Two novel kindreds with autosomal recessive STAT2 deficiency

Type I interferons (IFNs) and, to a lesser extent, type II and type III IFNs, are crucial signaling molecules in antiviral defense. While type I IFNs (IFN-α/β) and type II IFNs (IFN-γ) are ubiquitously secreted, type III IFNs (IFN-λ) are only secreted by epithelial cells and dendritic cells (1, 2). The type I IFN receptor, a heterodimer composed by the two subunits IFNAR1 and IFNAR2, is ubiquitously expressed, while the type III IFN receptor, composed by the subunits IFNLR1 and IL10RB, is only expressed on epithelial cells, blood–brain barrier endothelial cells, and a subset of human leukocytes (plasmacytoid dendritic cells and B cells) (3, 4, 5). After binding to their receptors, type I and III IFNs trigger the phosphorylation of kinases constitutively associated to their receptors (TYK2 and JAK1) and the recruitment of signal transducer and activator (STAT) molecules (6, 7). Although seven STAT molecules are known, STAT1 and STAT2 are the main players in antiviral defense (8). After their recruitment, STAT1 and STAT2 are phosphorylated and form dimers. STAT1 homodimerizes and binds to γ-activated sequence elements in the promoter of IFN-stimulated genes (ISGs). While STAT1 can form homodimers, STAT2 can only bind to STAT1 to form heterodimers. The heterodimer combines with IRF9, and the resulting complex, heterotrimer ISG factor 3 (ISGF3), translocates to the nucleus. There, ISGF3 can bind to the promoter region of ISGs to induce antiviral response. In addition, STAT2 recruits ubiquitin-specific protease 18 (USP18) to IFNAR2. USP18 inhibits the function of IFNAR2 and downregulates the type I IFN response, thus preventing hyperactivation and mediating the resolution of the IFN response (7, 9, 10, 11, 12). Stimulation with type I and type III IFN induces mainly the formation of STAT1/STAT2 heterodimers, while stimulation with type II IFN (IFN-γ) induces the formation of STAT1 monomers (2).

Five loss-of-function (LOF) defects in genes connected to this signaling cascade have been reported. IFNAR1, IFNAR2, IRF9, and STAT2 deficiencies are characterized by susceptibility to viral infections, including those caused by live-attenuated vaccines (LAVs), while STAT1 deficiency causes a broader susceptibility to viral and mycobacterial infections (7, 9, 11, 13, 14, 15, 16, 17, 18, 19). The typical presentation of autosomal recessive (AR) STAT2 deficiency is indeed disseminated infection and encephalitis following administration of LAV. Hence, most patients are diagnosed early in life following vaccination with the measles-mumps-rubella (MMR) vaccine, although penetrance is not complete for this manifestation (16). Patients are also susceptible to severe infection with naturally occurring viral agents such as herpes simplex virus 1, Epstein-Barr virus, respiratory syncytial virus, influenza A, or severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) (7, 9, 11, 13, 16, 17, 20). In addition, systemic inflammation following infection is a typical clinical presentation of STAT2 deficiency and is triggered by uncontrolled viral proliferation and a prominent neutrophil and monocyte signature (16). 25 patients with AR STAT2 deficiency have been reported so far, with mortality reaching 35% in early childhood (Table 1 and Table 2) (16, 21, 22, 23).

Here we describe two unrelated kindreds harboring two novel variants in STAT2, underlying AR STAT2 deficiency.

We studied three patients from two kindreds carrying biallelic STAT2 variants (Fig. 1 A and Fig. S1). Patients 1 (P1) and 2 (P2) are siblings born to healthy consanguineous parents (first cousins) from Portugal. They both presented with frequent recurrent infections and hyperinflammatory symptoms. Briefly, P1 suffered recurrent severe infections since the age of 22 days, including sepsis, viral meningitis, multiple episodes of pneumonia, purulent otitis, and gastroenteritis. 15 days after MMR vaccination, P1 presented with hepatosplenomegaly, generalized lymphadenopathies, and oral candidiasis. She developed asthma, chronic rhinosinusitis, localized bronchiectases with frequent chest infections necessitating antibiotics, and a pulmonary aspergilloma. She was treated with cotrimoxazole and itraconazole as prophylaxis. At the age of 31 years, she developed a chronic anal fissure and terminal ileitis due to Crohn’s disease was diagnosed.

