key: cord-1013157-g7qh4bur
authors: de Cevins, Camille; Luka, Marine; Smith, Nikaïa; Meynier, Sonia; Magérus, Aude; Carbone, Francesco; García-Paredes, Víctor; Barnabei, Laura; Batignes, Maxime; Boullé, Alexandre; Stolzenberg, Marie-Claude; Pérot, Brieuc P.; Charbit, Bruno; Fali, Tinhinane; Pirabarakan, Vithura; Sorin, Boris; Riller, Quentin; Abdessalem, Ghaith; Beretta, Maxime; Grzelak, Ludivine; Goncalves, Pedro; Di Santo, James P.; Mouquet, Hugo; Schwartz, Olivier; Zarhrate, Mohammed; Parisot, Mélanie; Bole-Feysot, Christine; Masson, Cécile; Cagnard, Nicolas; Corneau, Aurélien; Bruneau, Camille; Zhang, Shen-Ying; Casanova, Jean-Laurent; Meunier, Brigitte Bader; Haroche, Julien; Melki, Isabelle; Lorrot, Mathie; Oualha, Mehdi; Moulin, Florence; Bonnet, Damien; Belhadjer, Zahra; Leruez, Marianne; Allali, Slimane; Leguen, Christèle Gras; de Pontual, Loïc; Fischer, Alain; Duffy, Darragh; Laucat, Fredéric Rieux-; Toubiana, Julie; Ménager, Mickaël M.
title: A monocyte/dendritic cell molecular signature of SARS-CoV2-related multisystem inflammatory syndrome in children (MIS-C) with severe myocarditis
date: 2021-02-23
journal: bioRxiv
DOI: 10.1101/2021.02.23.432486
sha: c342c347dadbbef0b4077e47b765012768e7c117
doc_id: 1013157
cord_uid: g7qh4bur

SARS-CoV-2 infection in children is generally milder than in adults, yet a proportion of cases result in hyperinflammatory conditions often including myocarditis. To better understand these cases, we applied a multi-parametric approach to the study of blood cells of 56 children hospitalized with suspicion of SARS-CoV-2 infection. The most severe forms of MIS-C (multisystem inflammatory syndrome in children related to SARS-CoV-2), that resulted in myocarditis, were characterized by elevated levels of pro-angiogenesis cytokines and several chemokines. Single-cell transcriptomic analyses identified a unique monocyte/dendritic cell gene signature that correlated with the occurrence of severe myocarditis, characterized by sustained NF-κB activity, TNF-α signaling, associated with decreased gene expression of NF-κB inhibitors. We also found a weak response to type-I and type-II interferons, hyperinflammation and response to oxidative stress related to increased HIF-1α and VEGF signaling. These results provide potential for a better understanding of disease pathophysiology.

5 associated with a positive RT-PCR test for either Mycoplasma pneumoniae or 137 rhinovirus/enterovirus, and negative RT-PCR for SARS- Forty-three children displayed features of postacute hyperinflammatory illness ( Figure S1 , 139 Table S1 ). SARS-CoV-2 infection status of all samples was confirmed by specific antibody 140 determination (both IgG and IgA) in the plasma, using ELISA and flow cytometry-based had severe myocarditis (i.e. with elevated high-sensitivity cardiac troponin I and/or regional 147 wall motion abnormalities on echocardiography, and clinical signs of circulatory failure 148 requiring intensive care support; MIS-C_MYO (CoV2 + )). Thirteen tested negative for SARS-149

CoV-2 and fulfilled clinical criteria for complete (n=6) or incomplete (n=7) Kawasaki disease 150 (KD), and were therefore considered to have KD-like illness (KD (CoV2 -) group) ( Figure S1 , 151 Table S1 ). Clinical and biological characteristics, at time of disease activity and before 152 treatment, or within 24 hours of treatment onset, are presented in Table S1 . MIS-C cases had 153 low lymphocyte counts and those with severe myocarditis had in addition abnormally increased 154 neutrophil counts as compared to other groups, along with high levels of CRP, PCT, serum 155 alanine transaminases (ALT) and ferritin (Table S1). All cases responded favorably to 156 intravenous immunoglobulin injections (IVIG), some (N=12, Table S1) in combination with 157 glucocorticosteroids, received before sampling. Multi-parametric analyses were performed at 158 a median fever persistence of 9-10 days (Figures 1A, B) . 159 160 6 161 postacute hyperinflammatory conditions 162 We investigated plasma cytokine and chemokine levels in all patients, by Luminex and Simoa 163 assays. Hierarchical clustering analysis and stratification by patient groups revealed overall 164 elevated levels of immune and inflammatory markers, with 40/46 measured proteins 165 significantly elevated (q<0.05) as compared to healthy controls (Figure 2A ; global heat map). 166

Twelve cytokines were found to be elevated in all groups of patients as compared to healthy 167 controls ( Figure S2B ). High IL-8 and CXCL1 ( Figure S2C ) were more specific to children Figure 2D ) (Holbrook et al., 2019; 181 Varfolomeev and Ashkenazi, 2004 ). An increased level of CCL19, CCL20, CCL3 (cell 182 migration and chemotaxis) and IL-1 agonist/antagonist (IL-1, IL-1RA) were also observed, 183 as well as increased soluble PD-L1 ( Figure S2G ). Another noticeably elevated cytokine was 184 CSF2, known to be involved in myeloid cell differentiation and migration ( Figure 2D ).

