Fig 1: ELAVL1 is a potent positive regulator of cGAS and RIG-I signaling.(A) ISRE reporter activation in THP-1 WT, ELAVL1 KO (#1), ELAVL1 KO (#1) expressing ELAVL1-FLAG or GFP, RIG-I KO (clones #1–2), MAVS KO (clones #1–2), and cGAS KO cells 20 h after stimulation with 5′ppp-dsRNA (0.1 µg/ml), pDNA (0.1 µg/ml), G3-YSD (0.5 µg/ml), Pam3CSK4 (0.5 µg/ml), or IFNα (5000 U/ml) (mean ± SD; stimuli from left to right: n = 3, 3, 3, 3, 3, 3 independent experiments). ctrl, non-stimulated. (B) ISRE reporter activation in the cells depicted in (A) 20 h after challenge with 2'3′-cGAMP (10 µg/ml) (mean ± SD; stimuli from left to right: n = 3, 3 independent experiments). ctrl, non-stimulated. (C) NF-κB reporter activation in the cells depicted in (A) 20 h after challenge with the indicated stimuli (mean ± SD; stimuli from left to right: n = 3, 3, 3, 3, 3, 3, 3 independent experiments). (D) MTT assay of THP-1 WT, ELAVL1 KO (#1), and ELAVL1 KO (#1) expressing ELAVL1-FLAG or GFP 20 h after stimulation with 5′ppp-dsRNA (0.1 µg/ml), pDNA (0.1 µg/ml), G3-YSD (0.5 µg/ml), 2'3′-cGAMP (10 µg/ml), Pam3CSK4 (0.5 µg/ml), or IFNα (1000 U/ml). ctrl, non-stimulated (stimuli from left to right: n = 11, 6, 6, 6, 3, 3, 8 independent experiments). (E) CXCL10 ELISA with supernatants of the cells depicted in (D) collected 20 h after stimulation with 5′ppp-dsRNA (0.1 µg/ml), pDNA (0.1 µg/ml), G3-YSD (0.5 µg/ml), Pam3CSK4 (0.5 µg/ml), or IFNα (1000 U/ml). ctrl, non-stimulated (stimuli from left to right: n = 7, 7, 7, 7, 7, 7 independent experiments). (F) Quantification of pIRF3-Ser396 levels shown in Fig. 4J (mean ± SD; stimuli from left to right: n = 2, 2, 2 independent experiments). (G) Quantification of pTBK1-Ser172 levels shown in Fig. 4J (mean ± SD; stimuli from left to right: n = 2, 2, 2 independent experiments). (H) Quantification of pSTING-Ser366 levels shown in Fig. 4J (mean ± SD; stimuli from left to right: n = 2, 2, 2 independent experiments). (I) Quantification of pTBK1-Ser172 levels shown in Fig. 4K (mean ± SD; stimuli from left to right: n = 3, 3, 3 independent experiments). For (D, E): mean ± SD, two-way ANOVA, Dunnett’s multiple comparisons test. ns, P value > 0.05.
Fig 2: MS-444 reduces constitutive secretion of inflammatory cytokines in cells from patients with dermatomyositis, AGS, and SAVI/chilblain lupus.(A) THP-1 cells were pre-treated with MS-444/DMSO for 2 h and then stimulated with 5′ppp-dsRNA (0.1 µg/ml) or left non-stimulated (ctrl) in MS-444/DMSO-containing medium. After 16 h, ISRE reporter activation was determined (from left to right: n = 3, 3, 3, 3, 3, 3, 3, 3 independent experiments). (B) THP-1 cells were pre-treated with MS-444/DMSO for 2 h and then stimulated with G3-YSD (0.5 µg/ml) or left non-stimulated (ctrl) in MS-444/DMSO-containing medium. After 16 h, ISRE reporter activation was determined (from left to right: n = 3, 3, 3, 3, 3, 3, 3, 3 independent experiments). (C) Immunoblot analysis of THP-1 cells pre-treated with 20 µM MS-444 or DMSO for 16 h and then stimulated for 1 h with G3-YSD (0.5 µg/ml), 5′ppp-dsRNA (0.1 µg/ml), Pam3CSK4 (0.5 µg/ml), or left non-stimulated (ctrl). One representative experiment of four independent experiments is shown. (D) Primary human fibroblasts were pre-treated with MS-444/DMSO for 2 h and then stimulated with 5′ppp-dsRNA (0.1 µg/ml) or left non-stimulated (ctrl) in MS-444/DMSO-containing medium. After 16 h, CXCL10 secretion was analyzed. ctrl, non-stimulated (from left to right: n = 4, 4, 4, 4, 4, 4, 4, 4 independent experiments). (E) Primary human fibroblasts were pre-treated with MS-444/DMSO for 2 h and then stimulated with pDNA (1 µg/ml) or left non-stimulated (ctrl) in MS-444/DMSO-containing medium. After 16 h, CXCL10 secretion was analyzed. ctrl, non-stimulated (from left to right: n = 4, 4, 4, 4, 4, 4, 4, 4 independent experiments). (F) Primary human fibroblasts from a healthy donor (WT) or from patients with dermatomyositis, AGS, or SAVI/chilblain lupus were treated for 16 h with 20 µM MS-444 or DMSO and then CXCL10 secretion was analyzed (from left to right: n = 5, 5, 5, 5, 5, 5, 5, 5 independent experiments). (G) TNFα ELISA with supernatants of the cells depicted in (F) (from left to right: n = 5, 5, 5, 5, 5, 5, 5, 5 independent experiments). (H) Proposed model of how hnRNPM and ELAVL1 regulate signaling downstream of cGAS and RIG-I. For (A, B, D, E): mean ± SD, two-way ANOVA, Dunnett’s multiple comparisons test. For (F, G): mean ± SD, two-way ANOVA, Šídák’s multiple comparisons test. ns, P value > 0.05. Source data are available online for this figure.
