Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Jun 16.
Published in final edited form as: J Immunol. 2017 Jul 31;199(5):1846–1855. doi: 10.4049/jimmunol.1601493

PACT Facilitates RNA-Induced Activation of MDA5 by Promoting MDA5 Oligomerization

Pak-Yin Lui *,†,1, Lok-Yin Roy Wong *,†,1, Ting-Hin Ho *,†, Shannon Wing Ngor Au ‡, Chi-Ping Chan *,†, Kin-Hang Kok †,§, Dong-Yan Jin *,†
PMCID: PMC9202335  NIHMSID: NIHMS1812050  PMID: 28760879

Abstract

MDA5 is a RIG-I–like cytoplasmic sensor of dsRNA and certain RNA viruses, such as encephalomyocarditis virus, for the initiation of the IFN signaling cascade in the innate antiviral response. The affinity of MDA5 toward dsRNA is low, and its activity becomes optimal in the presence of unknown cellular coactivators. In this article, we report an essential coactivator function of dsRNA-binding protein PACT in mediating the MDA5-dependent type I IFN response. Virus-induced and polyinosinic-polycytidylic acid–induced activation of MDA5 were severely impaired in PACT-knockout cells and attenuated in PACT-knockdown cells, but they were potentiated when PACT was overexpressed. PACT augmented IRF3-dependent type I IFN production subsequent to dsRNA-induced activation of MDA5. In contrast, PACT had no influence on MDA5-mediated activation of NF-κB. PACT required dsRNA interaction for its action on MDA5 and promoted dsRNA-induced oligomerization of MDA5. PACT had little stimulatory effect on MDA5 mutants deficient for oligomerization and filament assembly. PACT colocalized with MDA5 in the cytoplasm and potentiated MDA5 recruitment to the dsRNA ligand. Taken together, these findings suggest that PACT functions as an essential cellular coactivator of RIG-I, as well as MDA5, and it facilitates RNA-induced formation of MDA5 oligomers.

Innate immunity constitutes the first-line defense against virus infection (1). During virus infection, viral genome and the intermediates generated from the replication cycle are sensed by ubiquitously expressed cytoplasmic helicases RIG-I and MDA5 to induce type I IFN production in the infected cells (2). RIG-I and MDA5 are characterized by a common DExD/H-box helicase domain with ATPase activity and a C-terminal domain (3), which work in coordination to encompass the dsRNA stem at the major groove (4). The ATPase activity of RIG-I and MDA5 helps to discriminate nonself from self-RNA (5). Stimulation by the dsRNA ligand results in exposure of the N-terminal caspase recruitment domains in RIG-I and MDA5 to transduce an activation signal to the downstream cascade leading to IFN production (3). Despite their similar domain architecture, the functional roles of RIG-I and MDA5 are different. Particularly, MDA5 recognizes long dsRNA and is responsible for detecting the infection of a group of viruses with a positive-sense ssRNA genome, such as encephalomyocarditis virus (EMCV) (6). MDA5 is also implicated in recurrent rhinovirus infection and certain autoimmune diseases (7–9).

Owing to the open and flat conformation of its C-terminal domain, MDA5 has a lower ligand-binding affinity (10). It requires cellular coactivators for its optimal function. For instance, a catalytically inactive homolog of RIG-I and MDA5, LGP2, positively modulates MDA5–RNA interactions and, hence, the activation of MDA5 (11, 12). RNase L and oligoadenylate synthetase–like protein might also serve as upstream regulators of MDA5 (13, 14). Regulatory roles of the posttranslational modifications of MDA5 have also been described (15, 16); however, additional coactivators of MDA5 remain to be identified and characterized. We have previously identified a dsRNA-binding protein, PACT, as a cellular activator of RIG-I (17). PACT directly interacts with and stimulates the ATPase activity of RIG-I (17). The importance of PACT in the RIG-I–mediated antiviral response has also been demonstrated in the context of infection with several human pathogenic viruses (18–22). In contrast, PACT is also known to be a coactivator of another DExD/H-box helicase, Dicer, which is structurally related to RIG-I and MDA5 (23). The ability of PACT to activate DExD/H-box helicases appears to be conserved from Dicer to RIG-I, but whether and how PACT might impact on MDA5 are still unclear. In this study, we set out to explore the role of PACT in the MDA5-dependent antiviral response, as well as the mechanism of action at the molecular level.

Materials and Methods

Plasmids, cells, and reagents

RIG-I expression and IFN-β–luc reporter plasmids were gifts from T. Fujita (3). The guide RNA (gRNA)-Cas9 coexpression plasmid PX459 for CRISPR/Cas9-mediated genome editing was a gift from F. Zhang (24). Other plasmids have been described elsewhere (17, 20). Point mutants were designed and constructed using a QuikChange XL Site-Directed Mutagenesis Kit (Agilent) and the following primers: PACT-M1 (sense: 5′-CAGGTGAAGGTACAAGTAGGAGGCTGGCGAGACATAGAGCTGCAG-3′), PACT-M2 (sense: 5′-ACTGGAAAGGGGGCATCAAGAAGGCAAGCCAGAAGGAATGCTGCTG-3′), MDA5-570/572 (sense: 5′-CAAACTTATTGTCAAATGAGTCCAAGGTCACGTTTTGGAACTCAACCCTATGAACAA-3′), MDA5-841/842 (sense: 5′-CCTGGTTGCTCACAGTGGTTCAGGAGTTAGGAGACATGAGACAGTTAATGATTTC-3′), MDA5-S88A (sense: 5′-CGGAGAACCGGCGCCCCTCTGGCCGC-3′), and MDA5-R337G (sense: 5′-CAGGGAGTGGAAAAACCGGAGTGGCTGTTTACATT-3′).

HEK293 cells, L929 cells, EMCV (EMC strain), and Sendai virus (SeV; Cantell strain) were purchased from the American Type Culture Collection. JEG-3 cells, HEK293 cells carrying a stably expressed IFN-β–luc plasmid, and wild-type (WT) and PACT−/− mouse embryonic fibroblasts (MEFs) have been used previously (18, 22, 25). All cell lines were maintained in DMEM with 10% FBS at 37°C supplemented with 5% CO2. Cells were passaged and split in a 1:10 ratio in complete DMEM when they reached 80% confluence. For virus infection, EMCV (multiplicity of infection [MOI] = 0.01) and SeV (80 HAU/ml) were diluted in serum-free DMEM and allowed to attach to PBS-washed cells at 37°C for 1 h before removal.