Altogether, high inflammatory cytokine levels were detected in both acute infection and 186 postacute inflammatory cases. The strongest inflammatory profile was observed in MIS-C with 187 severe myocarditis (MIS-C_MYO (CoV2 + )) and remarkably, a similar profile was observed 188 when comparing MIS-C without myocarditis (MIS-C (CoV2 + )) and KD-like illness unrelated 189 to SARS-CoV-2 (KD (CoV2 -)). Of note, the inflammatory profile was much reduced in 190 intensity in MIS-C cases with myocarditis under combined glucocorticosteroid and IVIG 191 treatment, as compared to IVIG alone ( Figure S2H) . (Figures 6 A, B, S6 B, D) . However, both groups of MIS-C patients 263 showed elevated plasma IFN-2 and IFN proteins (Figures 2A, B) . Gene expression 264 downregulation in monocytes and DCs of MIS-C patients with severe myocarditis, included 265 most of the MHC class II genes suggesting a decrease in antigen processing and presentation 266 pathways (Figures 6 C and S6D) . 267

Transcription factor prediction in EnrichR revealed overexpression of targets of the NF-B 268 complex in MIS-C patients with severe myocarditis as compared to MIS-C without myocarditis 269 ( Figure S6E ). This activation of NF-B signaling in MIS-C patients with severe myocarditis 270 was found to be associated with the strong downregulation of NF-B inhibitors as shown in 271 Figure 5E . 272 A strong overexpression of genes belonging to TNF- signaling, as well as inflammatory 273 responses, hypoxia and response to oxidative stress (HIF1A, HMOX1, HMBG1, etc.) was found 274 in cases with severe myocarditis (Figures 6D, E and S6E ). This was associated with a 275 downregulation of genes linked with oxidative phosphorylation, nitric oxide production and 276 iNOS signaling ( Figure S6B ). TGF- signaling and VEGF signaling were also found enriched 277 in monocytes and DCs of patients with myocarditis and to a lesser magnitude in B cells 278 (Figures S6 B, E) . Interestingly, an increased expression of several genes encoding S100 279 proteins and calcium-binding cytosolic proteins, all known to serve as danger signals to 280 regulate cell migration, homeostasis and inflammation, were noticed in the cases of severe 281 myocarditis ( Figure S6F ) (Xia et al., 2018) . 282

To summarize, NF-B activation, a decreased expression of NF-B inhibitors, TNF- 283 signaling, together with a hypoxic response to oxidative stress, VEGF signaling, 284 downregulation of MHC-II genes and low type-I and type-II IFN responses characterize the 285 monocytes and DCs of children with MIS-C and severe myocarditis. 286

To identify a potential clinical relevance of our study, we searched for a molecular signature 289 that correlated with the appearance of severe myocarditis among the monocytes/DCs of 290 children with SARS-CoV-2-related MIS-C. By using several SC-RNA-SEQ comparison 291 strategies ( Figure 7A) , we identified 329 genes upregulated in monocytes and DCs of the MIS-292 C group with myocarditis (N=6) as compared to all other groups ( Figure 7A ). To validate this 293 molecular signature, RNA from PBMCs were sequenced from an independent group of 294 patients. A scoring system was generated, based on normalized expression represented by a Z-295 score, coupled with hierarchical clustering, in order to identify genes that were overexpressed 296 in children with myocarditis (MIS-C_MYO (CoV2 + ) group) as compared to the other groups 297 ( Figure S7A) . Within the 329 genes identified by SC-RNA-SEQ in monocytes and DCs of 298 patients with severe myocarditis, expression of 116 genes were found upregulated in PBMCs 299 from an independent group of 9 patients belonging to the MIS-C_MYO (CoV2 + ) group with 300 myocarditis (Figures 7B). From these genes, a signature score (SignatureSCORE) was 301 determined for each sample processed by Bulk-RNA-SEQ ( Figure 7C ). We then further 302 developed a RankingSCORE (Figures S7 A, B) to identify the top genes that contributed the 303 most to the monocytes and DCs myocarditis signature. This led to the identification of a set of 304 25 genes that clearly segregate patients with severe myocarditis from other MIS-C and KD 305 (CoV2 -) ( Figure 7D) . Consistently, most of these 25 genes belong to functional pathways that 306 were previously identified (Figures 6 and S6) , such as inflammation, oxidative stress, TNF- 307 and/or NF-B signaling, and in some cases already known markers of myocarditis or MIS-C 308 and/or COVID-19, such as genes coding for S100 proteins (Figures S7 C, D) . 309

