Inborn errors of OAS-RNase L in SARS-CoV-2-related multisystem inflammatory syndrome in children

Abstract

Multisystem inflammatory syndrome in children (MIS-C) is a rare and severe condition that follows benign COVID-19. We report autosomal recessive deficiencies of OAS1, OAS2, or RNASEL in five unrelated children with MIS-C. The cytosolic double-stranded RNA (dsRNA)-sensing OAS1 and OAS2 generate 2'-5'-linked oligoadenylates (2-5A) that activate the single-stranded RNA-degrading ribonuclease L (RNase L). Monocytic cell lines and primary myeloid cells with OAS1, OAS2, or RNase L deficiencies produce excessive amounts of inflammatory cytokines upon dsRNA or severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) stimulation. Exogenous 2-5A suppresses cytokine production in OAS1-deficient but not RNase L-deficient cells. Cytokine production in RNase L-deficient cells is impaired by MDA5 or RIG-I deficiency and abolished by mitochondrial antiviral-signaling protein (MAVS) deficiency. Recessive OAS-RNase L deficiencies in these patients unleash the production of SARS-CoV-2-triggered, MAVS-mediated inflammatory cytokines by mononuclear phagocytes, thereby underlying MIS-C.

OAS–RNase L deficiency in MIS-C.

dsRNAs from SARS-CoV-2 or SARS-CoV-2–permissive cells engulfed by mononuclear phagocytes simultaneously activate the RIG-I/MDA5–MAVS pathway, inducing inflammatory cytokine production, and the OAS–RNase L pathway, exerting posttranscriptional control over inflammatory cytokine production. OAS–RNase L deficiency results in excessive inflammatory cytokine production by myeloid cells, triggering MIS-C, including lymphoid cell activation and multiple tissue lesions. NK, natural killer; IRF3, interferon regulatory factor 3; NF-κB, nuclear factor κB.

Fig. 1.. Biallelic OAS1, OAS2, and RNASEL variants in patients with MIS-C.

(A) Family pedigrees with allele segregation. Mutant, “MT” in red; wild-type, “WT” in black. (B to D) Functional assays for WT and mutant OAS1 (B), OAS2 (C), and RNase L (D). Variants for which homozygotes or compound heterozygotes were present in our MIS-C cohort were tested. (Upper panels) RNase L–mediated cleavage of rRNA in a cell-free system based on transfected HeLa M cells. (Lower panels) Immunoblots of the indicated proteins. EV, empty vector. Arrows indicate degraded rRNA species. OAS2 variants (C) were tested under two different sets of conditions (see methods). The results shown in (B) to (D) are representative of three independent experiments. (E and F) FRET assay of 2-5A synthesized in response to poly(I:C) stimulation by WT and MT OAS1 (E) or OAS2 (F). RFU, relative fluorescence units. The data shown are the means ± SEM of six biological replicates. Statistical analysis was performed as described in the methods. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. (G) Concentrations of various cytokines in plasma samples from OAS–RNase L–deficient patients during MIS-C (P1, P2, and P5); comparison with those of healthy controls (HC), pediatric (pC-19) or adult COVID-19 pneumonia (aC-19) patients, typical Kawasaki disease patients (KD), other MIS-C patients with no known genetic etiology (MIS-C), and patients with toxic shock syndrome (TSS). (H) PCA of gene expression quantified by whole-blood bulk RNA-seq for P1 and P2 during MIS-C relative to pediatric controls (pCtrls), previously published MIS-C patients, and a pediatric patient with mild COVID-19 (pC-19). (I) Relative levels of TRBV 11-2 (encoding Vβ21.3) RNA in blood samples from P1, P2, and P5 during MIS-C, relative to other MIS-C patients, adults with mild or severe COVID-19 (mild aC-19, sev aC-19), and healthy controls. (J to L) CADD-MAF graph of OAS1 (J), OAS2 (K), and RNASEL (L) variants for which homozygotes are reported in gnomAD and/or found in our MIS-C cohort. Single-letter abbreviations for the amino acid residues are as follows: A, Ala; C, Cys; D, Asp; E, Glu; F, Phe; G, Gly; H, His; I, Ile; K, Lys; L, Leu; M, Met; N, Asn; P, Pro; Q, Gln; R, Arg; S, Ser; T, Thr; V, Val; W, Trp; and Y, Tyr.