Polyinosinic-polycytidylic acid [poly(I:C)] for transfection, as well as poly(I:C)- and polyC-conjugated beads, were obtained as described (20). Plasmid transfection was achieved by GeneJuice (EMD Millipore), whereas poly(I:C) and short interfering RNA (siRNA) were transfected with Lipofectamine 2000 (Life Technologies, Grand Island, NY).

Construction of PACT-deficient HEK293 cells by CRISPR

Two gRNAs, designated gRNA–PACT-1 (5′-GAGCCTTGGTGCCGTAGCTC-3′) and gRNA–PACT-2 (5′-TTGTTATGTATCTTCTGGA-3′), were used for CRISPR-mediated disruption of PACT loci. The method of CRISPR editing has been described elsewhere (24, 26). In brief, HEK293 cells were transiently transfected with plasmids PX459–gRNA–PACT-1 and PX459–gRNA–PACT-2. After 48 h of incubation, transfected cells were subjected to clonal selection and expansion in DMEM with 3 mg/ml puromycin for 2 d. Surviving clones were allowed to recover for 7 d and were isolated into separate wells for further recovery for four additional days. Phenotypes of the surviving clones were verified by Western blotting.

Luciferase reporter assay, quantitative RT-PCR, and RNA interference

Transfected cells were harvested at the indicated time points and lysed with 100 μl of 1 × Passive Lysis Buffer (Promega). The luciferase reporter activity was measured in a 96-well plate using a Dual-Luciferase Reporter Assay System (Promega) and a MicroLumatPlus LB96V Microplate Luminometer (Berthold Technologies).

Total cellular RNA was extracted using RNAiso Plus reagent (TaKaRa) and reverse transcribed with a Transcriptor First Strand cDNA Synthesis Kit (Roche) and oligo-dT primer. The protocol for quantitative RT-PCR has been described (18, 22). The forward and reverse primers used for detection of endogenous genes and viral genome include mouse Ifna4 (5′-TTCTGCAATGACCTCCATCA-3′ and 5′-TATGTCCTCACAGCCAGCAG-3′), mouse Ifnb (5′-CAGCTCCAAGAAAGGACGAAC-3′ and 5′-GGCAGTGTAACTCTTCTGCAT-3′), and EMCV positive genome (5′-ACACAAACGCAACTGCTGAC-3′ and 5′-CATTAGAGAACGGGGCAAAA-3′).

Sequences of siRNAs used for knockdown of endogenous protein expression are as follows: PACT-specific siRNA (siPACT)-1 (5′-GAGGGAAUACACCACGAUCt t-3′), siPACT-2 (5′-CACCGAUUCAGGUAUUGCAt t-3′), and negative control GFP-specific siRNA (siGFP) (5′-GAACGGCAUCAAGGUGAACt t-3′). siRNA was transfected at a working concentration of 50 μM and incubated for 48 h prior to other treatments.

Coimmunoprecipitation, pull-down, Western blot analysis, and confocal microscopy

Mouse anti-FLAG (M2) and anti–β-actin Abs were from Sigma-Aldrich (St. Louis, MO), whereas mouse anti-Myc (9E10) and anti-V5 Abs were from Santa Cruz Biotechnology (Dallas, TX) and Life Technologies, respectively. Rabbit anti-MDA5 (AT113) antiserum was from Enzo Life Sciences (Farmingdale, NY), and rabbit anti-PACT (ab31967) Abs were from Abcam (Cambridge, U.K.). HRP-conjugated anti-mouse and anti-rabbit IgG secondary Abs were from GE Healthcare Bio-Sciences (Pittsburgh, PA). Using these Abs, poly(I:C) pull-down, coimmunoprecipitation, and Western blotting were performed as previously described (20, 27). Nondenaturing native PAGE for IRF3 dimerization was performed accordingly (28), whereas that for MDA5 oligomerization was modified to a 5% gel.

Confocal microscopy was performed using a Carl Zeiss LSM710 microscope, as described (25, 29). JEG-3 cells were transfected with expression plasmids for 48 h and then induced with poly(I:C) for 3 h. The cells were fixed with 4% paraformaldehyde, permeabilized with methanol-acetone (1:1), and blocked with 3% BSA. FLAG-MDA5 and V5-PACT were detected with rabbit anti-V5 (V8137) and mouse anti-FLAG (F1804; both from Sigma-Aldrich) Abs, respectively, and nuclei were visualized with DAPI.

Results

PACT is essential for EMCV-induced antiviral response

PACT−/− MEFs have been useful in elucidating the importance of PACT in RIG-I function (18, 19, 22). Its role in the MDA5-mediated antiviral response has not been clearly addressed but can be evaluated by the use of EMCV, which is sensed primarily by MDA5 (6). We challenged WT MEFs with EMCV and observed the induction of IFN-β mRNA from 15 to 21 h postinfection (h.p.i.), whereas the expression of IFN-β transcripts in PACT−/− MEFs was undetectable during the entire course of the experiment (Fig. 1A). A similar expression pattern was seen for IFN-α4, another type I IFN (Fig. 1B). In line with this, consistently more robust production of EMCV genome was observed in PACT−/− MEFs than in WT cells (Fig. 1C), collectively reflecting the importance of PACT in inducing IFN and restricting EMCV replication. The same PACT−/− MEFs were tested with SeV infection, which is predominantly detected by RIG-I (6). It has been demonstrated that SeV stimulates the expression of IFN-β as early as 3–6 h.p.i., whereas infection at late time points (after 12 h.p.i.) leads to suppression of IFN-β expression via IRF3 degradation (30). At 18 h.p.i., we observed that SeV infection could potently induce IFN-β expression in PACT−/− MEFs to a level even superior to that observed in WT MEFs (Fig. 1D). This demonstrated that PACT−/− MEFs were not defective in the IFN-β production pathway.

FIGURE 1. PACT is essential for EMCV-induced activation of MDA5.

FIGURE 1.

