleen and thymus, indicating a potential immune-related function (Supplementary Fig. 7b).We then crossed these mice with Lyz2-Cre transgenic mice to create myeloid-specific Wdr77 knockout mice.Successful targeting of Wdr77 was confirmed by sequencing the PCR fragments (Supplementary Fig. 7c) and immunoblotting.We analyzed WDR77 expression in peritoneal macrophages (PEM) and bone marrow-derived macrophages (BMDM) from Wdr77 fl//fl Lyz2-Cre+ (referred to as Wdr77 CKO ) and WT mice (Supplementary Fig. 7d-f).

Wdr77 regulates antiviral innate immunity in mice
To examine the role of WDR77 in antiviral immunity in vivo, we injected WT and Wdr77 CKO mice with VSV via the tail vein.Upon VSV infection, induction of Ifnb1, Ifna4 and ISGs in various organs (spleen, liver and lung) was significantly higher in Wdr77 CKO mice compared with WT mice (Fig. 7a, b, and Supplementary Fig. 8a-c).Additionally, Wdr77 CKO mice showed significantly reduced proliferation of VSV (Fig. 7c) and higher levels of serum IFN-β in response to VSV infection (Fig. 7d).Furthermore, hematoxylin and eosin (H&E) staining revealed more severe inflammation and immune cell infiltration in the lungs of WT mice compared with Wdr77 CKO mice after VSV infection (Fig. 7e).Consistently, Wdr77 CKO mice were less susceptible to VSV infection than WT mice (Fig. 7f).
To investigate the function of WDR77 in immune signaling further, we infected mice with influenza A virus (IAV).Wdr77 CKO mice showed lower mortality than WT mice following IAV infection (Fig. 7g).In contrast, there was no difference in survival between Wdr77 CKO and WT mice following HSV-1 infection (Fig. 7i).Moreover, the production of IFN-β and induction of Ifnb1, Ifna4 in response to HSV infection were not affected by the absence of Wdr77, confirming that WDR77 might not be involved in the DNA-sensing pathway (Fig. 7h and Supplementary Fig. 8d-f).Collectively, these data suggest that WDR77 is an essential negative regulator of the antiviral immune response against negative-strand RNA viruses in vivo.

Discussion
To maintain proper immune homeostasis, antiviral signaling must be precisely regulated.In RLR pathway, a series of crucial proteins are activated in response to viral infection and must be suppressed by distinct mechanisms to terminate the immune response once the pathogen is eliminated.MAVS is a critical adapter protein in the RLR pathway that forms prion-like aggregates, linking viral RNA sensing to downstream signaling activation during virus infection 39 .Our study demonstrates that WDR77, a member of the WD40 family, functions as an important negative regulator of innate antiviral immunity.WDR77 inhibits the prion-like aggregation of MAVS by binding to its prolinerich region, thereby decreasing RLR-mediated IFN-β production and preventing harmful inflammation (Supplementary Fig. 8g).
In our previous study, we discovered a mechanism that prevents spontaneous aggregation of MAVS in cells 38 .However, how the aggregation of MAVS is regulated after an effective antiviral response remains unclear.To investigate this, we used an affinity purification method to search for unknown factors that may associate with MAVS and regulate its activity.We identified the arginine methyltransferase PRMT5 and its cofactor WDR77 as factors that bind to MAVS.It has been reported that PRMT7 and PRMT9, but not PRMT5, covalently modify MAVS to negatively regulate its activity 40,41 .In agreement with these reports, we found that overexpression of PRMT5 had no effect on IFN-β induction mediated by RLR signaling.Surprisingly, overexpression of WDR77 significantly inhibited IFN-β induction in response to SeV infection.Moreover, we observed that WDR77 deficiency resulted in enhanced activation of TBK1/IRF3 and upregulation of IFN-β induction in HEK293T, MEF cells, and HeLa cells upon negative-strand RNA virus but not DNA stimulation.These findings suggest that WDR77 is an important negative regulator specifically involved in the RIG-I-MAVS pathway.
domain, which potentially contributes to the binding of WDR77 to MAVS, surpassing a threshold necessary for WDR77 to inhibit MAVS aggregation.An alternative possibility is that the WD1 domain of WDR77 may have a role in its inhibitory function, in addition to its binding to MAVS.Additionally, our research demonstrates that recombinant WDR77 can effectively impede the prion-like aggregation of MAVS in vitro, highlighting a mechanism facilitated by direct protein-protein interaction.Furthermore, our study showed that WDR77 is recruited to MAVS upon virus infection and prevents the formation of MAVS prion-like filaments, indicating its critical role in regulating IFN-β induction.Notably, mice lacking WDR77 in myeloid cells were less susceptible to negative-strand RNA virus infection and exhibited enhanced transcription of IFN-I and other antiviral genes.These findings suggest that WDR77 serves as a crucial negative regulator of innate antiviral immunity.In contrast, another WD repeatcontaining protein, WDR5, was shown to promote MAVS-mediated antiviral signaling 45 .
Firstly, it is important to acknowledge the limitations of this study.While we have demonstrated the interaction between WDR77 and MAVS, along with its inhibitory effect on MAVS aggregation, further investigations using structural analysis would provide valuable mechanistic insights into the specific interaction between WDR77 and MAVS, elucidating how WDR77 effectively hinders MAVS aggregation.Secondly, it is worth noting that this study does not provide evidence linking WDR77 with other regulators that may confer posttranslational modifications on MAVS.Exploring potential connections and interactions between WDR77 and such regulators would be an intriguing avenue for future research.Additionally, the potential involvement of dysregulated WDR77 in virus infection-related diseases remains to be determined.Future investigations are needed to assess whether any associations exist and to shed light on the implications of WDR77 dysregulation in such conditions.These areas warrant further exploration in subsequent studies.