Multi-parametric analysis of peripheral blood mononuclear cells from children with acute 311 respiratory infection and postacute hyperinflammation, collected during the COVID-19 312 pandemic, allowed to detect an inflammatory profile associated with a loss of circulating 313 monocytes and dendritic cells (DCs), as well as an upregulation of genes and pathways 314 involving NF-B signaling, oxidative stress with establishment of hypoxic conditions and 315 VEGF signaling. These pathways were upregulated in both acute and postacute groups of 316 patients, independently of SARS-CoV-2 infection. However, significant features of MIS-C 317 with severe myocarditis were detected specifically in monocytes and DCs including low type-318 I and type-II IFN responses, decreased expression of NF-kB inhibitors, increased TNF- 319 signaling and overexpression of HIF-1. 320

Acute cases were characterized by the detection of inflammatory markers in the plasma with a 321 particularly strong elevation of IL-8 and CXCL1, two chemokines known to mediate neutrophil 322 migration to the lung (Kunkel et al., 1991; Pease and Sabroe, 2002; Sawant et al., 2015) and a 323 modest elevation of IFN2 levels. These findings suggest that in some children, a suboptimal 324

anti-viral type-I interferon response, alongside a hyperinflammatory response (IL-6 levels and 325 exacerbation of the NF-kB pathway), could account for SARS-CoV-2 disease with pneumonia, 326 as compared to the very usual benign or even asymptomatic clinical course of SARS-CoV-2 327 infection in children. This has been previously observed in severe Respiratory Syncytial Virus 328 (RSV) infections (Hijano et al., 2019) . 329

In the postacute patients, elevated levels of plasma IFN-, IFN2, IL-10, IL-15, and, to a lesser 330 extent, TNF-, were found, as previously described in other cohorts (Brodsky et al., 2020; 331 Carter et al., 2020; Consiglio et al., 2020; Esteve-Sole et al., 2021; Gruber et al., 2020) . These 332 findings are typical of an ongoing anti-viral immune response, not directly related to SARS-333

CoV-2 infection. In addition, elevated chemokines such as CCL2, CCL3 and CCL4 may recruit 334 monocytes and DCs to tissues, possibly accounting for their reduced numbers observed in the 335 blood of those patients. Additional mechanisms such as apoptosis or other cell death pathways 336 may also be involved. 337

Cellular phenotypes that distinguish MIS-C from classical KD have been previously reported 338 (Brodsky et al., 2020; Consiglio et al., 2020; Esteve-Sole et al., 2021) . Brodin and colleagues 339 described several key differences such as elevated IL-17, IL-6 and CXCL10 that were only 340 observed in KD, associated with decreased naïve CD4 + T cells and increased central memory 341

and effector memory CD4 + T cells in MIS-C. In the present study, high levels of IL-17, IL-6 342 and CXCL10, were both found in MIS-C and KD (CoV2 -) groups. In addition, no major 343 differences in CD4 + T cell compartments were detected. Accordingly, only a few differentially 344 expressed genes were found between the MIS-C and KD (CoV2 -) groups. These data support 345 the hypothesis that MIS-C patients with KD features exhibit a molecular phenotype close to 346 the one seen in KD patients, suggesting overlapping pathogenesis mechanisms (Gruber et al., 347 2020) . Differences observed with previous reports by Brodin and colleagues, may be due to 348 inclusion of only patients with criteria for complete or incomplete KD among the MIS-C cases, 349 or technical differences in the respective studies, such as time of blood sampling relative to 350 admission to hospital and medical treatments. 351

However, we did find noticeable differences in MIS-C cases with severe myocarditis with 352 circulatory failure that required intensive care. The expression of a number of cytokines was 353 further increased in these cases, most of them related to the NF-B-TNF- signaling axis. 354