Open in a new tab

(A–C) WT and PACT−/− MEFs were infected with EMCV at MOI = 0.01. Expression levels of Ifnb (A) and Ifna4 (B) mRNAs, as well as EMCV genome levels (C), were analyzed by quantitative real-time RT-PCR. The relative expression level was calculated with respect to GAPDH mRNA using the 2−ΔCT method. Results are expressed as mean ± SD (n = 3). *p < 0.05, PACT−/− versus WT at the same time point, Student t test. (D) WT and PACT−/− MEFs were infected with EMCV (MOI = 0.01) or SeV (80 HAU/ml) for 18 h. Expression levels of Ifnb mRNA were analyzed by quantitative real-time RT-PCR. (E) PACT was reintroduced into PACT−/− MEFs by ectopic expression 24 h prior to EMCV infection at MOI = 0.01. The fold induction of the Ifnb gene was calculated with respect to the GAPDH transcript in mock-treated cells at time zero. *p < 0.05, PACT versus vector at the same time point, Student t test. ND, not detected.

To corroborate this observation in PACT−/− cells, the loss of endogenous PACT expression was remedied by plasmid transfection in a similar infection experiment. Although vector transfection could, to a certain extent, sensitize PACT−/− cells to IFN-β induction by EMCV, ectopic PACT expression further increased the magnitude of this induction (Fig. 1E), confirming a positive regulatory role for PACT in IFN production. As a complementary piece of evidence, we independently depleted endogenous PACT with an siRNA (siPACT-1) in WT MEFs. The level of endogenous MDA5 remained unchanged, but IFN-β expression was less robust when challenged with EMCV (Fig. 2A). Likewise, in the mouse fibrosarcoma cell line L929, depletion of PACT with siPACT-1 yielded similar results (Fig. 2B). The suppressive effect on IFN-β expression could still be observed when another independent siRNA (siPACT-2) was used (Fig. 2C), indicating that the observation was specific to PACT depletion but was not likely due to an offtarget effect of siRNA. Hence, PACT is an essential coactivator in the MDA5-dependent antiviral response against EMCV.

FIGURE 2. Knockdown of PACT attenuates EMCV-induced activation of MDA5.

FIGURE 2.

Open in a new tab

WT MEFs (A) and L929 cells (B and C) were transfected with siPACT-1, siPACT-2, or siGFP for 48 h. Endogenous PACT mRNA levels (left panels) and protein levels (middle panels), as well as relative expression levels of Ifnb mRNA (right panels), in EMCV-infected cells (MOI = 0.01). *p, 0.05, siPACT versus siGFP at the same time point, Student t test. ND, not detected.

PACT augments dsRNA-induced IFN production

Long dsRNA is the prototypic ligand of MDA5 (31). To mimic long dsRNA, we used high m.w. poly(I:C) as a direct agonist of MDA5. When we ectopically expressed PACT and/or MDA5 in a reporter cell line stimulated with poly(I:C), we found that prior MDA5 expression potently increased the magnitude of IFN-β promoter activation (Fig. 3A), supporting that MDA5 is responsible for the detection of long dsRNA. Although PACT expression alone had minimal effect on the IFN-β promoter, coexpression of PACT with MDA5 further augmented the promoter activity from 6 to 12 h after poly(I:C) induction (Fig. 3A). In addition, when we assessed the effect of loss of PACT on IFN production in PACT−/− cells, we observed impairment in IFN-β and IFN-α4 mRNA expression in response to poly(I:C) induction (Fig. 3B, 3C). As in the case of EMCV infection (Fig. 1E), vector transfection alone sensitized PACT−/− MEFs to poly(I:C) induction. However, the stimulatory effect on the IFN-β promoter caused by ectopic expression of PACT was still visible (Fig. 3D). To further study the effect of PACT deficiency, especially in human cells, we used genome-editing technology by CRISPR to knockout endogenous PACT protein in two independent clones of HEK293 cells. It was observed that IFN-β promoter activation was diminished in both clones in the absence of endogenous PACT protein (Fig. 3E).

FIGURE 3. PACT augments dsRNA-induced activation of MDA5.

FIGURE 3.

Open in a new tab

(A) HEK293 cells carrying a stably expressed IFN-β–luc plasmid were transfected with PACT and MDA5 expression plasmids for 24 h before poly(I:C) induction (1 μg/ml). A luciferase reporter assay was carried out; relative luciferase activity is shown in arbitrary units. Data are the mean and SD of three replicates. *p < 0.05, PACT + MDA5 versus PACT at the same time point, Student t test. WT and PACT−/− MEFs were induced with poly(I:C) as above and analyzed for the expression of Ifnb (B) and Ifna4 (C) mRNAs. *p < 0.05, PACT−/− versus WT at the same time point, Student t test. (D) PACT was reintroduced into PACT−/− MEFs by ectopic expression 24 h prior to poly(I:C) induction at 1 μg/ml. The fold induction of the Ifnb gene was calculated with respect to GAPDH transcripts in mock-treated cells at time zero. *p < 0.05, PACT versus vector at the same time point, Student t test. (E) HEK293 cells were rendered deficient in PACT expression by double-nicking via CRISPR/Cas9-mediated genome editing. Two positive clones (clones 1 and 2) were validated for the expression of PACT protein by Western blotting using parental untreated WT cells as control (left panel). Luciferase reporter assay of IFN-β promoter activity was performed in these cells with overexpression of MDA5 and induction by poly(I:C) (right panel). Results are representative of three independent experiments. Error bars indicate SD. *p < 0.05, clones 1/2 versus WT, Student t test. ND, not detected.

Upon ligand stimulation, MDA5 mobilizes downstream signal transducers and primarily activates IRF3 and NF-κB to drive IFN production (3). To determine how these two transcription factors might be affected, we made use of two reporter constructs, each containing tandem copies of the IRF3 or κB element in the promoter region. Although PACT could robustly potentiate MDA5-dependent activation of IRF3-driven promoter activity, no enhancement of NF-κB–driven promoter activation was observed (Fig. 4A). As a more direct and sensitive way to evaluate IRF3 activation, we visualized IRF3 dimerization by nondenaturing native PAGE. When we expressed PACT with IRF3 and challenged the cells with poly(I:C), potent stimulation of IRF3 dimerization was seen in the presence of PACT (Fig. 4B). Collectively, PACT augments IRF3 activation induced by MDA5.

FIGURE 4. PACT facilitates IRF3 activation.

FIGURE 4.