Fig 3: ELAVL1 interacts with hnRNPM to regulate type I IFN induction downstream of both cGAS and RIG-I.(A) Left panel: Expression of ELAVL1 mRNA in THP-1 cells expressing control shRNAs (shC001, shC002) or ELAVL1-specific shRNAs (shELAVL1.1, shELAVL1.2) (mean ± SD, one-way ANOVA, Dunnett’s multiple comparisons test; from left to right: n = 3, 3, 3, 3 independent experiments), right panel: ISRE reporter activation 20 h after stimulation with 5′ppp-dsRNA (0.1 µg/ml), pDNA (0.1 µg/ml), or G3-YSD (0.5 µg/ml) (mean ± SD, two-way ANOVA, Dunnett’s multiple comparisons test; stimuli from left to right: n = 3, 3, 3, 3 independent experiments). ctrl, non-stimulated. (B) Expression of HNRNPM and ELAVL1 mRNA in primary human fibroblasts stably expressing control shRNA (shC001) or shRNAs against HNRNPM or ELAVL1 (shRNA.1, shRNA.2) (mean ± SD; from left to right: n = 2, 2, 2, 2, 2, 2 independent experiments). (C) CXCL10 ELISA with supernatants of the cells depicted in (B) 20 h after stimulation with 5′ppp-dsRNA (0.1 µg/ml). ctrl, non-stimulated (mean ± SD, two-way ANOVA, Dunnett’s multiple comparisons test; stimuli from left to right: n = 3, 3 independent experiments). (D) CXCL10 ELISA with supernatants of the cells depicted in (B) 20 h after stimulation with pDNA (1 µg/ml). ctrl, non-stimulated (stimuli from left to right: n = 2, 2 independent experiments). (E) hnRNPM-GFP, ELAVL1-GFP, and GFP were immunoprecipitated from lysates of non-stimulated THP-1 cells and bound proteins were identified by LC-MS/MS. Proteins detected in hnRNPM and ELAVL1 precipitates were statistically compared to GFP as control using a two-sided Welch’s t test (S0 = 0.1, permutation-based FDR < 0.05). The Venn diagram shows the absolute and relative proportions of unique and shared interactors of hnRNPM and ELAVL1. (F) Shared interaction network of hnRNPM and ELAVL1. Blue circles indicate proteins annotated with the gene ontology (GO) term immune system process (GO.0002376, FDR = 0.0034). The statistical tests used are described in detail in the methods. ns, P value > 0.05. Source data are available online for this figure.
Fig 4: hnRNPM and ELAVL1 interact with TBK1, IKKε, IKKβ, and NF-κB p65.(A) GFP-specific beads were used to immunoprecipitate hnRNPM-GFP and GFP from lysates of non-stimulated or pDNA-stimulated (0.1 µg/ml, 3 h) THP-1 cells. If indicated, IPs were treated with RNase A (100 µg/ml, 1.5 h). The IPs and 5.0% of the cleared cellular lysate used for IP (input) were analyzed by immunoblotting with the indicated antibodies. IPs and input controls of hnRNPM-GFP and GFP were analyzed on the same membrane, with empty lanes removed. (B) FLAG tag-specific beads were incubated with lysates of WT or ELAVL1-FLAG-expressing HEK293FT cells. If indicated, the IPs were treated with RNase A (100 µg/ml, 1.5 h). IPs and input controls were analyzed by immunoblotting with the indicated antibodies. (C) Differentiated THP-1 cells stably expressing hnRNPM-GFP or GFP were transfected with Alexa Fluor 647-labelled G3-YSD (0.5 µg/ml, 4 h) or left non-stimulated (ctrl). Anti-GFP and anti-pTBK1-Ser172 monoclonal antibodies were used to analyze the interactions between hnRNPM-GFP/GFP and pTBK1-Ser172 in cellulo by PLA. Z stack images were recorded on a confocal microscope. Image sequences of hnRNPM-GFP-expressing cells are shown as maximum intensity projections (left panel). The PLA signals per cell in different focal planes were counted and the counted number of puncta per cell was defined as the PLA score (right panel). hnRNPM-GFP:pTBK1, interactions between hnRNPM-GFP and pTBK1-Ser172; GFP:pTBK1, interactions between GFP and pTBK1-Ser172. Blue, Hoechst 33342; yellow, G3-YSD-AF647; cyan, PLA signal. (D) The PLA was performed as described in (C) with differentiated THP-1 cells stably expressing ELAVL1-GFP or GFP. (E) The PLA was performed as described in (C) using anti-GFP and anti-ELAVL1 monoclonal antibodies to analyze the interactions between hnRNPM-GFP/GFP and ELAVL1. For (A, B): One representative experiment of two independent experiments is shown. For (C–E): mean, two-way ANOVA, Tukey’s multiple comparisons test; n is the number of analyzed cells. ns, P value > 0.05. Source data are available online for this figure.