Elevated VEGF and TGF- and TGF- are potential drivers of angiogenesis and vascular 355 homeostasis, whereas elevated chemokines (CCL2, CCL3, CCL20, CX3CL1, CXCL10) could 356 mediate increased cell migration towards inflamed tissues. Molecular analysis confirmed an 357 upregulation of genes belonging to the TNF- and NF-B signaling pathways that were 358 specifically found in monocytes and DCs of MIS-C patients with severe myocarditis. A lower 359 expression of NF-B complex inhibitors, including TNFAIP3 (A20), TNFAIP2, NFKBIA, 360 NFKBIZ, was detected, suggesting a possible mechanism for NF-B sustained activation which 361 could then potentially lead to exacerbated TNF- signaling. Overall, these results point to a 362 potential role of monocytes and DCs in the pathogenesis of MIS-C with severe myocarditis, 363 which might not be directly driven by SARS-CoV-2 infection, but rather due to a tolerance 364 defect in limiting a pathological immune response, as already observed for other pathogens 365 (Goodnow, 2021). It would be interesting to investigate the presence of genetic variants among 366 MIS-C with severe myocarditis, in genes such as TNFAIP3, as already discussed (Goodnow, 367 2021) . The apparent hypoxic conditions detected in children with myocarditis, could also 368 account for the exacerbation of NF-B signaling. HIF-1, a sensor of oxidative stress, is well-369 known for being able to induce a switch from oxidative phosphorylation to glycolysis to limit 370 generation of reactive oxygen species (ROS). It can also activate NF-B signaling (D'Ignazio 371 and Rocha, 2016; D'Ignazio et al., 2016) . Additional environmental factors and/or genetic 372 predispositions could also be involved. Another striking feature was the low expression of 373 genes involved in type-I and type-II interferon responses, specifically in monocytes and DCs 374 of children with myocarditis, although IFN- and IFN2 proteins were elevated in the plasma 375 of all MIS-C patients. This reduced response to type-I IFN in the most severe forms of MIS-C 376 (with myocarditis and circulatory failure) is in part reminiscent of the impaired type-I IFN 377 activity observed in the most severe forms of COVID-19 in adults (Bastard et al., 2020; Hadjadj 378 et al., 2020; Zhang et al., 2020) . The search for auto-antibodies against IFN2 were negative 379 (data not shown) but presence of autoantibodies to ISGs (interferon stimulated genes) cannot 380 be excluded (Combes et al., 2021) . with elevated high-sensitivity cardiac troponin I levels (>26 ng/mL) and /or decreased cardiac 458 function (diastolic or systolic ventricular dysfunction at echocardiography), were considered 459 to have MIS-C with severe myocarditis (Brissaud et al., 2016; Canter Charles E. and Simpson 460 Kathleen E., 2014) . 461

For each included patient, we collected demographic data, symptoms, results of SARS-CoV-2 462 testing and other laboratory tests, echocardiograms, and treatments. All patients with negative 463 initial serology testing were retested after an interval of at least 3 weeks (Architect SARS-CoV-464 2 chemiluminescent microparticle immunoassay; Abbott Core Laboratory). 465

Healthy controls were recruited before the COVID-19 pandemic (before November 2019). 

For each patient and healthy donor, peripheral blood samples were collected on EDTA and 469 lithium heparin tubes. After a centrifugation of the EDTA tube at 2300rpm for 10 minutes, 470 plasma was taken and stored at -80°C before cytokine quantification. PBMCs were isolated 471 from the lithium heparin samples, frozen as described below and stored at -80°C and were used 472 for both bulk and single-cell RNAseq, as well as cell phenotyping by CyTOF. The workflow 473 is summarized in Figure 1B . Prior to protein analysis plasma samples were treated in a BSL3 laboratory for viral 487 decontamination using a protocol previously described for SARS-CoV (Darnell and Taylor, 488 2006) , which we validated for SARS-CoV-2. Briefly, samples were treated with TRITON 489 X100 (TX100) 1% (v/v) for 2hrs at Room Temperature. IFNα2, IFNγ, IL-17A, (triplex) and 490

IFNβ (single plex) protein plasma concentrations were quantified by Simoa assays developed 491 with Quanterix Homebrew kits as previously described (Rodero et al., 2017) . The limit of 20 detection of these assays were 0.6 pg/mL for IFNβ, 2 fg/mL for IFNα2, 0.05 pg/ml for IFNγ 493 and 3 pg/mL for IL17A including the dilution factor. IL-6, TNFα, and IL-10 were measured 494 with a commercial triplex assay (Quanterix). Additional plasma cytokines and chemokines (44 495 analytes) were measured with a commercial Luminex multi-analyte assay (Biotechne, R&D 496 systems). 497 498 Serology assays 499 SARS-CoV-2 specific antibodies were quantified using assays previously described (Grzelak 500 et al., 2020) . Briefly, a standard ELISA assay using as target antigens the extracellular domain 501 of the S protein in the form of a trimer (ELISA tri-S) and the S-Flow assay, which is based on 502 the recognition of SARS-CoV-2 S protein expressed on the surface of 293T cells (293T-S), 503 were used to quantify SARS-CoV-2 specific IgG and IgA subtypes in plasma. Assay 504 characteristics including sensitivity and specificity were previously described (Grzelak et al., 505 2020) . Then 270 L of the samples were directly added to the dry antibody cocktail for 30 minutes. 3 514 mL of MaxPar Water was added to each tube for an additional 10-min incubation. Three 515 washes were performed on all the samples using MaxPar Cell Staining Buffer and they were 516 fixed using 1.6% paraformaldehyde (Sigma-Aldrich, France). After one wash with MaxPar 517