Open in a new tab

(A) HEK293 cells were transfected with PACT and MDA5 expression plasmids, together with a luciferase construct driven by tandem copies of IRF3 (IRF3-luc; upper panel) or κB (κB-luc; lower panel) elements, for 40 h. Results are representative of three independent experiments. Error bars indicate SD. (B) HEK293 cells were transfected with IRF3 expression plasmid and increasing doses of PACT expression plasmid for 24 h followed by poly(I:C) induction (1 μg/ml). Total lysate was resolved on nondenaturing native PAGE (upper panel) to visualize IRF3 dimerization or on denaturing SDS-PAGE (lower panel) to verify protein expression. *p < 0.01, PACT + MDA5 versus MDA5, Student t test.

PACT and MDA5 are physically associated and responsive to dsRNA

Because it has been observed that PACT functions as an essential activator of MDA5, it will be of tremendous interest to understand how PACT performs its role under physiological conditions. To determine how PACT and MDA5 might interact with viral dsRNA during infection, we recreated this scenario with a biochemical assay using a pull-down experiment. We ectopically expressed PACT and MDA5 in cultured cells and subjected the lysate to pull-down by poly(I:C)-conjugated agarose beads. Although PACT and MDA5 could be detected in the bound fraction of poly(I:C) beads, neither could be recovered from that of polyC beads (Fig. 5A), demonstrating that PACT and MDA5 could be specifically recruited to dsRNA but not ssRNA. To reflect the physiological amount of the two proteins present in cells, we also performed pull-down experiments with endogenous proteins. Because the endogenous level of MDA5 was below the detection level, we pretreated the cells with rIFN-β to induce MDA5 expression; PACT expression was unaffected throughout (Fig. 5B). When the cell lysate was subjected to pull-down, we discovered endogenous PACT and MDA5 in the bound fraction of poly(I:C), but not polyC, beads (Fig. 5B), suggesting that both proteins might be recruited to viral dsRNA during infection.

FIGURE 5. PACT interacts with MDA5, and its activation of MDA5 requires dsRNA binding.

FIGURE 5.

Open in a new tab

(A) HEK293 cells were transfected with MDA5 and PACT expression plasmids for 48 h. Total lysate was subjected to pull-down with poly(I:C)- or polyC-conjugated agarose beads. Bound fraction and input were analyzed by Western blotting. (B) A poly(I:C) pull-down experiment was carried out with endogenous MDA5 and PACT. HEK293 cells were mock treated or treated with rIFN-β (1000 U/ml) for 24 h before harvest. (C and D) HEK293 cells were transfected with MDA5 and PACT expression plasmids, alone or in combination, for 30 h before poly(I:C) stimulation (1 μg/ml). Coimmunoprecipitation was carried out with anti-V5 (C) or anti-FLAG (D) Ab. (E) HEK293 cells carrying a stably expressed IFN-β–luc were transfected with expression plasmids for MDA5 and either PACT-WT or its dsRNA-binding defective mutant (PACT-M1, PACT-M2, or PACT-DM) for 24 h, followed by poly(I:C) induction (1 μg/ml) for 6 h. Protein expression was validated by Western blotting (inset). *p < 0.01 versus PACT-WT, Student t test.

To further explore the physical interaction between the two proteins, we performed a reciprocal coimmunoprecipitation experiment in cultured cells with ectopically expressed PACT and MDA5 in the presence of poly(I:C). When we immunoprecipitated PACT with anti-V5 Ab, MDA5 could be detected in the precipitate (Fig. 5C). Alternatively, when MDA5 was immunoprecipitated with anti-FLAG Ab, PACT could be detected in the precipitate (Fig. 5D). These results show that PACT and MDA5 are physically associated in the presence of dsRNA.

To shed light on the molecular determinant of PACT-mediated activation of MDA5, we created PACT mutants that are defective in dsRNA binding. Among the three dsRNA-binding domains (dsRBDs) of PACT, two contain three conserved lysine residues essential for dsRNA binding (32). We mutated them to arginine to disrupt the salt bridge formed between the dsRBD and dsRNA in one, two, or both dsRBDs to create PACT-M1, PACT-M2, and PACT-DM mutants. Their ability to augment MDA5-induced IFN-β promoter activation in response to poly(I:C) was compared with that of PACT-WT. Interestingly, none of these mutants exhibited an augmentation effect (Fig. 5E) (i.e., disruption of either dsRBD was detrimental to PACT activation of MDA5). Thus, PACT requires its dsRNA-binding capacity to activate MDA5.

PACT induces IFN-β production through MDA5 oligomerization

We next explored the molecular mechanism by which PACT activates MDA5. MDA5 was known to oligomerize upon stimulation with dsRNA ligand (33, 34). To test whether PACT would directly activate and stimulate MDA5 oligomerization, we assessed the oligomerization of endogenous MDA5 protein in cultured cells with ectopic expression of PACT and induction by poly(I:C). Although poly(I:C) induction over time could potently stimulate MDA5 oligomerization, as visualized by native PAGE, PACT expression further enhanced the degree of oligomerization (Fig. 6A), indicating that PACT promotes MDA5 oligomerization.

FIGURE 6. PACT facilitates MDA5 oligomerization.

FIGURE 6.

Open in a new tab

(A) HEK293 cells were transfected with PACT expression plasmid for 24 h followed by poly(I:C) induction (1 μg/ml). Total lysate was resolved on nondenaturing native PAGE (upper panel) to visualize endogenous MDA5 oligomerization or denaturing SDS-PAGE (lower panel) to confirm protein expression. HEK293 cells carrying a stably expressed IFN-β–luc were transfected with expression plasmids for PACT and MDA5-WT, its monomer-interface mutant (MDA5-570/572 or MDA5-841/842) (B), or its constitutively active mutant (MDA5-S88A or MDA5-R337G) (C). Error bars indicate SD (n = 3). *p, 0.01, MDA5 mutant versus MDA5-WT, Student t test.