Ethics Statement
All animal experiments were performed in accordance with the guidelines of the Institutional Animal Care and Use Committee (IACUC) at the Shanghai Institute of Biochemistry and Cell Biology, CAS (Approved No. S346-1704-004).The mice were maintained in a specific pathogen-free (SPF) environment, with appropriate temperature and humidity settings (kept at 25°C and suitable humidity, typically around 50%).They were also subjected to a 12-h dark/light cycle and fed with sufficient water and food.

Mice
Wdr77 fl/fl mice in C57BL/6 background were created by flanking the fifth to ninth exons of Wdr77 with two loxP sites through CRISPR-Cas9mediated gene editing (Cyagen Biosciences).Lyz2-Cre transgenic mice expressing Cre recombinase were generously provided by Dr. Hongyan Wang (SIBCB).Mavs knockout mice were generated as described 26 .Wdr77 fl/fl mice were crossed with Lyz2-Cre mice to get F1 generation.Wdr77 fl/fl Lyz2-Cre + (Wdr77 CKO ) and Wdr77 fl/fl Lyz2-Cre-(control) littermates from the F2 generation were used as future breeders.The genotyping was performed by standard PCR.

Generation of knock-out cell lines based on HEK293T, HEK293, HeLa and MEF
WDR77 -/-and PRMT5 -/-knock-out cells were established by CRISPR/Cas9 technique and the guide RNA sequences are described in Supplemental Table 1.Briefly, guide RNAs were cloned into a CRISPR/Cas9based vector pX330 with a puromycin-resistant selection marker.These vectors were transfected into HEK293T cells by Lipofectamine 3000 (Invitrogen, L3000015).After puromycin (0.5 μg/ml) selection, single colonies were picked and verified by DNA sequencing and immunoblotting.Flag-WDR77 in HEK293T cells for 36 h before immunoprecipitation and immunoblotting (bottom panel) were performed.Asterisk indicates nonspecific bands.g Diagram of WDR77 truncations used in Co-IP experiments (top panel).Various WDR77 truncations were co-expressed with HA-MAVS in HEK293T cells for 36 h before immunoprecipitation and immunoblotting (bottom panel) were performed.h WT or WDR77 -/-#1 HEK293T were transfected with various WDR77 truncations for 24 h, and then stimulated with or without SeV for 12 h.IFNB1 induction was measured by qPCR (IFNB1: all ****p < 0.0001; ns p = 0.9998).Data are representative of three independent experiments with similar results (d, e (left), f and g), or three independent experiments (a-c, e (right) and h) (mean ± SD of three biological replicates).P-values were determined by two-way ANOVA (Šídák's test) (a-c, h) or unpaired two-sided t-test (e).n.s.indicates no statistical significance.Source data are provided as a Source Data file. https://doi.org/10.1038/s41467-023-40567-5 Nature Communications | (2023) 14:4824

Immunoprecipitations and immunoblot analysis
For immunoprecipitation, cells were harvested 36 h after transfection and lysed in IP buffer (HEPES 20 mM pH 7.5, 5 mM MgCl 2 , KCl 10 mM, EGTA 0.5 mM, 1% Triton X-100 and protease inhibitor cocktail (Roche, 11697498001)).Following a brief centrifugation to clear up cellular debris, anti-FLAG M2 affinity gel (Sigma, A2220) or anti-HA agarose beads (Thermo, 26181) were added to cell lysate.After incubation for 4 h at 4 °C, FLAG or HA beads were spun down and washed with lysis buffer three times followed by immunoblotting analysis.For coimmunoprecipitation, cell lysates were collected and incubated with protein G Plus-Agrose Immunoprecipitation reagent (Santa Cruz Biotechnology, sc-2003) together with 1 µg of antibodies as indicated.