P2 developed varicella at the age of 15 mo, which resolved without complications. Nine days after MMR vaccination, P2 was hospitalized with a working diagnosis of atypical Kawasaki disease characterized by fever, rash, conjunctivitis, rhinitis, cheilitis, pharyngitis, and edema of the extremities, complicated by a Staphylococcus epidermidis sepsis and meningitis. Notably, he had persistent fever for 41 days during this infectious episode. He had an episode of cellulitis of the leg requiring surgical debridement at the age of 31 years old. He was treated with prophylactic cotrimoxazole. Both patients had normal serum levels of IgA, IgM, IgG, and IgE. Specific IgG antibodies against pneumococcal polysaccharides, tetanus toxoid, and rubella were in normal range. Moreover, normal complement activity and oxidative burst were observed.

P3 was born to nonconsanguineous parents of Algerian origin and presented at the age of 12 mo with a severe adverse reaction following MMR vaccine with polymorphic cutaneous rash and fever. On admission, she had bicytopenia (anemia and thrombocytopenia), hyperferritinemia (3,400 µg/L), hypofibrinogenemia (0.4 g/L), and hepatosplenomegaly. Her immunophenotyping showed markedly increased effector memory (EM) CD8⁺ T cells. A lumbar puncture revealed lymphocytic hypercellularity but was negative for hemophagocytosis, while a bone marrow aspirate showed a normocellular marrow enriched in macrophages and with ongoing hemophagocytosis. A brain MRI was performed and showed diffuse parenchymal atrophy associated with a washed-out appearance and T2/FLAIR hyperintensity in the deep white matter and mild signal abnormality in the lenticular nuclei with increased cerebral blood flow, compatible with central hemophagocytic lymphohistiocytosis (HLH). Systemic and central nervous system HLH was diagnosed based on the HLH-2024 criteria (28). She was treated with one course of high-dose intravenous immunoglobulins (IVIG), steroids, and ruxolitinib (20 mg/m²/day) for 2 mo with complete resolution of the cytopenia, normalization of fibrinogen, triglycerides, and cellularity in the cerebrospinal fluid, reduction of ferritin levels, and resolution of inflammation and the other HLH symptoms.

Whole-exome sequencing was performed on P1, P2, and their healthy mother to search for candidate genetic variants, testing the AR inheritance model. The variants were then filtered according to their frequency (minor allele frequency <0.01) and quality (read depth > 10). As a result, a homozygous c.1467_1468insC (K490Qfs*41) STAT2 variant (NM_005419.4, 12:56349033-5649033) was identified in P1 and P2, which was further confirmed by Sanger sequencing (Fig. 1, A and B). The healthy mother and healthy sibling were heterozygous, and no material was available from the father (Fig. 1, A and B). The mutation was predicted to be pathogenic in silico (combined annotation dependent depletion [CADD] score: 24, mutation significance cutoff [MSC] 10) and to lead to an early stop interrupting the translation (ExPASy, Swiss-Prot Group, https://web.expasy.org/translate/). Five other homozygous rare nonsynonymous variations were also identified, but none of them resided in known inborn error of immunity (IEI)–related genes (Table S3).

A novel private homozygous essential splice variant in the STAT2 gene (12:56351291, c.941+1G>T) was found in P3 by a next generation sequencing–based panel, consisting of 500 known IEI-related genes. This variant was predicted to be pathogenic in silico (CADD score: 33, SpliceAI: donor loss). Both parents were heterozygous carriers of this variant (Fig. 1, A–C). Upon diagnosis of STAT2 deficiency, IVIG substitution was started and ruxolitinib was discontinued. The population genetics and position in STAT2 of the two novel variants relative to the other reported STAT2 mutations are shown in Fig. 1, D and E.