To evaluate whether PACT augments IFN production by stimulating MDA5 oligomerization, we used two MDA5 monomer-interface mutants, MDA5-570/572 and MDA5-841/842, which carry M570R/E572R and I841R/E842R mutations, respectively, and are defective in forming long oligomeric filaments (4). Although PACT expression could augment IFN-β promoter activation mediated by MDA5-WT in response to poly(I:C), little effect was observed with the two monomer-interface mutants (Fig. 6B). We also assessed the effect of PACT on IFN-β promoter activity activated by two constitutively active mutants of MDA5: MDA5-S88A and MDA5-R337G. The MDA5-S88A mutant carries a nonphosphorylatable residue, which is no longer susceptible to inhibitory phosphorylation under unstimulated conditions (35), whereas the MDA5-R337G mutant carries an Aicardi-Goutières syndrome mutation and constitutively forms a stable oligomer (9). Although PACT expression was fully competent in activating IFN-β promoter activity mediated by MDA5-S88A, no stimulatory effect was observed with the MDA5-R337G mutant (Fig. 6C). Altogether, these data are compatible with the model that PACT modulates IFN-β expression by promoting MDA5 oligomerization.

PACT potentially enhances MDA5 recruitment to dsRNA ligand

So far, we have demonstrated that PACT requires its dsRNA-binding activity to modulate MDA5 activity leading to MDA5 oligomerization. With these observations, we opted to explore the possible role of PACT in promoting MDA5 oligomerization from the perspective of the temporal interaction between MDA5 and its ligand. In particular, we focused on the recruitment of MDA5 to dsRNA in the presence or absence of PACT during early ligand exposure. A similar in vitro poly(I:C) pull-down experiment was repeated to evaluate the recruitment of MDA5 and PACT at early time points, before the attainment of system equilibration (Fig. 7A). As expected, PACT was efficiently recruited to poly(I: C)-conjugated beads as early as 15 min after incubation, owing to its high affinity for dsRNA. However, in the poly(I:C)-bound fraction, MDA5 was only noted at the 30-min time point. In the presence of PACT, MDA5 could be detected in the bound fraction at the 15-min time point, and its recruitment was minimal in the absence of PACT. Coimmunoprecipitation experiments in the presence and absence of the ligand also confirmed a more robust recruitment of MDA5 to the RNA–protein complex that contains poly(I:C) and PACT (Fig. 7B). These results support the notion that PACT potentially activates MDA5 by enhancing its recruitment to the dsRNA ligand.

FIGURE 7. PACT potentiates MDA5 recruitment to dsRNA ligand.

FIGURE 7.

Open in a new tab

(A) HEK293 cells were transfected with MDA5 expression plasmid, alone or together with PACT expression plasmid, and harvested for a poly (I:C) pull-down experiment. The bound fraction of poly(I:C)-conjugated beads was evaluated for MDA5 recruitment at 15 and 30 min after lysate incubation. (B) HEK293 cells were transfected with MDA5 and PACT expression plasmids, alone or in combination for 30 h, and then subjected to mock treatment or poly (I:C) induction. Coimmunoprecipitation was carried out with anti-FLAG or anti-V5 Ab. (C) JEG-3 cells were transfected with expression plasmids of MDA5 and PACT and subjected to mock treatment or poly(I: C) induction. Subcellular localization of PACT (red) and MDA5 (green) was visualized by confocal microscopy. Nuclear morphology (blue) was revealed by DAPI staining. Relative colocalization (yellow) is shown in the merged image. Condensation of the merged signal is indicated by the arrowhead. Scale bars, 20 μm.

The spatial distribution of MDA5 and PACT within host cells was also examined. Confocal microscopy was performed for a JEG-3 trophoblast cell line overexpressing MDA5 and PACT (Fig. 7C). Although it appeared that MDA5 and PACT colocalized in unstimulated cells, stimulation with poly(I:C) resulted in condensation of the signals from the two proteins in an uncharacterized subcellular compartment, which might lead to downstream signaling activation. Therefore, these data suggest that PACT and MDA5 interact with each other in cultured cells, particularly in the presence of dsRNA ligand.

Discussion

In this article, we reported that the dsRNA-binding protein PACT functions as an essential cellular coactivator of cytoplasmic DExD/H-box helicase MDA5 in the innate antiviral response.

This was demonstrated in PACT-deficient MEFs that showed impaired type I IFN production and elevated accumulation of viral genome when challenged by EMCV, which is known to be recognized by MDA5 (Fig. 1A–C). This notion was further supported by our rescue experiment, in which the impairment was reversed with ectopic expression of PACT, as well as by independent siRNA-knockdown and CRISPR/Cas9-knockout experiments, which resulted in the similar suppression of EMCV- or poly(I:C)-induced IFN-β expression as in PACT−/− cells (Figs. 2, 3). We reported previously that PACT could modulate the antiviral activity of RIG-I (17). Although PACT−/− cells were not included in any of the loss-of-function experiments in that study, depletion of PACT by siRNA revealed that IFN-β expression was impaired during SeV infection, which was known to be specific to RIG-I. In this study, we contrasted the impact of PACT deficiency on the immune response induced by RIG-I– and MDA5-dependent viruses (Fig. 1D). To our surprise, IFN-β expression was dampened in WT cells, but not in PACT−/− cells, late during SeV infection. One possible explanation for this phenomenon is that infection at late time points (after 12 h.p.i.) caused a negative regulation of IFN-β expression by IRF3 degradation (30), which could explain the decreased response in WT cells. In addition, PACT−/− cells might have greater NF-κB activity, leading to compensation for the blunted IRF3 signaling, as evidenced by enhanced IL-6 induction in our previous study (18). On the one hand, this observation demonstrated that the deficient cells that we used as a control were competent in the IFN-β–production pathway, whereas, on the other hand, it potentially distinguished the role of PACT in the early or late time course of viral infection.

Well before the discovery of RIG-I–like receptors, the antiviral role of PACT was originally implicated in infection with Newcastle disease virus, in which PACT overexpression augmented viral induction of IFN-β expression (36). Similar observations were noted in this study when a commonly used dsRNA analog of poly(I:C) was adopted (Fig. 3A). However, Sen and coworkers (37), who developed PACT-deficient mice, did not suggest the same. These mice transcribed PACT mRNA with an unexpectedly retained intron 7 carrying a premature stop codon, but they did not express detectable levels of the truncated protein; hence, they were used as PACT−/− mice in later studies. When these mice were tested for several immune pathways, it was observed that PACT was not involved in the antiviral response (38). Later, independent groups, including ours, carried out careful experiments; however, we all arrived at the opposite conclusion: that PACT deficiency attenuates IFN production induced by different types of viruses (18, 19, 22). Despite the use of the knockout cell line from the same source, the discrepancy was likely due to differences in experimental design, as well as the choice of end points. Nevertheless, the antiviral role of PACT has been widely recognized since then.