Cell Staining Buffer, cells were incubated one hour in Fix and Perm Buffer with 1:1000 of 518

Iridium intercalator (pentamethylcyclopentadienyl-Ir (III)-dipyridophenazine, Fluidigm, Inc 519

Maxpar Cell Acquisition Solution, a high-ionic-strength solution, and mixed with 10% of EQ 521

Beads immediately before acquisition. 522

Acquisition of the events was made on the Helios mass cytometer and CyTOF software version 523 6.7.1014 (Fluidigm, Inc Canada) at the "Plateforme de Cytométrie de la Pitié-Salpetriere 524 (CyPS)." An average of 500,000 events were acquired per sample. Dual count calibration, 525 noise reduction, cell length threshold between 10 and 150 pushes, and a lower convolution 526 threshold equal to 10 were applied during acquisition. Mass cytometry standard files produced 527 by the HELIOS were normalized using the CyTOF Software v. 6.7.1014. For data cleaning, 4 528 parameters (centre, offset, residual and width) are used to resolve ion fusion events (doublets) 529 from single events from the Gaussian distribution generated by each event (Bagwell et al., 530 2020). Subsequent to data cleaning, the program produces new FCS files consisting of only 531 intact live singlet cells. These data were analyzed in FlowJo v10.7.1 using 3 plugins 532 (DownSampleV3, UMAP and FlowSOM) with R v4.0.2. To increase efficiency of the analysis, 533 samples were downsampled to 50 000 cells, using the DownSample V3 plugin. All samples 534 were concatenated and analyzed in an unsupervised manner. Anti-CD127 antibody had to be 535 excluded due to poor staining. Clustering was performed using FlowSOM (Van Gassen et al., 536 2015) . The number of clusters was set to forty-five in order to overestimate the populations 537 and detect smaller subpopulations. Grid size of the self-organizing map was set to 20x20. 538

Resulting clusters were annotated as cell populations following the kit manufacturer's 539 instruction. When several clusters were identified as the same cell types, they were Sequencing reads were demultiplexed and aligned to the human reference genome (GRCh38, 556 release 98, built from Ensembl sources), using the CellRanger Pipeline v3.1. Unfiltered RNA 557 UMI counts were loaded into Seurat v3.1 (Stuart et al., 2019) for quality control, data 558 integration and downstream analyses. Apoptotic cells and empty sequencing capsules were 559 excluded by filtering out cells with fewer than 500 features or a mitochondrial content higher 560 than 20%. Data from each sample were log-normalized and scaled, before batch correction 561 using Seurat's FindIntegratedAnchors. For computational efficiency, anchors for integration 562 were determined using all control samples as reference and patient samples were projected onto 563 the integrated controls space. On this integrated dataset, we computed the principal component 564 analysis on the 2000 most variable genes. UMAP was carried out using the 20 most significant 565 principal components (PCs), and community detection was performed using the graph-based 566 modularity-optimization Louvain algorithm from Seurat's FindClusters function with a 0.8 567 resolution. Cell types labels were assigned to resulting clusters based on a manually curated 568 list of marker genes as well as previously defined signatures of the well-known PBMCs 569 subtypes (Monaco et al., 2019). Despite filtering for high quality cells, five clusters out of the 570 twenty-six stood out as poor quality clusters and were removed from further analysis, namely: 571 one erythroid-cell contamination; one low UMI cluster from a single control; two clusters of 572 proliferating cells originating from a patient with EBV co-infection and one megakaryocytes 573 cluster. In total 152,201 cells were kept for further analysis. Courtaboeuf, France). To note, the optional step with the DNase was performed. RNA integrity 584 and concentration were assessed by capillary electrophoresis using Fragment Analyzer 585 (Agilent Technologies). RNAseq libraries were prepared starting from 100 ng of total RNA 586 using the Universal Plus mRNA-Seq kit (Nugen) as recommended by the manufacturer. The 587 oriented cDNA produced from the poly-A+ fraction was sequenced on a NovaSeq6000 from 588

Illumina (Paired-End reads 100 bases + 100 bases). A total of ~50 millions of passing-filters 589 paired-end reads was produced per library. 590

Paired-end RNA-seq reads were aligned to the human Ensembl genome GRCh38.91 using 591