Fluorescence microscopy
Wild type and WDR77 -/-HEK293T cells were seeded in 6-well dish with a density of 60%. 12 h post-infection with VSV, fluorescent images were taken for live cells using Olympus IX71 inverted fluorescence microscope.To determine subcellular localization of WDR77, HEK293T cells were transfected with pcDNA3-HA-WDR77 and Flag-MAVS by Lipofectamine 3000.Twenty-four hours after transfection, cells were infected with or without SeV for 8 h.Cells were then washed with pre-warmed 1×PBS, fixed with 4% formaldehyde for 15 min, and permeabilized with 0.5% Triton X-100 for 15 min.After blocking with 3% bovine serum albumin (Roche,10735078001) at room temperature for 1 h, the cells were incubated with anti-HA or anti-Flag antibodies respectively overnight at 4 °C.After incubation, cells were washed with pre-warmed 1×PBS three times.Before fluorescent imaging, cells were incubated with secondary antibodies as indicated and 1 μg/ml DAPI (Beyotime, C1002).All the images were taken with Leica TCS SP8 STED X laser scanning confocal microscope.

Generation of knock-down cell lines by shRNA
Sequences of shRNAs targeting WDR77 and PRMT5 are described in Supplementary Data 1.HEK293T WT cells were seeded in 6-well plates and transfected with shRNA-harboring vector pLKO.1 and viral packaging vectors using Lipofectamine 2000 (Invitrogen, 11668019).Culture medium containing packaged lentiviral particles was collected after 48 h.HEK293T cells were infected with lentiviral particles for 24 h and treated with puromycin for 48 h.Cells were then harvested for various stimulations.Stable knockdown cell lines were obtained by select single colony, which was verified by DNA sequencing and immunoblotting.

Luciferase reporter assay
HEK293T cells were seeded in 12-well plates at a density of 1 × 10 5 cells per well and transfected with 20 ng of reporter gene (IFNB1-luciferase, ISRE-luciferase or NFKB1-luciferase), 20 ng of pCMV-LacZ, and expression vectors as indicated by Lipofectamine 3000.Twenty-four hours post-transfection, cells were stimulated with or without VSV or SeV.After cells were harvested and lysed in Passive Lysis Buffer (Promega, E1941), firefly luciferase activities were measured with a luminometer using the Luciferase Reporter Kit (Promega, E2710) and LacZ activities were measured o-Nitrophenyl-b-D-Galactopyranoside (ONPG) assay following a protocol provided by Sigma Technical Bulletin (GALA-1KT).Induction of firefly luciferase was normalized to LacZ activity.Data were shown as induction fold over controls transfected with empty vectors.

Quantitative PCR
Total RNA was extracted from cells using Trizol (Tiangen, 4992730), and the reverse transcription was performed using HiScript® III SuperMix for qPCR (+gDNA wiper) (Vazyme, R323-01).cDNAs were then used as templates for qPCR assay on Roche Applied Science LightCycler 480 with ChamQ Universal SYBR qPCR Master Mix (Vazyme, Q711-02/03).The induction fold was determined with the ΔΔCq method and qPCR primers used to amplify specific genes were listed in Supplementary Data 1.

ELISA
Concentrations of the IFN-β in cell culture medium and mouse serum were measured by Human Interferon β ELISA kit (Cusabio, CSB-E09889h) or Mouse Interferon β ELISA kit (Cusabio, CSB-E04945m) according to the manufacturer's instructions.

Subcellular fractionation
HEK293T or MEF cells were homogenized in hypotonic buffer (Tris-Cl 10 mM pH 7.5, KCl 10 mM, EGTA 0.5 mM and MgCl 2 1.5 mM) by douncing.After centrifugation at 1,000 g for 10 min at 4 °C, supernatant (S1) and pellet (P1) were obtained.The S1 was further centrifuged at 10,000 g for 10 min at 4 °C, the supernatant was S5 fraction and the pellet was P5 fraction.P5 and S5 were used for IRF3 dimerization assay in vitro followed by native PAGE and immunoblotting as described.