Fig 5: ELAVL1 promotes the cGAS- and RIG-I-mediated induction of type I IFNs and activation of NF-κB.(A) Immunoblot analysis of THP-1 WT and ELAVL1 KO cells using an ELAVL1-specific antibody. (B) ISRE reporter activation in the cells depicted in (A) 20 h after stimulation with 5′ppp-dsRNA (0.1 µg/ml), pDNA (0.1 µg/ml), G3-YSD (0.5 µg/ml), or IFNα (1000 U/ml). ctrl, non-stimulated (left panel: stimuli from left to right, n = 4, 4, 4, 4 independent experiments; right panel: stimuli from left to right, n = 3, 3 independent experiments). (C) ISRE reporter activation in THP-1 WT, ELAVL1 KO (#1), and ELAVL1 KO (#1) expressing ELAVL1-FLAG or GFP 20 h after stimulation with 5′ppp-dsRNA (0.1 µg/ml), pDNA (0.1 µg/ml), G3-YSD (0.5 µg/ml), or Pam3CSK4 (0.5 µg/ml). ctrl, non-stimulated (stimuli from left to right: n = 4, 4, 4, 4, 4 independent experiments). (D) ISRE reporter activation in the cells depicted in (C) 20 h after stimulation with 2'3′-cGAMP (10 µg/ml). ctrl, non-stimulated (stimuli from left to right: n = 3, 3 independent experiments). (E) ISRE reporter activation in the cells depicted in (C) 20 h after stimulation with IFNα (1000 U/ml). ctrl, non-stimulated (stimuli from left to right: n = 3, 3 independent experiments). (F) NF-κB reporter activation in the cells depicted in (C) 20 h after stimulation with 5′ppp-dsRNA (0.1 µg/ml), pDNA (0.1 µg/ml), G3-YSD (0.5 µg/ml), Pam3CSK4 (0.5 µg/ml), 2'3′-cGAMP (10 µg/ml), or IFNα (1000 U/ml). ctrl, non-stimulated (stimuli from left to right: n = 4, 4, 4, 4, 4, 4, 4 independent experiments). (G) Expression of IFNB1 mRNA in the cells depicted in (C) 6 h after stimulation with 5′ppp-dsRNA (0.1 µg/ml), pDNA (0.1 µg/ml), G3-YSD (0.5 µg/ml), or IFNα (1000 U/ml). ctrl, non-stimulated (stimuli from left to right: n = 6, 6, 3, 6, 6 independent experiments). (H) Expression of CXCL10 mRNA in the cells depicted in (C) 6 h after stimulation with 5′ppp-dsRNA (0.1 µg/ml), pDNA (0.1 µg/ml), G3-YSD (0.5 µg/ml), or IFNα (1000 U/ml). ctrl, non-stimulated (stimuli from left to right: n = 6, 6, 3, 6, 6 independent experiments). (I) Expression of IFIT1 mRNA in the cells depicted in (C) 6 h after stimulation with 5′ppp-dsRNA (0.1 µg/ml), pDNA (0.1 µg/ml), G3-YSD (0.5 µg/ml), or IFNα (1000 U/ml). ctrl, non-stimulated (stimuli from left to right: n = 6, 6, 3, 6, 6 independent experiments). (J) Immunoblot analysis of the cells depicted in (C) 3 h after stimulation with G3-YSD (0.5 µg/ml) or 5′ppp-dsRNA (0.1 µg/ml). ctrl, non-stimulated. (K) Immunoblot analysis of THP-1 WT and ELAVL1 KO (#1) cells after stimulation with Pam3CSK4 (0.5 µg/ml). ctrl, non-stimulated. (L) ISRE reporter activation in the cells depicted in (C) 24 h after infection with HSV-1 (MOI 5). ctrl, non-stimulated (stimuli from left to right: n = 3, 3 independent experiments). (M) ISRE reporter activation in the cells depicted in (C) 24 h after infection with SeV (MOI 1). ctrl, non-stimulated (stimuli from left to right: n = 3, 3 independent experiments). For (B–I, L, M): mean ± SD, two-way ANOVA, Dunnett’s multiple comparisons test. For (J, K): One representative experiment of at least two independent experiments is shown. ns, P value > 0.05. Source data are available online for this figure.
Supplier Page from Sino Biological, Inc. for Human ELAVL1 Gene ORF cDNA clone in cloning vector