MDA5 and PACTare known to intrinsically bind dsRNA (3, 39). In our in vitro binding assays, both proteins could be efficiently and specifically recruited to poly(I:C) (Fig. 5A, 5B). Although we did not investigate recruitment during actual virus infection, the two proteins are likely to localize to and concentrate in the same subcellular compartment, as observed in our confocal experiment using dsRNA stimulation (Fig. 7C). From this we theorize that the colocalization of MDA5 and PACT, with the downstream signaling molecule MAVS to convey activation signal, is also possible in view of the colocalization of MDA5 with MAVS during virus infection, as demonstrated in a previous study (40). Under physiological condition of active replication of RNA viruses, transient foci (termed virus-induced stress granules) which are enriched in viral genome and proteins, as well as host immune mediators, are formed in the cytoplasm (41). Particularly, MDA5 was recruited to these stress granules containing viral dsRNA during an early phase of infection with the mengovirus strain of EMCV (42). In contrast, PACT has not been shown to associate with virus-induced stress granules but was found in Ago2 containing stress granules for RNA interference (43). Moreover, the PACT-interacting partner PKR has been demonstrated as an essential modulator of stress granule formation during virus infection (42). In light of this, the presence of PACT in virus-induced stress granules is plausible but remains to be elucidated. In contrast, PKR was recently shown to have a nonredundant and kinase-dependent role downstream of MDA5 in the activation of MAVS (44). This raises the possibility of a similar role for PACT in transducing the activation signal from MDA5 to MAVS. Although this merits further analysis, the loss of MDA5-activating activity by PACT-M1, PACT-M2, and PACT-DM mutants (Fig. 5E), which carry an intact PKR-activating domain (32), suggested that the ability of PACT to activate MDA5 shown in our work might be separate from the activity of PKR that operates downstream of MDA5 in the other study (44).

We demonstrated that MDA5 and PACT could interact with each other in the presence of dsRNA (Figs. 5C, 5D, 7B, 7C). Our findings are generally consistent with the model in which PACT facilitates dsRNA-induced activation of MDA5. PACT mutants defective in dsRNA binding had no MDA5-activating property (Fig. 5E). It is also noteworthy that dsRNA is the physiological inducer of MDA5 oligomerization and filament assembly (33, 34). MDA5 and LGP2 preferentially bind with certain types of viral RNA, such as the N region in the genome of measles virus (45). It is plausible, as in the case of Dicer (46, 47), that PACT could select, concentrate, and transmit RNA ligands to MDA5 specifically to facilitate oligomerization and filament assembly. Indeed, our exploratory experiment intended to study the temporal dynamic interaction between MDA5 and its ligand showed that PACT might facilitate the recruitment of MDA5 to dsRNA (Fig. 7A). Further investigations using other biochemical systems will be required to elucidate the complete molecular basis of this function of PACT in the activation of RIG-I and MDA5. In view of our model for the PACT-dependent facilitation of dsRNA-induced oligomerization of MDA5, it remains incompletely answered why overexpression of PACT alone could promote MDA5 oligomerization (Fig. 6A) and, consequently, IFN-β promoter activation (Fig. 3A), even without dsRNA induction. One possibility could be the sequestration and concentration of endogenous MDA5 ligands by PACT. Another theory relates to the competition for endogenous ligand between PACT and other dsRNA-binding proteins, which prevents the signaling pathway from unwanted activation. For instance, dsRBD-carrying ADAR1 performs A-to-I editing on endogenous dsRNA to evade recognition by MDA5 (48). These two possibilities could be supported by the observations that overexpressed PACT and MDA5 were likely interacting with each other, even in the absence of exogenous stimulation [e.g., poly(I:C) induction], in our confocal and coimmunoprecipitation experiments (Fig. 7B, 7C). However, under physiological conditions where exogenous viral ligand strikes high during infection, endogenous PACT is sufficient and proficient to capture abundant dsRNA entities, such as defective-interfering RNA (21). Nevertheless, we have collectively demonstrated that PACT acts as an essential cellular coactivator of MDA5 in the innate antiviral response.

Acknowledgments

This work was supported by Hong Kong Health and Medical Research Fund Grants 14130862 (to K.-H.K.) and HKM-15-M01 (to D.-Y.J.), Hong Kong Research Grants Council Grants C7011-15R (to D.-Y.J.) and HKU271215/15M (to K.-H.K.), and National Natural Science Foundation of China Grant 81271852 (to K.-H.K.).

Abbreviations used in this article:

dsRBD

dsRNA-binding domain

EMCV

encephalomyocarditis virus

gRNA

guide RNA

h.p.i.

h postinfection

MEF

mouse embryonic fibroblast

MOI

multiplicity of infection

poly(I:C)

polyinosinic-polycytidylic acid

SeV

Sendai virus

siGFP

GFP-specific siRNA

siPACT

PACT-specific siRNA

siRNA

short interfering RNA

WT

wild-type

Footnotes

Disclosures

The authors have no financial conflicts of interest.