Native PAGE and SDD-AGE
Native PAGE was employed to analyze IRF3 dimerization as described previously 51 .In brief, the samples were mixed with 5×native loading buffer (Tris-HCl 100 mM pH 8.8, 50% Glycerol, 0.04% Bromophenol Blue), and loaded onto a 9% polyacrylamide gel without SDS, which was pre-run in the native running buffer (Tris-HCl 25 mM pH 8.8, Glycine 19.2 mM, 0.4% Doc-Na) for 60 min with a constant voltage of 200 V at 4 °C.After electrophoresis for another 60 min, immunoblotting was performed.SDD-AGE was used to analyze MAVS aggregation.A vertical 1.5% agarose gel was prepared and crude mitochondrial fractions (P5) were obtained by differential Fig. 4 | WDR77 inhibits MAVS aggregation and activation.a HEK293T cells were co-transfected with Flag-WDR77 and HA-MAVS for 24 h followed by SeV stimulation.Cells were harvested and subjected to immunoprecipitation assay and immunoblotting.b WT or WDR77 -/-#1 HEK293T cells were stimulated with or without SeV or VSV for 12 h.Cells were collected and subjected to immunoprecipitation assay and immunoblotting.c HEK293T cells were stimulated with SeV and harvested at various time points as indicated for immunoprecipitation assay and immunoblotting.The sample shown in lane 7 was 4% of input (whole cell lysate (WCL) harvested 32 h post infection).d HEK293T cells were transfected with increasing amounts of Flag-WDR77-expressing plasmids for 24 h, and then transfected with Flag-MAVS plasmids for 12 h.Cells were harvested for subcellular fractionation.P5 fractions were subjected to SDD-AGE to examine MAVS aggregation.e WDR77 -/-#1 cells were transfected with increasing amounts of Flag-WDR77expressing plasmids for 24 h, and then stimulated with or without SeV for 8 h.Cells were harvested for subcellular fractionation.P5 fractions were subjected to SDD-AGE to examine MAVS aggregation.f WT and WDR77 -/-HEK293T cells transfected with or without Flag-WDR77 were stimulated with or without VSV for 8 h.Cells were harvested for subcellular fractionation.S5 fractions were subjected to native PAGE to examine IRF3 dimer, and P5 fractions were subjected to SDD-AGE to examine MAVS aggregation.WCLs were subjected to SDS-PAGE to examine IRF3 phosphorylation.g WT or Wdr77 -/-MEF cells were stimulated with SeV for indicated times.Cells were harvested for subcellular fractionation.S5 fractions were subjected to native PAGE to examine IRF3 dimer, and P5 fractions were subjected to SDD-AGE to examine MAVS aggregation.h WT or WDR77 -/-#1 HEK293T cells were stimulated with SeV.Cells were harvested at specified time points post infection and subjected to immunoblotting.i WT or WDR77 -/-#1 HEK293T cells were stimulated with SeV.Cells were harvested at specified time points post infection and IFNB1 induction was measured by qPCR (IFNB1: all ns p > 0.9999, ****p < 0.0001, **p = 0.0018).Data are representative of two independent experiments with similar results (a-h), or three independent experiments (i) (mean ± SD of three biological replicates with 2 technical replicates).P-values were determined by two-way ANOVA (Šídák's test) (i).Source data are provided as a Source Data file. https://doi.org/10.1038/s41467-023-40567-5 Nature Communications | (2023) 14:4824 centrifugation.P5 was resuspended in 1×SDD loading (10% Glycerol, 0.5×TBE, 2% SDS, and 0.0025% Bromophenol Blue) and loaded onto the agarose gel.Electrophoresis in the running buffer (0.5×TBE, 0.1% SDS) was executed for 40 min with a constant voltage of 100 V at 4 °C, followed by immunoblotting analysis.

Electron microscopy
Copper grids (Ted Pella Inc.) coated with a layer of thin carbon film (5-10 nm) were rendered hydrophilic by glow-discharge in air.Seven microliters of purified MAVS samples were loaded onto the grids, incubated for 3 min, and stained with 2% uranyl acetate.After air drying for a few minutes, the grids were examined under a FEI Tecnai G2 Spirit 120 KV electron microscope.Images were taken in low-dose mode (20 electron/A˚2) at 67,000 x with defocus levels varying between 0.5 to 1.2 microns.Regular calibrations of the microscope magnifications found that the nominal 67,000x used here was always accurate within 3.0% error.