Hisat2 (v2.0.4) (Kim et al., 2019) and counted using featureCounts from the Subread R 592 package. The raw count matrix was analyzed using DESeq2 (version 1.28.1) (Love et al., 593 2014) . No pre-filtering was applied to the data. Differential expression analysis was performed 594 using the "DESeq" function with default parameters. For visualization and clustering, the data 595 was normalized using the `variant stabilizing transformation` method implemented in the "vst" 596 function. Plots were generated using ggplot2 (version 3.3.2), and pheatmap (version 1.0.12). 597

During exploratory analyses, it was noted that the clustering was mainly driven by the sex of 598 the patients. To remove this effect, it was included in the regression formula for DESeq (~sex 599 + groups), and then removed following vst transformation, using "removeBatchEffect" from 600 the "limma" package (version 3.44.3). 601 602 Gene signature analysis 603 To identify genes that could be used as markers of severe myocarditis in the SC-RNA-SEQ 604 dataset, three initial strategies were used, all based on differential expression and selection of 605 the upregulated genes. First, we performed the differential expression between MIS-C_MYO 606 (CoV2 + ) samples and all other samples. Second, differential analysis was computed between 607 MIS-C_MYO (CoV2 + ) and other samples with postacute hyperinflammatory illness. In the last 608 strategy, we selected genes that were upregulated between the MIS-C_MYO (CoV2 + ) and the 609 CTL, but not upregulated in any other group compared to the CTL ( Figure 7A ). These three 610 strategies allowed us to identify 329 unique genes. 611

To further explore whether these genes could be considered as markers of severe myocarditis, 612

we analyzed their expression profile in our bulk RNA-SEQ dataset. This dataset excluded 613 samples from patients of the MIS-C_MYO (CoV2 + ) that were included in the SC-RNA-SEQ 614 cohort. Vst-transformed counts were log2-normalized and converted to z-score using the scale 615 function in R (v 4.0.2). A GeneSCORE was computed for each group as the mean z-score of 616 the samples of a group. Heatmaps representing this GeneSCOREgroup were performed using 617 pheatmap. Hierarchical clustering of the 329 previously identified genes was performed using 618 the complete method on the distance measured using Pearson's correlation, as implemented by 619 pheatmap. The hierarchical clustering was divided into 15 main clusters, 4 of which had the 620 expected pattern of expression: Clusters that had a higher expression in MIS-C_MYO (CoV2 + ) 621 than any other group were selected, resulting in 116 genes. A signature score for each sample 622 was performed on these genes, corresponding to the mean expression (z-score) of these N genes 623 in each sample (SignatureSCORE). 624

These genes were subsequently ranked based on the following equation: 625

where the SCOREs represent the mean expression (z-score) in each disease groups, and the 629 SignatureScore was computed on the top 25 genes. Immunitaires Héréditaires (CEREDIH), the Agence National de la Recherche (ANR-flash 667

Covid19 "AIROCovid" to FRL and "CoVarImm" to DD and JDS), and by the FAST 668 

Groups of patients 1

Acute-Inf (CoV2+) n= 9

General Positive NP SARS-CoV-2 RT-PCR 0 9 (100) 4 (44) 10 (48) IL2RB  NCAM1  TRDC  KLRD1  KLRC1  GZMM  GZMH  GZMK  GZMB  GZMA  NKG7  CCL4  CD8B  CD8A  IL32  CCR6  KLRB1  TNFRSF4  FOXP3  TIMP1  LRRN3  CCR7  CD4  TRAC  CD3D  DNASE1L3  LILRB4  JCHAIN  TNFRSF17  TNFRSF13B  CD80  TCL1A  CD24  BLK  CD79A CD19 S100A9 S100A8 HLA−DRA

LGALS1 Figure 3B . PBMCs represent all clusters; monocytes/DCs cells, clusters 5, 11, 12, 17, 20, 21, 24; T cells, clusters 0, 1, 2, 4, 6, 7, 10, 13, 14, 15, 16, 18 and 23; and B cells, clusters 3, 8, 9, 19, and 22 

CTL MIS-C (CoV2 + ) F S100A14 S100A1 S100A16 S100A3 S100A13 S100A2 S100A4 S100A5 S100A6 S100A8 S100A12 S100A9 S100A11 S100A10