Full-length STAT2 protein expression was absent in patients’ primary cells (peripheral blood mononuclear cells [PBMCs] for P3 and fibroblasts for P1 and P2), as measured by immunoblot (Fig. 2 A and Fig. 3 A). Quantitative PCR (qPCR) revealed complete absence of STAT2 mRNA in primary fibroblasts of P1 and P2 carrying the K490Qfs*41 variant (Fig. 2 B). P3’s variant c.941+1G>T is located within a splice donor site, suggesting a possible effect on the splicing of STAT2 mRNA. Sequencing of P3’s cDNA after pJet cloning identified six different splice variants but no wild-type STAT2 (Fig. 2 C). The three most frequent transcripts, namely the deletion of exon 9 (DELex9), the insertion of intron 9 (INSin9), and the insertion of the first 47 nucleotides of intron 8 (INSin8), as well as P1 and P2’s K490Qfs*41 variant, were used to generate plasmids via site-directed mutagenesis and overexpressed in HEK293T cells. The variant R510X was included as a known LOF stop-mutation (25) Variable amounts of truncated STAT2 proteins of different size were detected for each variant by immunoblot (Fig. 2D). Finally, FACS analysis confirmed the absence of STAT2 protein expression in P3-derived T cell blasts (Fig. S2).

The phosphorylation of STAT2 and STAT1 was induced through 30 min of IFNα and IFNγ stimulation, respectively, both in P1 and P2’s primary fibroblasts (Fig. 3 A) and in HEK293T cells transfected with plasmids containing the STAT2 variants K490Qfs*41, DELex9, INSin9, and INSin8 (Fig. 3 B). STAT1 phosphorylation was normal for all variants and controls after stimulation with IFNα and IFNγ (Fig. 3, A and B). None of the patients’ variants in the HEK overexpression system showed any phosphorylation of STAT2 after stimulation with IFNα, while this was normal in healthy controls (Fig. 3 B). Absence of STAT2 phosphorylation after treatment with IFNα for 30 min was also shown in primary fibroblast of P1 and P2 (Fig. 3 A) and in T cell blasts of P3 (Fig. S2, showing also normal STAT1 phosphorylation in the same cells), while it was preserved in P1 and P2’s carrier mother (Fig. 3 A).

We next measured the induction of ISGs expression after 6 h of IFNα and IFNγ stimulation in STAT2-deficient U6A fibrosarcoma cells transfected with plasmids containing the STAT2 variants K490Qfs*41, DELex9, INSin9, and INSin8. The upregulation of the ISGs RSAD2, IFIT1, GBP1, and USP18 was identified by qPCR. The variant R510X was included as a known LOF stop-mutation (25). The studied STAT2 variants failed to induce ISG transcription following type I IFN stimulation (Fig. 4 A). In contrast, stimulation with IFNγ led to a normal upregulation of transcription of ISGs (Fig. 4 B).

We report two novel homozygous LOF mutations in STAT2 leading to AR STAT2 deficiency in two unrelated families from Portugal and Algeria. Both mutations (K490Qfs*41 in P1 and P2 and c.941+1G>T in P3) lead to the complete absence of full-length STAT2 protein in the patients. Functional analysis of these variants show that they lead to absent or truncated proteins that undergo degradation and are not functional, as they do not support STAT2 phosphorylation and the induction of ISG transcription in response to IFNα.

All three patients reported in this study experienced life-threatening infection with live-attenuated MMR. Severe reaction to LAVs is a key presentation of AR STAT2 deficiency. In a recent study, 12 out of 18 patients with STAT2 deficiency experienced infectious and/or inflammatory complications, including HLH and Kawasaki-like disease, after LAV administration (16). HLH as complication of LAV administration was also described in IFNAR1, IFNAR2, IRF9, and STAT1 deficiency (26, 27, 29) and has recently been reported in another STAT2 deficient patient after MMR vaccination (27). Interestingly, this patient also suffered from multisystem inflammatory syndrome in children after SARS-CoV-2 infection, despite being vaccinated against it (27). In this study, P3 presented with HLH after MMR vaccination and was treated with immune suppression including ruxolitinib, an inhibitor of JAK1 and JAK2 (Jakinib). The patient’s cytopenia resolved during ruxolitinib treatment, although it was discontinued after the diagnosis of STAT2 deficiency because of the risk of viral infections. Nevertheless, its successful and safe use raises the question whether patients suffering from hyperinflammation in the context of type I IFN defects could indeed benefit from Jakinibs or other targeted anti-inflammatory treatments. Hyperinflammation, especially in the form of HLH, has been extensively described in patients with STAT2, IFNR1, and IRF9 deficiency (1, 9, 12, 16, 21, 26, 27). A proposed underlying mechanism could be a prolonged type I IFN signaling due to insufficient negative regulation by STAT2 and USP18, although we previously showed an increased tumor necrosis factor (TNF)/nuclear factor κ-B (NF-κB) and JAK/STAT3 signaling and roles for circulating monocytes, neutrophils, and CD8 EM cells in inflammation in patients with STAT2 or related deficiencies (9, 16). We speculate that ruxolitinib, acting directly on the JAK–STAT complexes, was effective in interrupting this proinflammatory cascade and assisted in the downregulation of this abnormal response.