Plaque assay
Plaque assay was performed as described 38 .In brief, Vero cells in 6-well plates were infected with serial dilutions of VSV for 1 h.Serum-free medium was removed and replaced with MEM medium containing 1% low melting temperature agarose (Lonza) and 10% FBS, followed by incubation for 48 h at 37 °C.The wells were stained with 0.1% crystal violet solution for 5 min at room temperature, and the number of plaques in the plate was counted.

Statistical analysis
Data are presented as mean value ± SD (standard deviation) or ± SEM (standard error of the mean) of three independent experiments unless otherwise stated.Statistical analyses were performed with GraphPad Prism 9.0.0 software.Unpaired two-sided Student's t-test, ordinary oneway ANOVA and two-way ANOVA were used to analyze the significance of the differences between groups.Survival data of mice were analyzed using Gehan-Breslow-Wilcoxon test.P < 0.05 was considered statistically significant and n.s.indicates no statistical significance (P > 0.05).

Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability
A reporting summary for this article is available as a supplementary file.The main data supporting the findings of this study are available within In this study, the identification of WDR77 was accomplished through mass spectrometry analysis.Furthermore, the analysis also revealed other MAVS-associated proteins; however, as they are not directly pertinent to the focus of this study, they were not included in the manuscript.Nonetheless, additional information regarding these identified proteins can be provided upon reasonable request.Source data are provided with this paper.

Fig. 1 |
Fig.1| WDR77 binds to MAVS and negatively regulates IFN-β induction.a Immunoblot analysis of MAVS expression in P5 fraction containing crude mitochondria.b Purification procedure for Flag-MAVS.c Silver staining of purified Flag-MAVS.The Flag-MAVS band is indicated by an arrow and the asterisk indicated interested bands.d Peptides identified by mass spectrometry.e Plasmids as indicated were co-transfected into HEK293T cells.36 h after transfection, cells were harvested for immunoprecipitation and immunoblotting.Asterisk indicated nonspecific bands.f-h HEK293T cells were transfected with luciferase reporters and increasing amounts of plasmids as indicated for 24 h, and then stimulated with or without SeV for 12 h.Cells were collected and IFNB1 promotor activation was detected by luciferase assay (f).IFNB1 induction was measured by quantitative PCR (qPCR) (g).Culture medium was collected and IFN-β was detected by ELISA (h) (For f, IFNB1: ****p < 0.0001, ****p < 0.0001, ns p = 0.9540, ns p = 0.4002 in sequence; For g, IFNB1: ****p < 0.0001, ****p < 0.0001, ns p = 0.6511, ns p = 0.1233 in sequence; For
cancers, and is associated with enhanced tumor growth and poor disease prognosis.Our study suggests that dysfunction of WDR77 could lead to inflammation, which may contribute to the development of cancer since

Fig. 5 |
Fig.5| WDR77 impedes MAVS to form prion-like filaments.a A two-step assay was performed.In step 1, HEK293T cells were infected with SeV for 8 h and harvested to get S5 fraction.Increasing amounts of S5 fractions were incubated with P5 fractions from WT or WDR77 -/-HEK293T cells at 30 °C for 1 h.P5 fractions were then isolated.In step 2, P5 fractions from step 1 were incubated with S5 fractions from WT HEK293T cells at 30 °C for another 1 h.IRF3 dimerization was then analyzed by native PAGE and MAVS aggregation was analyzed by SDD-AGE.Blots shown are representative of n = 2 biological replicates.b A two-step assay was performed.In step 1, S5 fractions were prepared from WT and WDR77 -/-HEK293T cells infected with SeV for 8 h, which were incubated with P5 fractions from WT HEK293T cells at 30 °C for 1 h.P5 fractions were then isolated.In step 2, P5 fractions from step 1 were incubated with S5 fractions from WT HEK293T cells at 30 °C for another 1 h.IRF3 dimerization was analyzed by native PAGE and MAVS aggregation was analyzed by SDD-AGE.Blots shown are representative of n = 2 biological replicates.c Recombinant MAVS was separated into two peaks by gel filtration chromatography, i.e., Peak-I and Peak-II.MAVS (Peak-II) was incubated with BSA or recombinant WDR77 at 4 °C.At various time points, the mixtures were subjected to negative staining and examination under electron microscopy.Scale bars indicate 100 nm.Representative images of n = 3. d MAVS (Peak-I) was incubated with BSA or recombinant WDR77 at 4 °C.At various time points, the mixtures were subjected to negative staining and examination under electron microscopy.Scale bars indicate 100 nm.Data are representative of two independent experiments with similar results (a, b), or three independent experiments with similar results (c, d).Source data are provided as a Source Data file.