Features Percent Expression of S100 genes RNF24  S100A12  S100A8  PGD  SPI1  RETN  SDCBP  FOXO3  BACH1  QSOX1  TMEM167B  FTL  TOP1MT  RBM3  HADHB  CAPNS1  APLP2  MBOAT7  SLC44A1  RNF141  ASAH1  TNNT1  KRT10  FNDC3B  QKI  CMTM6  BRI3  ADAP2  POLR2L  S100A6  FCAR  HBEGF  FBP1  TIMP2  IRF2BP2  SIRPA  ZYX  LGALS1  TMA7  GSTO1  NDUFB9  ASPH  CHMP4B  THBD  TPD52L2  GADD45GIP1  ATP6V1F  RPL37A  CCDC71L  CD9  TLNRD1  IL1R1  FAM20C  PHLDA1  SRA1  RPL21  PFDN1  RUNX1  RPL37  AATK  CARD19  PGPEP1  RIT1  BNIP2  STK24  RPL22  ZFAND3  VIM  GAPDH  NAMPT  ADD3  ANPEP  NRIP1  IFNGR2  GMFB  P2RX1  CLU  PHC2  AREG  GLUL  LAPTM5  SLC25A37  ATF4  ALOX5AP  RPL38  TPST1  CSGALNACT2  CD63  RPL35A  ADAM9  ARL8A  SLC6A6  MMP24OS  PF4  PPBP  RPS4Y1  FAM49B  RFLNB  GNG5  S100A10  SMIM3  RPS9  RAC1  RPL7  CTSD  SH3BGRL3  RPL6  ATP6V0B  DAZAP2  CTSA  C5AR1  LEPROT Figure S7A ). D.

Boxplot showing the SignatureScore computed on the expression of the top 25 genes, as ranked in Figure S7B , in the Bulk-RNA-SEQ dataset.

Each dot represents a sample. Boxes range from the 25th to the 75th percentiles. The upper and lower whiskers extend from the box to the largest and smallest values respectively.

Any samples with a value at most x1.5 the inter-quartile range of the hinge is considered an outlier and plotted individually. CAPNS1  S100A8  PPBP  CTSA  PF4  PGD  P2RX1  S100A12  IFNGR2  TOP1MT  SLC25A37  VAT1  RBM3  CTSD  PGPEP1  TPST1  RPS9  GADD45GIP1  GAPDH  ALOX5AP  SH3BGRL3  PFDN1  LGALS1  PHC2  ATF4  RAC1  RPL6  LAPTM5  RPL38  SLC44A1  GMFB  HADHB  NAMPT  STK24  VIM  FTL  NDUFB9  RNF24  MMP24OS  GLUL  DAZAP2  RNF141  AREG  SPI1  CCDC69  APLP2  S100A6  SMIM3  RPL22  RPL7  CD63  LEPROT  RPS4Y1  CTSB  ASPH  ARL8A  ANPEP  BNIP2  ATP6V1F  BACH1  RPL37  ATP6V0B  POLR2L  ZYX  GSTO1  S100A10  GNG5  RPL35A  CHMP4B  QSOX1  FOXO3  CSGALNACT2  RUNX1  ASAH1  RFLNB  TNNT1  BRI3  RPL37A  SDCBP  AATK  CCDC71L  CARD19  TPD52L2  ADAM9  RPL21  TMEM167B  MBOAT7  ADD3  TIMP2  SRA1  THBD  CD9  ZFAND3  QKI  IL1R1  CXXC5  NRIP1  FBP1  CMTM6  SIRPA  C5AR1  TMA7  IRF2BP2  PHLDA1  TLNRD1  SLC6A6  FNDC3B  ADAP2  FAM49B  FAM20C  KRT10  HBEGF  RIT1  FCAR   RETN  CLU  CAPNS1  S100A8  PPBP  CTSA  PF4  PGD  P2RX1  S100A12  IFNGR2  TOP1MT  SLC25A37  VAT1  RBM3  CTSD  PGPEP1  TPST1  RPS9  GADD45GIP1  GAPDH  ALOX5AP  SH3BGRL3  PFDN1  LGALS1  PHC2  ATF4  RAC1  RPL6  LAPTM5  RPL38  SLC44A1  GMFB  HADHB  NAMPT  STK24  VIM  FTL  NDUFB9  RNF24  MMP24OS  GLUL  DAZAP2  RNF141  AREG  SPI1  CCDC69  APLP2  S100A6  SMIM3  RPL22  RPL7  CD63  LEPROT  RPS4Y1  CTSB  ASPH  ARL8A  ANPEP  BNIP2  ATP6V1F  BACH1  RPL37  ATP6V0B  POLR2L  ZYX  GSTO1  S100A10  GNG5  RPL35A  CHMP4B  QSOX1  FOXO3  CSGALNACT2  RUNX1  ASAH1  RFLNB  TNNT1  BRI3  RPL37A  SDCBP  AATK  CCDC71L  CARD19  TPD52L2  ADAM9  RPL21  TMEM167B  MBOAT7  ADD3  TIMP2  SRA1  THBD  CD9  ZFAND3  QKI  IL1R1  CXXC5  NRIP1  FBP1  CMTM6  SIRPA  C5AR1  TMA7  IRF2BP2  PHLDA1  TLNRD1  SLC6A6  FNDC3B  ADAP2  FAM49B  FAM20C  KRT10  HBEGF  RIT1  FCAR   RETN  CLU  CAPNS1  S100A8  PPBP  CTSA  PF4  PGD  P2RX1  S100A12  IFNGR2  TOP1MT  SLC25A37  VAT1  RBM3  CTSD  PGPEP1  TPST1  RPS9  GADD45GIP1  GAPDH  ALOX5AP  SH3BGRL3  PFDN1  LGALS1  PHC2  ATF4  RAC1  RPL6  LAPTM5  RPL38  SLC44A1  GMFB  HADHB  NAMPT  STK24  VIM  FTL  NDUFB9  RNF24  MMP24OS  GLUL  DAZAP2  RNF141  AREG  SPI1  CCDC69  APLP2  S100A6  SMIM3  RPL22  RPL7  CD63  LEPROT  RPS4Y1  CTSB  ASPH  ARL8A  ANPEP  BNIP2  ATP6V1F  BACH1  RPL37  ATP6V0B  POLR2L  ZYX  GSTO1  S100A10  GNG5  RPL35A  CHMP4B  QSOX1  FOXO3  CSGALNACT2  RUNX1  ASAH1  RFLNB  TNNT1  BRI3  RPL37A  SDCBP  AATK  CCDC71L  CARD19  TPD52L2  ADAM9  RPL21  TMEM167B  MBOAT7  ADD3  TIMP2  SRA1  THBD  CD9  ZFAND3  QKI  IL1R1  CXXC5  NRIP1  FBP1  CMTM6  SIRPA  C5AR1  TMA7  IRF2BP2  PHLDA1  TLNRD1  SLC6A6  FNDC3B  ADAP2  FAM49B  FAM20C  KRT10  HBEGF  RIT1  FCAR   RETN  CLU  CAPNS1  S100A8  PPBP  CTSA  PF4  PGD  P2RX1  S100A12  IFNGR2  TOP1MT  SLC25A37  VAT1  RBM3  CTSD  PGPEP1  TPST1  RPS9  GADD45GIP1  GAPDH  ALOX5AP  SH3BGRL3  PFDN1  LGALS1  PHC2  ATF4  RAC1  RPL6  LAPTM5  RPL38  SLC44A1  GMFB  HADHB  NAMPT  STK24  VIM  FTL  NDUFB9  RNF24  MMP24OS  GLUL  DAZAP2  RNF141  AREG  SPI1  CCDC69  APLP2  S100A6  SMIM3  RPL22  RPL7  CD63  LEPROT  RPS4Y1  CTSB  ASPH  ARL8A  ANPEP  BNIP2  ATP6V1F  BACH1  RPL37  ATP6V0B  POLR2L  ZYX  GSTO1  S100A10  GNG5  RPL35A  CHMP4B  QSOX1  FOXO3  CSGALNACT2  RUNX1  ASAH1  RFLNB  TNNT1  BRI3  RPL37A  SDCBP  AATK  CCDC71L  CARD19  TPD52L2  ADAM9  RPL21  TMEM167B  MBOAT7  ADD3  TIMP2  SRA1  THBD  CD9  ZFAND3  QKI  IL1R1  CXXC5  NRIP1  FBP1  CMTM6  SIRPA  C5AR1  TMA7  IRF2BP2  PHLDA1  TLNRD1  SLC6A6  FNDC3B  ADAP2  FAM49B  FAM20C  KRT10  HBEGF  RIT1 Top 25 genes of the signature S100A7A S100A1 S100A7 S100A3 S100A4 S100A8 S100A9 S100A12 S100A6 S100A10 S100A11 S100A14 S100A13 S100A5 S100A2 S100A16 