In young patients with severe viral infections and/or complications after LAVs, STAT2 deficiency should be excluded. Since STAT2 deficiency is fatal in 35% of patients in early childhood (7, 16, 30), early genetic testing is recommended. LAVs are to be avoided in STAT2-deficient patients, and early intervention in case of viral infections is crucial. The mortality rate seem to decrease with age, probably due to compensation from the adaptive immune system and other innate immune components that seem to rely less on type I IFN for their development and function, such as natural killer cells (31). Nevertheless, exposure to novel viruses can still cause severe and even fatal outcomes in adulthood, as the recent SARS-CoV-2 pandemic demonstrated (16, 22, 32, 33). Despite a still significant mortality after LAV and during acute viral infections, STAT2-deficient patient experience relatively few life-threatening infections or other complications, not warranting the risk of curative treatments like hematopoietic stem cell transplantation. They are usually managed with antiviral and supportive treatment during infectious/inflammatory episodes and sometimes with prophylactic acyclovir and immunoglobulin substitution. The use of ruxolitinib as a targeted anti-inflammatory drug during acute episodes of hyperinflammation/HLH could be an interesting addition to the management of STAT2-deficient patients, as successfully demonstrated in P3. An alternative could be emapalumab, a monoclonal antibody that has been approved for the treatment of primary HLH (34, 35). Targeting IFN-γ specifically, emapalumab could counteract STAT1-mediated hyperinflammation while presenting a safer profile in the context of viral infections compared to the broader action of ruxolitinib. Nevertheless, monoclonal/immunomodulatory agents should be tested with caution, ideally in the context of a multicenter clinical study with protocols to assess efficacy and safety.

Overall, our case report contributes to a better understanding of STAT2 deficiency and human antiviral immunity. Our patients’ presentations confirm once again that the phenotypic spectrum of STAT2 deficiency is relatively narrow, although characterized by incomplete penetrance. In addition, we report the first successful use of a Jakinib in a patient with a JAK–STAT defect and HLH, highlighting the complexity of the regulation of the type I IFN pathway and the need for more studies in order to offer these patients better therapeutic options.

Patients were included through referral by the treating physicians upon clinical presentation and upon identification of a STAT2 variant in the whole-exome sequencing analysis. All patients or their legal guardians gave consent for participation in the study. The study was approved by the Ethical Research Committee of Leuven University Hospitals (study number S60905). Informed consent was obtained in France in accordance with local regulations and a human-subjects research protocol approved by the institutional review board of the Institut National de la Santé et de la Recherche Médicale (INSERM). Approval was obtained from the French Ethics Committee (Comité de Protection des Personnes), the French National Agency for Medicine and Health Product Safety, and INSERM in Paris (protocol C10-13). The data were collected via a case record form.

For P1 and P2: Whole-exome sequencing was performed as previously described (36). In brief, exome capture was performed with SureSelect Human All Exon 50 Mb kit (Agilent Technologies). Paired-end sequencing was performed on an Illumina HiSeq 2000 (Illumina), generating 100-base reads. The reads were then mapped onto the human reference genome (GRCh38/hg38) with the Burrows-Wheeler Aligner (37). Downstream processing was performed with the Genome Analysis Toolkit (GATK) according to documented best practice. All variants were then filtered and annotated with in-house developed software.