Fig. 2.. Expression pattern of the OAS–RNase L pathway genes and their role in SARS-CoV-2 restriction.

(A and B) Relative OAS1, OAS2, OAS3, and RNASEL mRNA levels measured by bulk RNA-seq (A) or RT-qPCR (B), in various cell types. hPSC, human pluripotent stem cell; ClassMonocytes, classical monocytes; NClassMonocytes, nonclassical monocytes; MDM, monocyte-derived macrophages; MDDC, monocyte-derived dendritic cells; Log2RC, log₂ read count. (C and D) Immunoblot of the indicated proteins (C) and immunofluorescence (IF) of SARS-CoV-2 nucleocapsid (N) protein (D) in A549+ACE2/TMPRSS2 cells with and without knockout (KO) of OAS1, OAS2, or RNase L. IF analysis for N protein was performed 24 hours after infection with various dilutions of SARS-CoV-2. Dilution factors of ¹/₄, ¹/₂, and 1 correspond to MOI values of 0.0002, 0.0005, and 0.001, respectively. GAPDH, glyceraldehyde-3-phosphate dehydrogenase; NI, noninfected. (E and F) Immunoblot of the indicated proteins (E) and IF analysis for the SARS-CoV-2 N protein (F) in SV40-fibroblasts+ACE2 from healthy controls (Ctrl1 and Ctrl2), patients with OAS-RNASEL mutations (P1, P3, P4, and P5), and a previously reported patient with complete IFNAR1 deficiency (IFNAR1^−/−). IF analysis for N protein was performed at various time points after infection at a MOI of 0.08. (G and H) Immunoblot of the indicated proteins (G) and IF analysis for the SARS-CoV-2 N protein (H) in THP-1 cells with and without KO of OAS1, OAS2, or RNase L. IF analyses for N protein were performed in PMA-primed THP-1 cells 24 hours after infection with various dilutions of SARS-CoV-2. Dilution factors of ¹/₄, ¹/₂, and 1 correspond to MOI values of 0.012, 0.025, and 0.05, respectively. WT A549+ACE2/TMPRSS2 cells were included as a positive control for SARS-CoV-2 infection. The data points are means ± SEM from three [(D) and (F)] or means from two [(B) and (H)] independent experiments with three to six technical replicates per experiment. Statistical analyses were performed as described in the methods. ns, not significant; *P < 0.05, **P < 0.01.

Fig. 3.. Exaggerated inflammatory responses of OAS–RNase L-deficient THP-1 cells.

(A) Concentrations of various cytokines in the supernatant of OAS1 KO, OAS2 KO, RNase L KO, or parental THP-1 cells (upper panels) or PMA-primed THP-1 cells (lower panels) treated as indicated for 24 hours. (B) IFN-λ1 and IL-6 concentrations in the supernatant of RNase L KO THP-1 cells transduced with the WT or P5’s variant RNASEL cDNA, or empty vector (EV), and treated as indicated for 24 hours. On the right, RNase L protein levels, as assessed by immunoblotting. NT, not transfected. (C) IFN-λ1 and IL-6 concentrations in the supernatant of parental, RIG-I KO, MDA5 KO, or MAVS KO THP-1 cells with or without (WT sh-ctrl) RNase L knockdown (KDn), treated as indicated for 24 hours. (D) IFN-λ1 and IFN-β concentrations in the supernatant of parental or RNase L KO THP-1 cells, treated as indicated for 24 hours. (E) Immunoblot of phosphorylated P65 and IRF3 in parental and RNase L KO THP-1 cells treated as indicated. The results shown are representative of two independent experiments. (F) IFN-λ1 and IL-6 concentrations in the supernatant of parental, OAS1 KO, or RNase L KO THP-1 cells treated as indicated for 24 hours. (G) IFN-λ1 and IL-6 concentrations in WT THP-1 cells treated as indicated for 24 hours. In (A) to (D), (F), and (G), the data points are means ± SEM from three to five independent experiments with one to two technical replicates per experiment. Statistical analysis was performed as described in the methods. ns, not significant; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. NS, nonstimulated; Lipo, lipofectamine only; poly(I:C), extracellularly added poly(I:C); poly(I:C)+Lipo, intracellular poly(I:C) in the presence of lipofectamine; 2-5A+Lipo, intracellular 2-5A in the presence of lipofectamine; 2-5A+poly(I:C)+Lipo, intracellular poly(I:C) in addition to intracellular 2-5A.