Mapping Systemic Inflammation and 755

Antibody Responses in Multisystem Inflammatory Syndrome in Children (MIS-C)

A comparison of four 759 serological assays for detecting anti-SARS-CoV-2 antibodies in human serum samples from 760 different populations

Spread of 763 SARS-CoV-2 in the Icelandic Population

Impaired type I interferon activity and 766 inflammatory responses in severe COVID-19 patients

TRRUST v2: an expanded reference database of human and mouse transcriptional 769 regulatory interactions

IFN) in the Respiratory Syncytial Virus (RSV) Immune Response and 772 Disease Severity

Tumour 774 necrosis factor signalling in health and disease

Single-Cell

COVID-19 and Kawasaki Disease: Novel Virus and 780 Novel Case

Recognition of a Kawasaki disease shock syndrome

Graph-based genome 785 alignment and genotyping with HISAT2 and HISAT-genotype

Enrichr: a comprehensive gene set 788 enrichment analysis web server 2016 update

KD (CoV2 -) vs CTL Monocytes/cDCs/pDCs

MIP-1 / CCL3 (pg/mL) *** *** CTL Acute-Inf (CoV2 -) Acute-Inf (CoV2 + ) MIS-C (CoV2 + ) MIS-C_MYO (CoV2 + ) KD (CoV2 -)