References

  • 1.Hoffmann HH, Schneider WM, and Rice CM. 2015. Interferons and viruses: an evolutionary arms race of molecular interactions. Trends Immunol. 36: 124–138. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Yoo JS, Kato H, and Fujita T. 2014. Sensing viral invasion by RIG-I like receptors. Curr. Opin. Microbiol 20: 131–138. [DOI] [PubMed] [Google Scholar]
  • 3.Yoneyama M, Kikuchi M, Matsumoto K, Imaizumi T, Miyagishi M, Taira K, Foy E, Loo YM, Gale M Jr., Akira S, et al. 2005. Shared and unique functions of the DExD/H-box helicases RIG-I, MDA5, and LGP2 in antiviral innate immunity. J. Immunol 175: 2851–2858. [DOI] [PubMed] [Google Scholar]
  • 4.Wu B, Peisley A, Richards C, Yao H, Zeng X, Lin C, Chu F, Walz T, and Hur S. 2013. Structural basis for dsRNA recognition, filament formation, and antiviral signal activation by MDA5. Cell 152: 276–289. [DOI] [PubMed] [Google Scholar]
  • 5.Louber J, Brunel J, Uchikawa E, Cusack S, and Gerlier D. 2015. Kinetic discrimination of self/non-self RNA by the ATPase activity of RIG-I and MDA5. BMC Biol. 13: 54. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Kato H, Takeuchi O, Sato S, Yoneyama M, Yamamoto M, Matsui K, Uematsu S, Jung A, Kawai T, Ishii KJ, et al. 2006. Differential roles of MDA5 and RIG-I helicases in the recognition of RNA viruses. Nature 441: 101–105. [DOI] [PubMed] [Google Scholar]
  • 7.Lamborn IT, Jing H, Zhang Y, Drutman SB, Abbott JK, Munir S, Bade S, Murdock HM, Santos CP, Brock LG, et al. 2017. Recurrent rhinovirus infections in a child with inherited MDA5 deficiency. J. Exp. Med 214: 1949–1972. jem.20161759. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Funabiki M, Kato H, Miyachi Y, Toki H, Motegi H, Inoue M, Minowa O, Yoshida A, Deguchi K, Sato H, et al. 2014. Autoimmune disorders associated with gain of function of the intracellular sensor MDA5. Immunity 40: 199–212. [DOI] [PubMed] [Google Scholar]
  • 9.Rice GI, del Toro Duany Y, Jenkinson EM, Forte GM, Anderson BH, Ariaudo G, Bader-Meunier B, Baildam EM, Battini R, Beresford MW, et al. 2014. Gain-of-function mutations in IFIH1 cause a spectrum of human disease phenotypes associated with upregulated type I interferon signaling. Nat. Genet 46: 503–509. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Takahasi K, Kumeta H, Tsuduki N, Narita R, Shigemoto T, Hirai R, Yoneyama M, Horiuchi M, Ogura K, Fujita T, and Inagaki F. 2009. Solution structures of cytosolic RNA sensor MDA5 and LGP2 C-terminal domains: identification of the RNA recognition loop in RIG-I-like receptors. J. Biol. Chem 284: 17465–17474. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Bruns AM, Leser GP, Lamb RA, and Horvath CM. 2014. The innate immune sensor LGP2 activates antiviral signaling by regulating MDA5-RNA interaction and filament assembly. Mol. Cell 55: 771–781. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Uchikawa E, Lethier M, Malet H, Brunel J, Gerlier D, and Cusack S. 2016. Structural analysis of dsRNA binding to anti-viral pattern recognition receptors LGP2 and MDA5. Mol. Cell 62: 586–602. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Luthra P, Sun D, Silverman RH, and He B. 2011. Activation of IFN-β expression by a viral mRNA through RNase L and MDA5. Proc. Natl. Acad. Sci. USA 108: 2118–2123. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Li LF, Yu J, Zhang Y, Yang Q, Li Y, Zhang L, Wang J, Li S, Luo Y, Sun Y, and Qiu HJ. 2017. Interferon-inducible oligoadenylate synthetase-like protein acts as an antiviral effector against classical swine fever virus via the MDA5-mediated type I interferon-signaling pathway. J. Virol 91: e01514–e01516. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Lang X, Tang T, Jin T, Ding C, Zhou R, and Jiang W. 2017. TRIM65-catalized ubiquitination is essential for MDA5-mediated antiviral innate immunity. J. Exp. Med 214: 459–473. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Hu MM, Liao CY, Yang Q, Xie XQ, and Shu HB. 2017. Innate immunity to RNA virus is regulated by temporal and reversible sumoylation of RIG-I and MDA5. J. Exp. Med 214: 973–989. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Kok KH, Lui PY, Ng MH, Siu KL, Au SW, and Jin DY. 2011. The double-stranded RNA-binding protein PACT functions as a cellular activator of RIG-I to facilitate innate antiviral response. Cell Host Microbe 9: 299–309. [DOI] [PubMed] [Google Scholar]
  • 18.Kew C, Lui PY, Chan CP, Liu X, Au SW, Mohr I, Jin DY, and Kok KH. 2013. Suppression of PACT-induced type I interferon production by herpes simplex virus 1 Us11 protein. J. Virol 87: 13141–13149. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Luthra P, Ramanan P, Mire CE, Weisend C, Tsuda Y, Yen B, Liu G, Leung DW, Geisbert TW, Ebihara H, et al. 2013. Mutual antagonism between the Ebola virus VP35 protein and the RIG-I activator PACT determines infection outcome. Cell Host Microbe 14: 74–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Siu KL, Yeung ML, Kok KH, Yuen KS, Kew C, Lui PY, Chan CP, Tse H, Woo PCY, Yuen KY, and Jin DY. 2014. Middle east respiratory syndrome coronavirus 4a protein is a double-stranded RNA-binding protein that suppresses PACT-induced activation of RIG-I and MDA5 in the innate antiviral response. J. Virol 88: 4866–4876. [DOI] [PMC free article] [PubMed] [Google Scholar] [Retracted]
  • 21.Tawaratsumida K, Phan V, Hrincius ER, High AA, Webby R, Redecke V, and Häcker H. 2014. Quantitative proteomic analysis of the influenza A virus nonstructural proteins NS1 and NS2 during natural cell infection identifies PACT as an NS1 target protein and antiviral host factor. J. Virol 88: 9038–9048. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Ho TH, Kew C, Lui PY, Chan CP, Satoh T, Akira S, Jin DY, and Kok KH. 2015. PACT- and RIG-I-dependent activation of type I interferon production by a defective interfering RNA derived from measles virus vaccine. J. Virol 90: 1557–1568. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Kok KH, Ng MH, Ching YP, and Jin DY. 2007. Human TRBP and PACT directly interact with each other and associate with dicer to facilitate the production of small interfering RNA. J. Biol. Chem 282: 17649–17657. [DOI] [PubMed] [Google Scholar]