Fig. 4.. Exaggerated inflammatory responses to SARS-CoV-2 of OAS–RNase L–deficient THP-1 cells.

(A and B) PCA of RNA-seq–quantified gene expression for OAS1 KO, OAS2 KO, RNase L KO, and parental (WT) THP-1 cells left nonstimulated (NS), treated as indicated for 2 or 8 hours (A), or stimulated with SARS-CoV-2 (SCV2) at a MOI of 0.01 for 8 hours (B). (C and D) Differential expression analysis (DEA) and gene set enrichment analysis (GSEA) for genes induced by 8 hours of intracellular poly(I:C) stimulation (C) or by 8 hours of SCV2 stimulation (D). The OAS1 KO, OAS2 KO, and RNase L KO THP-1 cells were compared with parental (WT) THP-1 cells. Volcano plots show immune system–related pathways. NES, normalized enrichment score. Heatmaps show gene expression for the “IFN-γ response” (C) or “inflammatory response” (D) Hallmark gene sets. (E) IL-6 and CXCL10 concentrations in the supernatant of parental or RNase L KO THP-1 cells treated as indicated for 24 hours. The data points are means ± SEM from three independent experiments with three technical replicates per experiment. Statistical analysis was performed as described in the methods. *P < 0.05. (F) PCA of RNA-seq–quantified gene expression, for RNase L KO and parental THP-1 cells cocultured with Vero cells with or without SCV2 infection for 24 hours (left) or transfected for 8 hours with RNA from Vero cells with or without SCV2-infection (right). (G and H) DEA and GSEA for genes induced in RNase L KO THP-1 cells, compared with parental THP-1 cells after 24 hours of coculture with SCV2-infected or mock-infected Vero cells (G), or after 8 hours of transfection with RNA from SCV2-infected or mock-infected Vero cells (H). Volcano plots show immune system–related pathways. Heatmaps show gene expression for the indicated Hallmark gene sets. Heatmaps represent Z-score–scaled log₂ read counts per million. NS, nonstimulated; Lipo, lipofectamine; SCV2, SARS-CoV-2.

Fig. 5.. Exaggerated myeloid cell activation in response to SARS-CoV-2 underlies MIS-C.

(A) Concentrations of cytokines in the supernatant of PBMCs from OAS–RNase L–deficient patients (grouped in the pink violin zone) and three healthy pediatric and two healthy adult controls (Ctrls; gray violin zone). The data points are means of biological duplicates. (B) Fold-increase in the concentrations of cytokines in the supernatant of MDDCs with KDn of OAS1, OAS2, or RNase L, or transduced with control shRNA (WT sh-ctrl). The fold-change is expressed relative to the values for poly(I:C)+lipo-stimulated WT sh-ctrl cells. Data shown are means ± SEM from three independent experiments, with one to two technical replicates per experiment. For (A) and (B), statistical analysis was performed as described in the methods. NS, nonstimulated. ns, not significant; *P < 0.05, **P < 0.01, ***P < 0.001. (C to E) scRNA-seq of PBMCs from OAS–RNase L–deficient patients (OAS–RNase L-MT) or healthy controls after 6 hours of incubation with SARS-CoV-2 (SCV2) or mock infection (NS). (C) Uniform manifold approximation and projection (UMAP) of single PBMC transcriptomes. (D) Cell type–specific transcriptional responses. Genes passing the FDR < 0.01 and |log2FC| > 0.5 thresholds are shown. (E) GSEA of SCV2-induced genes across immune-related Hallmark gene sets. PBMCs from three patients with type I IFN pathway deficiency are controls for defective type I IFN responses. Gray zone highlights the expected enrichment scores under the null hypothesis (95% CI calculated over 100 randomized genes). (F to I) scRNA-seq of PBMCs from P5 and from healthy controls. A published dataset for pediatric patients with acute SARS-CoV-2 infection (pC-19) and MIS-C was also integrated. (F) UMAP of clustering analysis. (G) Pseudobulk differential expression analysis with GSEA. P5 (convalescent phase) was compared with local pediatric controls (pCtrls). Immune-related pathways are shown. [(H) and (I)] Intercellular communication analysis with CellChat. (H) Incoming signal strength and (I) the number of interactions for representative cell subsets.