For P3: Genomic DNA was isolated by phenol-chloroform extraction from peripheral blood cells or primary fibroblasts from the patient. DNA (3 µg) was sheared with a Covaris S2 Ultrasonicator (Covaris). An adapter-ligated library was prepared with the TruSeq DNA Sample Prep Kit (Illumina). Exome capture was done using the SureSelect Human All Exon 50 Mb kit (Agilent Technologies). Paired-end sequencing was done on an Illumina HiSeq 2000 (Illumina), generating 100-base reads. The sequences were aligned with the human genome reference sequence (hg19 build), with the Burrows-Wheeler aligner (v.0.7.12). Downstream processing was done using GATK (v.3.4), SAMtools (v.1.0) and Picard Tools (http://picard.sourceforge.net; v.1.92). Substitution and insertion or deletion (indel) calls were made using a GATK unified genotyper and GATK IndelGenotyperV2, respectively. All calls with a Phred-scaled single-nucleotide polymorphism quality of up to and including 20 and a read coverage of 2 or less were filtered out. All variants were annotated with annotation software developed in-house.

The STAT2 gene was sequenced in PBMCs of P3 as well as in fibroblasts of P1, P2, and their mother. Sanger sequencing was performed on an ABI 3730 XL Genetic Analyzer (Applied Biosystems) at the LGC Genomics Facility in Berlin, Germany. The regions of interest in Exon17 (P1 and P2) as well as Intron9 (P3) were amplified using the following primers: 5′-UM13-CTGCAGCTAGAAACATCA-3′ and 5′-UM13-AAAAGGAAATCTGTACCGAA-3′ (P1 and P2); 5′-UM13-GGAGAAAGGAGGTGCTGGATG-3′ and 5′-UM13-AACTTCCGGAAGCCTTGTAAT-3′ (P3).

Sequencing was performed with the universal UM13 primers. A control sample of an unrelated control was included for both sequencing runs.

Variants ClinVar accession numbers: SCV007595874 and SCV007595875.

Frozen PBMCs of P3 were available for analysis. After defrosting, 1 × 106 cells were used for the Sanger Sequencing (above), and a whole-cell extract was made in 100 µl NP40 buffer supplemented with protease inhibitors.

For P3 and a healthy control, T cell blasts were generated as follows: PBMCs were isolated by Ficoll-Hypaque density centrifugation (Amersham-Pharmacia-Biotech). For T-blast induction, PBMCs were cultured in ImmunoCult-XF T Cell Expansion Medium (STEMCELL) in the presence of ImmunoCultTM Human CD3/CD28/CD2 T cell activator (12.5 μl/ml) and human recombinant IL-2 (100 ng/ml, Novartis).

Primary fibroblasts were cultured in DMEM/F-12 (1:1) containing L-glutamine and HEPES supplemented with 10% FCS and 1% Pen-Strep and 0.1% normacin. Cells were incubated at 37°C under the atmosphere containing 5% CO₂. When the cells reached 80% confluency, the cells were trypsinized and split 1:4.

Fibrosarcoma cells deficient for STAT2 (U6A, kindly gifted by Prof. S. Boisson-Dupuis, Human Genetics of Infectious Diseases Laboratory, Rockefeller University, New York, NY, USA) were cultured in medium DMEM (1×) + GlutaMax-1 (cat #61965-026; Thermo Fisher Scientific) complemented with 10% FCS (cat #51818-500), Penicillin (100 IU/ml), and Streptomycin (100 µg/ml) (cat #15140-122; Thermo Fisher Scientific).

The splice variant identified in P3 was analyzed via the CloneJET PCR Cloning Kit (cat #K1232; Thermo Fisher Scientific) following the manufacturer’s instructions. In brief, the isolated RNA of P3 was reverse transcribed with Superscript Vilo cDNA synthesis kit (Thermo Fisher Scientific). The STAT2 gene of the relevant fragment was amplified as described above. Afterward, the amplified fragments were cloned into the pJET 1.2/blunt cloning vector and amplified in heat shock–transformed Escherichia coli (#C3040H; New England Biolabs). Purification of 182 clones was performed with the QIAprep Spin Miniprep kit (#27104; Qiagen) according to the manufacturer’s instructions. The sequences inserted were determined by Sanger sequencing with the manufacturers primers: 5′-CGA​CTC​ACT​ATA​GGG​AGA​GCG​GC-3′.

The coding sequence for STAT2 (NM_005419.4) was cloned into an untagged pCMV6 vector (OriGene). The Q5 Site-directed Mutagenesis Kit (New England Biolabs) was used to generate the indicated STAT2 variants. All constructs were re-sequenced (LGC Genomics) to ensure that no adventitious mutations were generated during cloning. The used primer sets are displayed in Table S1. In addition, each construct was expressed in HEK293T cells and analyzed via immunoblotting to ensure no damage to the plasmid’s backbone.

HEK293T- or STAT2-deficient U6A Fibrosarcoma cells were transfected with 100 ng of WT or mutant plasmid. Proteins of primary fibroblasts, PBMCs, and transfected cells were extracted in NP40 buffer supplemented with protein inhibitors. The lysates were run on 4–12% Tris-glycerine SDS-PAGE (Invitrogen), and the resulting bands were transferred to a polyvinyl difluoride membrane with 0.45-µm pores (Thermo Fisher Scientific), which was blocked with 5% bovine serum albumin in Tris-buffered saline. It was probed with unconjugated primary antibodies: STAT1 (sc-464; Santa Cruz Biotechnology, 1/500), pSTAT1 (9167S [Tyr701], Cell Signaling Technology 1/400), STAT2 (sc-514193; Santa Cruz Biotechnology, 1/400), pSTAT2 (88410; Cell Signaling Technology, 1/500), and HRP-conjugated secondary antibodies (goat anti-mouse [71045-3; Merck Life Science, 1/10,000], and mouse anti-rabbit [sc-2357; Sant Cruz Biotechnology 1/10,000]). An anti–β-actin antibody (Sigma-Aldrich, 1/10,000) was used as a loading control. SuperSignal West Pico Plus chemiluminescent substrate and Pierce ECL western blotting substrate (Thermo Fisher Scientific) were used to visualize HRP activity. Chemiluminescent signals were detected with a Bio-Rad Imager, and Image Lab 6.0.1 software was used for analysis.

RNA was extracted from primary fibroblasts (P1 and P2), frozen PBMCs (P3), or STAT2-deficient U6A fibrosarcoma cells (overexpression) that were collected in Trizol Reagent (Ambion). cDNA was generated with the SuperScript VILO cDNA synthesis kit (Thermo Fisher Scientific) according to the manufacturer’s protocols. qPCR for STAT2 was performed using SsoAdvanced SYBR Green Supermix (Bio-Rad) with a QuantStudio Real-Time PCR System (Thermo Fisher Scientific). The results are expressed according to the ΔCt method, with GAPDH as the housekeeping gene.

The induction of IFN genes was measured in patient derived primary fibroblasts of P1 and P2 as well as in the overexpression system of STAT2 deficient fibrosarcoma cells previously transfected with STAT2 wild-type and mutant plasmids as indicated. After transfection cells were rested for 48 h and stimulated with 10,000 U/ml IFN-α2A (cat #H6041; Merck), 1,000 U/ml IFN-γ (R&D Systems) for the indicated time points. Trizol was added to the cells and RNA was extracted as well as cDNA generated (see RNA analysis). The upregulation of IFN-induced genes, RSAD2, ISG15, Mx1, IFIT1, GBP1, and USP18 were analyzed via RT-qPCR. The used primer sets are displayed in Table S2. The results are displayed according to the ΔΔCt method, with GAPDH as the housekeeping gene. Data are normalized to the healthy control/WT conditions.

For measurement of the total cell expression of STAT2, control or patient PBMCs or T-blast cells were plated in 96-well plates, at a density of 5 × 105 cells per well, and surface-stained with D3 PE, CD14 FITC, and then permeabilized and stained intracellularly with pSTAT2 AF647 or pSTAT1 AF647 antibodies. The cells were then washed twice with PBS and analyzed by flow cytometry. Data were acquired on a Gallios flow cytometer, and the results were analyzed with FlowJo (Tree Star).

Fig. S1 shows clinical presentation of the patients; and Fig. S2 shows expression of phosphorylated STAT2 (pSTAT2), STAT2, pSTAT1, and STAT1 in lymphocytes derived from P3 and two healthy controls. Table S1 shows primers for site-directed mutagenesis; Table S2 shows primers for qPCR; and Table S3 shows list of variants identified after analysis of the WES of the P1 and P2.

The data underlying the figures are available in the published article and its online supplemental material.

We thank the patients and their families for placing their trust in us. We thank the members of both branches of the Laboratory of Human Genetics of Infectious Diseases. We thank Y. Nemirovskaya, M. Woollett, D. Liu, S. Boucherit, A. Geraldo, G. Goutbi, M. Chrabieh, and L. Lorenzo for administrative assistance.

I. Meyts is a senior clinical investigator at the Research Foundation Flanders (FWO) and is supported by the KU Leuven C1 Grant C16/18/007, by the FWO Grant G0B5120N, by the Horizon Europe Research Council (ERC) Undine grant 101057100, and by the Jeffrey Modell Foundation. I. Meyts is a recipient of the ERC starting grant MORE2ADA2. The Laboratory of Human Genetics of Infectious Diseases is supported by the Howard Hughes Medical Institute, the Rockefeller University, the St. Giles Foundation, the National Institutes of Health (NIH) (R01AI163029), the National Center for Advancing Translational Sciences, NIH Clinical and Translational Science Award program (UL1TR001866), the Fisher Center for Alzheimer’s Research Foundation, the Meyer Foundation, the JPB Foundation, the Stavros Niarchos Foundation (SNF) as part of its grant to the SNF Institute for Global Infectious Disease Research at The Rockefeller University, the French Agence Nationale de la Recherche (ANR) under the France 2030 program (ANR-10-IAHU-01), the Integrative Biology of Emerging Infectious Diseases Laboratory of Excellence (ANR-10-LABX-62-IBEID), the French Foundation for Medical Research (FRM) (EQU202503020018), the ANR-RHU program ANR-21-RHUS-0008, ANR GENVIR (ANR-20-CE93-003), ANR AABIFNCOV (ANR-20-CO11-0001) and ANR GenMISC (ANR-21-COVR-0039), AI2D (ANR-22-CE15-0046) projects, the European Union’s Horizon 2020 research and innovation program under grant agreement no. 824110 (EASI-genomics), the HORIZON-HLTH-2021-DISEASE-04 program under grant agreement 01057100 (UNDINE), the Square Foundation, Grandir - Fonds de solidarité pour l’enfance, the Fondation du Souffle, the SCOR Corporate Foundation for Science, the Battersea and Bowery Advisory Group; the French Ministry of Higher Education, Research, and Innovation (MESRI-COVID-19), William E. Ford, General Atlantic’s Chairman and Chief Executive Officer, Gabriel Caillaux, General Atlantic’s Co-President, Managing Director, and Head of Business in EMEA, and the General Atlantic Foundation, Institut National de la Santé et de la Recherche Médicale (INSERM), REACTing-INSERM, and Paris Cité University. P. Bastard. was supported by the FRM (EA20170638020), the MD-PhD program of the Imagine Institute (with the support of the Fondation Bettencourt-Schueller), and a “Poste CCA-INSERM-Bettencourt” (with the support of the Fondation Bettencourt-Schueller).

Author contributions: Verena Kienapfel: formal analysis, investigation, methodology, supervision, validation, visualization, and writing—original draft, review, and editing. Lotte Cresens: investigation. Lucy Bizien: investigation. Julia Vasconcelos: resources and writing—review and editing. Marwa Chbihi: resources. Margarida Guedes: investigation and validation. António Marinho: resources. Anneleen Hombrouck: investigation. Marjon Wouters: investigation and writing—review and editing. Dylan Laurens: investigation and writing—review and editing. Koji Nakajima: resource and writing—original draft, review, and editing. Jean-Laurent Casanova: funding acquisition, resources, supervision, and writing—review and editing. Jacinta Bustamante: investigation and resources. Shen-Ying Zhang: conceptualization, funding acquisition, investigation, supervision, and writing—review and editing. Paul Bastard: funding acquisition, investigation, resources, supervision, validation, and writing—review and editing. Leen Moens: investigation and visualization. Isabelle Meyts: conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, project administration, resources, software, supervision, validation, visualization, and writing—original draft, review, and editing. Giorgia Bucciol: conceptualization, supervision, visualization, and writing—original draft, review, and editing.