  • 24.Ran FA, Hsu PD, Wright J, Agarwala V, Scott DA, and Zhang F. 2013. Genome engineering using the CRISPR-Cas9 system. Nat. Protoc 8: 2281–2308. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Chaudhary V, Yuen KS, Chan JF, Chan CP, Wang PH, Cai JP, Zhang S, Liang M, Kok KH, Chan CP, et al. 2017. Selective activation of interferon-γ signaling by Zika virus NS5 protein. J. Virol 91: e00163–e17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Yuen KS, Chan CP, Wong NH, Ho CH, Ho TH, Lei T, Deng W, Tsao SW, Chen H, Kok KH, and Jin DY. 2015. CRISPR/Cas9-mediated genome editing of Epstein-Barr virus in human cells. J. Gen. Virol 96: 626–636. [DOI] [PubMed] [Google Scholar]
  • 27.Lui PY, Wong LY, Fung CL, Siu KL, Yeung ML, Yuen KS, Chan CP, Woo PCY, Yuen KY, and Jin DY. 2016. Middle east respiratory syndrome coronavirus M protein suppresses type I interferon expression through the inhibition of TBK1-dependent phosphorylation of IRF3. Emerg. Microbes Infect 5: e39. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Iwamura T, Yoneyama M, Yamaguchi K, Suhara W, Mori W, Shiota K, Okabe Y, Namiki H, and Fujita T. 2001. Induction of IRF-3/-7 kinase and NF-kappaB in response to double-stranded RNA and virus infection: common and unique pathways. Genes Cells 6: 375–388. [DOI] [PubMed] [Google Scholar]
  • 29.Siu KL, Kok KH, Ng MH, Poon VK, Yuen KY, Zheng BJ, and Jin DY. 2009. Severe acute respiratory syndrome coronavirus M protein inhibits type I interferon production by impeding the formation of TRAF3.TANK. TBK1/IKKepsilon complex. J. Biol. Chem 284: 16202–16209. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Ye J, and Maniatis T. 2011. Negative regulation of interferon-β gene expression during acute and persistent virus infections. PLoS One 6: e20681. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Kato H, Takeuchi O, Mikamo-Satoh E, Hirai R, Kawai T, Matsushita K, Hiiragi A, Dermody TS, Fujita T, and Akira S. 2008. Length-dependent recognition of double-stranded ribonucleic acids by retinoic acid-inducible gene-I and melanoma differentiation-associated gene 5. J. Exp. Med 205: 1601–1610. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Gupta V, Huang X, and Patel RC. 2003. The carboxy-terminal, M3 motifs of PACT and TRBP have opposite effects on PKR activity. Virology 315: 283–291. [DOI] [PubMed] [Google Scholar]
  • 33.Berke IC, Yu X, Modis Y, and Egelman EH. 2012. MDA5 assembles into a polar helical filament on dsRNA. Proc. Natl. Acad. Sci. USA 109: 18437–18441. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Peisley A, Lin C, Wu B, Orme-Johnson M, Liu M, Walz T, and Hur S. 2011. Cooperative assembly and dynamic disassembly of MDA5 filaments for viral dsRNA recognition. Proc. Natl. Acad. Sci. USA 108: 21010–21015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Wies E, Wang MK, Maharaj NP, Chen K, Zhou S, Finberg RW, and Gack MU. 2013. Dephosphorylation of the RNA sensors RIG-I and MDA5 by the phosphatase PP1 is essential for innate immune signaling. Immunity 38: 437–449. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Iwamura T, Yoneyama M, Koizumi N, Okabe Y, Namiki H, Samuel CE, and Fujita T. 2001. PACT, a double-stranded RNA binding protein acts as a positive regulator for type I interferon gene induced by Newcastle disease virus. Biochem. Biophys. Res. Commun 282: 515–523. [DOI] [PubMed] [Google Scholar]
  • 37.Rowe TM, Rizzi M, Hirose K, Peters GA, and Sen GC. 2006. A role of the double-stranded RNA-binding protein PACT in mouse ear development and hearing. Proc. Natl. Acad. Sci. USA 103: 5823–5828. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Marques JT, White CL, Peters GA, Williams BR, and Sen GC. 2008. The role of PACT in mediating gene induction, PKR activation, and apoptosis in response to diverse stimuli. J. Interferon Cytokine Res. 28: 469–476. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Patel RC, and Sen GC. 1998. PACT, a protein activator of the interferon-induced protein kinase, PKR. EMBO J. 17: 4379–4390. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Triantafilou K, Vakakis E, Kar S, Richer E, Evans GL, and Triantafilou M. 2012. Visualisation of direct interaction of MDA5 and the dsRNA replicative intermediate form of positive strand RNA viruses. J. Cell Sci 125: 4761–4769. [DOI] [PubMed] [Google Scholar]
  • 41.Onomoto K, Yoneyama M, Fung G, Kato H, and Fujita T. 2014. Antiviral innate immunity and stress granule responses. Trends Immunol. 35: 420–428. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Langereis MA, Feng Q, and van Kuppeveld FJ. 2013. MDA5 localizes to stress granules, but this localization is not required for the induction of type I interferon. J. Virol 87: 6314–6325. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Pare JM, Tahbaz N, López-Orozco J, LaPointe P, Lasko P, and Hobman TC. 2009. Hsp90 regulates the function of argonaute 2 and its recruitment to stress granules and P-bodies. Mol. Biol. Cell 20: 3273–3284. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Pham AM, Santa Maria FG, Lahiri T, Friedman E, Marié IJ, and Levy DE. 2016. PKR transduces MDA5-dependent signals for type I IFN induction. PLoS Pathog. 12: e1005489. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Sanchez David RY, Combredet C, Sismeiro O, Dillies MA, Jagla B, Coppée JY, Mura M, Guerbois Galla M, Despres P, Tangy F, and Komarova AV. 2016. Comparative analysis of viral RNA signatures on different RIG-I-like receptors. eLife 5: e11275. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Lee HY, Zhou K, Smith AM, Noland CL, and Doudna JA. 2013. Differential roles of human Dicer-binding proteins TRBP and PACT in small RNA processing. Nucleic Acids Res. 41: 6568–6576. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Heyam A, Lagos D, and Plevin M. 2015. Dissecting the roles of TRBP and PACT in double-stranded RNA recognition and processing of noncoding RNAs. Wiley Interdiscip. Rev. RNA 6: 271–289. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Liddicoat BJ, Piskol R, Chalk AM, Ramaswami G, Higuchi M, Hartner JC, Li JB, Seeburg PH, and Walkley CR. 2015. RNA editing by ADAR1 prevents MDA5 sensing of endogenous dsRNA as nonself. Science 349: 1115–1120. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES