A Rare RIPK3 Variant Enhances Necroptosis and Promotes Inflammation in a Still Disease–Like Autoinflammatory Syndrome

Abstract

Objective

Still disease represents a prototypical polygenic systemic autoinflammatory disease, characterized by recurrent systemic inflammation and dysregulation of innate immunity. Despite extensive clinical characterization, familial clustering Still disease remains unreported. This study investigates a novel variant underlying a distinct Still disease–like autoinflammatory syndrome.

Methods

Whole‐exome sequencing was performed in a three‐generation family with three affected individuals. Functional analyses were conducted to assess the impact of the receptor‐interacting protein kinase 3 (RIPK3) p.Q134K variant on kinase activity, necroptosis, and inflammatory signaling. Transcriptomic profiling of peripheral blood mononuclear cells (PBMCs) and in vitro cytokine assays were used to characterize inflammatory responses.

Results

A heterozygous RIPK3 p.Q134K variant was identified in all affected individuals. Functional analyses demonstrated that this variant exhibits gain‐of‐function activity, characterized by enhanced RIPK3 kinase activity and increased RIPK3 self‐association, which in turn promote mixed‐lineage kinase domain‐like protein phosphorylation and necroptosis. RNA sequencing of PBMCs from one patient revealed marked up‐regulation of NF‐κB and MAPK signaling pathways and elevated expression of interleukin‐6, and tumor necrosis factor‐α, which was effectively suppressed by pharmacologic inhibition of RIPK3.

Conclusion

This study describes a familial Still disease–like autoinflammatory syndrome caused by a gain‐of‐function RIPK3 variant. Our findings establish aberrant RIPK3 activation as a driver of autoinflammation and expand the spectrum of human RIPK3‐associated diseases.

INTRODUCTION

Systemic autoinflammatory diseases (SAIDs) comprise a group of complex disorders characterized by recurrent systemic inflammation arising from dysregulation of the innate immune system. Still disease represents a prototypical polygenic SAID characterized by spiking fever, evanescent salmon‐colored rash, arthritis, and markedly elevated inflammatory cytokines. Unlike autoimmune diseases, which are mediated by adaptive immune responses and autoantibodies, SAIDs are primarily driven by genetic mutations that alter the encoded protein and disrupt innate immune regulation, leading to uncontrolled inflammatory responses. ¹ , ² These pathogenic gene mutations often affect key signaling pathways, including interleukin‐1β (IL‐1β), IL‐18, NF‐κB, and type I interferon (IFN) pathways, which in turn results in diverse clinical manifestations such as recurrent fever, rash, and organ damage. ³ , ⁴ , ⁵ Despite these advances, familial clustering Still disease has not been previously reported. Within the expanding spectrum of SAIDs, we identified a family with three affected members presenting a classic Still disease–like phenotype that did not meet the full classification criteria. This distinct presentation defines a novel autoinflammatory syndrome. ⁶

Building on these insights, >50 genes have now been identified as causative for SAIDs, many of which regulate inflammatory responses triggered by cell death pathways. ⁷ Among these, receptor‐interacting protein kinase 3 (RIPK3), a member of the RIPK family characterized by serine/threonine kinase activity, is best known for its essential role in mediating necroptosis, a regulated form of necrotic cell death. ⁸ , ⁹ In response to necroptotic stimuli, activated RIPK3 phosphorylates and facilitates the recruitment of mixed‐lineage kinase domain‐like protein (MLKL), leading to membrane rupture and the release of damage‐associated molecular patterns, thereby amplifying inflammation. ⁹ Dysregulated RIPK3 signaling has been implicated in the pathogenesis of multiple inflammatory diseases, including ischemic injury, tumor necrosis factor‐α (TNF‐α)–induced systemic inflammatory response syndrome, and organ‐specific inflammatory diseases affecting the liver, lungs, kidneys, and intestines. ¹⁰ , ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ Furthermore, pharmacologic inhibition of RIPK3 has been shown to attenuate excessive inflammation and mitigate lung damage in severe influenza infections, highlighting its potential as a therapeutic target in inflammatory disorders. ¹⁶

Despite increasing evidence demonstrating the critical role of cell death–related gene mutations in autoinflammation, including variations or deficiencies in RIPK1, disease‐associated RIPK3 variants remain undocumented in familial Still disease–like syndromes. Recent studies, including our work on the JAK‐STAT pathway in patients with refractory macrophage activation syndrome, have highlighted the importance of genetic variants in autoinflammatory phenotypes. ¹⁷ These observations raise the possibility that rare, functionally relevant variants in other key immune regulators, such as RIPK3, may underlie certain undiagnosed SAIDs.

PATIENTS AND METHODS

Study design

Whole‐exome sequencing (WES) was first performed on three patients with systemic inflammatory symptoms to identify potential pathogenic genes. Subsequently, we validated the function of the mutant by examining the expression and function of wild‐type (WT) or mutant RIPK3 proteins through transient transfection into HeLa cells as well as the expression and function of RIPK3 in primary peripheral blood mononuclear cells (PBMCs) from the patients. Finally, to investigate the specific pathogenic mechanisms of RIPK3 mutations, we examined the expression of known MAPK and NF‐κB pathways in HeLa cells transfected with the mutant and further evaluated the therapeutic potential of inhibitors of these pathways in the disease.

Patients

This study was conducted following the Declaration of Helsinki and Good Clinical Practice guidelines. Ethical approval was granted by the Institutional Research Ethics Committee of Ruijin Hospital in Shanghai, China (approval ID: 2023‐273), and the patient assessments were conducted in accordance with the approved protocol. All participants provided written informed consent before participating in the study.

WES

Genomic DNA was isolated from peripheral blood samples and evaluated for quality using agarose gel electrophoresis and quantified using a Qubit fluorometer. WES was conducted on affected individuals to identify potential pathogenic variants. Libraries were prepared through enzymatic or ultrasonic DNA fragmentation into 150‐ to 250‐bp fragments followed by end repair, adapter ligation, and polymerase chain reaction (PCR) amplification. Target regions were enriched using a liquid‐phase hybridization capture method with biotin‐labeled RNA probes specific to exonic regions. Enriched libraries were quantified and assessed for quality before high‐throughput sequencing on a NovaSeq 6000 platform (Illumina). Data analysis included quality filtering of raw reads, alignment to the reference genome (GRCh38), and annotation of variants.

Sanger sequencing

Candidate variants detected by WES were confirmed through Sanger sequencing. Corresponding primers were employed to amplify the genomic regions flanking the variants via PCR. The PCR products were then purified and subjected to bidirectional sequencing on an ABI 3730 DNA Analyzer. Phred/Phrap software was used to analyze the sequence data, confirming the presence of the variants and their segregation within the family.

RNA sequencing

PBMCs were isolated using Polymorphprep (1895, Serumwerk Bernburg AG) for RNA extraction. Messenger RNA was enriched using oligo(dT) magnetic beads, fragmented, and complementary DNA synthesis with random hexamer primers. Libraries were constructed, circularized, amplified into DNA nanoballs, and sequenced on the BGISEQ G400 platform (BGI‐Shenzhen). Differentially expressed genes were identified using DESeq2 (|log2 fold change| > 1, P < 0.05). Pathway enrichment was performed using KEGG, and gene set enrichment analysis (GSEA) was conducted. Data visualization was generated using ggplot2.

Cell culture and stimulation

Hela cells were obtained from the ATCC. The total white blood cells (WBCs) were acquired by processing peripheral blood with a red blood cell lysis buffer (G2015, Servicebio) following the manufacturer's protocol. HeLa cells were maintained in Dulbecco Modified Eagle Medium (Servicebio) supplemented with 10% fetal bovine serum (FBS; S711, Lonsera) and penicillin‐streptomycin (G4003, Servicebio). PBMCs and WBCs were cultured in RPMI 1640 medium containing 10% FBS and the same concentration of penicillin‐streptomycin. For stimulation, cells were treated with TNF‐α (30001A, PeproTech), SM164 (S7089, Selleck), Z‐VAD‐FMK (S7023, Selleck), and GSK872 (S8465, Selleck) under appropriate conditions as specified for each experimental setup.

Plasmid construction and transfection

Human WT RIPK3 and the K50A, Q134K, Q134H, and Q134R mutants were cloned into the cytomegalovirus promoter–based vector. Site‐directed mutagenesis was then used to generate the mutant plasmids. After plasmids were confirmed by Sanger sequencing, transfection was conducted using Lipofectamine 3000 (L3000, Thermo Fisher Scientific) as per the manufacturer's protocol.

Western blotting

Hela cells and PBMCs were stimulated as described above and then lysed with radioimmunoprecipitation assay lysis buffer (P0013, Beyotime) containing phosphatase inhibitor cocktails (4906845001, Roche) and protease inhibitor cocktails (4693116001, Roche). The separation of nuclear and cytoplasmic proteins was performed using the Nuclear and Cytoplasmic Protein Extraction Kit (KGB5302, Keygen Biotech) following the manufacturer's protocol. Protein concentration was measured using a BCA protein assay kit (ZJ101, Epizyme). Immunoblotting was performed using primary antibodies diluted at 1:1,000, unless otherwise indicated, following previously described methods. ¹⁸ Antibodies against GAPDH (GB11002) and Flag (AB3058) were purchased from Servicebio and Beyotime, respectively. Other antibodies listed below were obtained from Cell Signaling Technology: MLKL (14993), phospho‐MLKL (91689), RIPK3 (13526), phospho‐RIPK3 (93654), phospho‐p38 (4511), p38 (8690), phospho‐ERK (4370), ERK (4695), phospho‐JNK (4668), JNK (9258), NFκB p65 (8242), phospho‐NFκB p65 (3033), and RIPK1 (14993).

Lactate dehydrogenase release assay

The lactate dehydrogenase (LDH) release assay was conducted using the LDH Cytotoxicity Assay Kit (C0016, Beyotime). Cells were treated as indicated, and 10% LDH release reagent was added to control wells 1 hour before measurement. Plates were centrifuged at 400g for 5 minutes, and 120 μL of the supernatants was transferred to a new 96‐well plate. After addition of 60 μL of LDH detection solution, samples were incubated at room temperature in the dark for 30 minutes, and absorbance was measured at 490 nm.

Cell death assay

An Annexin V‐FITC/PI Apoptosis Detection Kit (KGA1102, Keygen Biotech) was used to evaluate cell death. After the aforementioned treatments, cells were resuspended in 500 μL of binding buffer, then incubated with 5 μL of FITC‐conjugated Annexin V and 5 μL of PI for 10 minutes at room temperature in the dark. Apoptotic and necrotic cells were subsequently analyzed using a FACS Calibur flow cytometer (BD Biosciences).

Cytokine detection

Plasma levels of IFN‐α, IFN‐γ, IL‐1β, IL‐2, IL‐4, IL‐5, IL‐6, IL‐8, IL‐10, IL‐12p70, IL‐17, IL‐18, and TNF‐α were determined using a cytometric bead array (BD Biosciences) following the manufacturers’ instructions. Additionally, the concentrations of TNF‐α, IL‐1β, and IL‐6 in the cell supernatants were quantified by enzyme‐linked immunosorbent assay (DY210, DY201, and DY206, R&D systems), according to the respective manufacturer's guidelines.

ADP‐Glo kinase assay

The kinase activity of WT and Q134K mutant RIPK3 was assessed using the ADP‐Glo Kinase Assay (Promega, V6930) according to the manufacturer's instructions. Recombinant human WT RIPK3 (MCE, HY‐P71855), Q134K RIPK3 (custom synthesized, MCE, HYP7S2531), and MLKL protein (MCE, HY‐P702574) were reconstituted in kinase reaction buffer (40 mM Tris‐HCl, pH 7.5; 20 mM MgCl₂; 0.1 mg/mL BSA) with ATP (20 μM) at room temperature for 1 hour. The reaction was terminated with ADP‐Glo Reagent, followed by Kinase Detection Reagent, and luminescence was measured using a microplate luminometer.

Blue native Western blotting

To assess whether the RIPK3 Q134K mutation affects RIPK3 self‐dimerization, HeLa cells were stimulated as described above and lysed on ice for 15 minutes in a native lysis buffer containing 5% Digitonin (BN20061, Invitrogen). Lysates were clarified by centrifugation at 20,000g for 30 minutes at 4°C. Equal amounts of protein were then mixed with nondenaturing loading buffer (BN20041, Invitrogen) and subjected to blue native PAGE using precast gels (P0546S, Beyotime). Proteins were subsequently transferred to PVDF membranes and analyzed by immunoblotting using standard procedures.

Co‐immunoprecipitation

To examine the interaction between RIPK1 and RIPK3 in WT and Q134K mutant, clarified cell lysates were incubated with anti‐Flag magnetic beads (90 μL, resuspended in 1 × Tris‐buffered saline [TBS]; Beyotime) overnight at 4°C on a rotator. Beads were captured using a magnetic rack and washed four times with 1 × TBS. Bound proteins were eluted in 1× SDS‐PAGE sample buffer by heating at 95°C for 5 minutes and subsequently analyzed by Western blotting.

Dual‐luciferase reporter assay

NF‐κB– and AP‐1–dependent transcriptional activity was assessed using dual‐luciferase reporter assays. HeLa cells were cotransfected with pNFκB‐TA‐luc (D2207, Beyotime), pAP1‐TA‐luc (D2109, Beyotime), and pRL‐TK (D2760, Beyotime) in 96‐well plates and treated as indicated. Following stimulation, cells were lysed with Dual‐Lumi Firefly Luciferase Reporter Cell Lysis Buffer (RG126S, Beyotime), and luciferase activity was measured using the Dual‐Lumi Luciferase Reporter Assay Kit (RG088S, Beyotime) according to the manufacturer's instructions. Firefly luciferase activity was normalized to Renilla luciferase activity and expressed as relative luciferase units.

Statistical analysis

Statistical analysis was performed using R v4.3 or GraphPad Prism v9.0. Quantitative data are presented as the mean ± SD. For data with a Gaussian distribution, an unpaired two‐sided t‐test or one‐way analysis of variance was used, and nonparametric data were analyzed with the Mann‐Whitney U test. A P value <0.05 was considered statistically significant. All experiments were performed with at least three independent biologic replicates. Each experiment included three technical replicates. The data that support the findings of this study are available from the corresponding author upon reasonable request.

RESULTS

Case presentations and identification of the RIPK3 variant

All three patients (Figure 1A; Table 1) exhibited a spectrum of Still disease–like symptoms. The proband (P1), a 34‐year‐old female, presented with recurrent fever, rash, arthritis, and diarrhea. Laboratory tests revealed elevated C‐reactive protein (117 mg/L, normal range <10 mg/L) and a raised erythrocyte sedimentation rate (75 mm/h, normal range <20 mm/h). Autoimmune‐related antibody test results, including antinuclear antibody (ANA), extractable nuclear antigen (ENA), antineutrophil cytoplasmic antibody, rheumatoid factor (RF), and cyclic citrullinated peptide antibody, were all negative. Comprehensive imaging studies and bone marrow evaluation excluded infection, malignancy, and hematologic disorders. P2, a 57‐year‐old male and the father of P1, experienced similar symptoms of recurrent fever, rash, arthritis, diarrhea, and myalgia at the age of 50 years. Laboratory findings included an increased erythrocyte sedimentation rate and an ANA titer of 1:320, whereas ENA and RF results were negative. Imaging examinations revealed diffuse lymphadenopathy. Both P1 and P2 received methylprednisolone treatment, resulting in marked alleviation of symptoms. P3, a 9‐year‐old male and the son of P1, developed a fever of unknown origin refractory to antimicrobial therapy but responsive to glucocorticoids at the age of 7 years.

Figure 1.

A rare heterozygous variant of RIPK3 identified in a family with Still disease–like autoinflammatory syndrome. (A) Pedigrees of the family with variants in RIPK3. Black solid symbols represent individuals affected with a hemizygous RIPK3 variant. Genotypes are indicated below the symbols. Squares indicate male family members and circles female members. Roman numerals indicate generations. (B) Whole‐exome sequencing data‐filtering approach under the assumption of genetic homogeneity. (C) Schematic showing the protein domain of RIPK3 with the position of Q134K marked. Pathogenic variants associated with viral encephalitis (shown in black font) are included for comparison. (D) Conservation analysis of the RIPK3 p.Gln134 residue across species. Amino acids are colored according to physicochemical properties. (E) The structure of ATP interacting with RIPK3 and expanded view of the effects of WT and Q134K mutation on ATP and RIPK3 interactions. Protein and small‐molecule interaction structure predictions were performed using AlphaFold3, and structural analysis was executed with PyMOL. Light blue represents ATP, dark blue represents amino acids in RIPK3 protein interacting with ATP, yellow represents hydrogen bonds, and red represents amino acid position 134 in RIPK3 protein. *Genes associated with autoimmunity and autoinflammation. MAF, minor allele frequency; P, patient; RHIM: RIP homotypic interaction motif; RIPK3, receptor‐interacting protein kinase 3; SNP, single‐nucleotide polymorphism; WT, wild‐type.

Table 1.

Baseline characteristics of the three patients affected by a Still disease–like syndrome^*

Characteristic	P1	P2	P3
Sex	Female	Male	Male
Age at onset, y	34	50	8
Fever >39°C	Y	Y	Y
Rash	Y	Y	N
Sore throat	Y	Y	N
Arthralgia	Y	Y	N
Pleuritis	N	N	N
Pneumonia	N	N	N
Pericarditis	N	N	N
Hepatomegaly	N	N	N
Splenomegaly	N	N	N
Lymphadenopathy	Y	Y	N
Myalgia	N	Y	N
Diarrhea	Y	Y	N
ANA	Neg	1:320	–
RF	Neg	Neg	–
ACPA	Neg	Neg	–
ALT, U/L	20	130	–
AST, U/L	22	103	–
ESR, mm/h	75	28	–
CRP, mg/L	117	17	–
Ferritin, ng/mL	>1,500	391.4	–

^*ACPA, anticitrullinated protein antibody; ALT, alanine aminotransferase; ANA, antinuclear antibody; AST, aspartate aminotransferase; CRP, C‐reactive protein; ESR, erythrocyte sedimentation rate; N, no; Neg, negative; P, patient; RF, rheumatoid factor; Y, yes.

We performed WES on the three patients and their family members. By filtering for mutation quality and priority under the assumption of genetic homogeneity, we identified a heterozygous mutation in RIPK3 (c.400C>A, p.Gln134Lys) in all three patients (Figure 1A and B; Supplementary Table S1). The variant was extremely rare in population databases (gnomAD v4.1.0 minor allele frequency = 6.2 × 10⁻⁷) and absent in the East Asian subpopulation. In silico analyses predicted a deleterious effect (Combined Annotation Dependent Depletion 12.2; PolyPhen “probably damaging”; Functional Analysis through Hidden Markov Models “deleterious”; missense tolerance ratio = 0.93), supporting functional constraint (Table 2). Segregation was confirmed by Sanger sequencing, which detected the variant exclusively in affected family members (Supplementary Figure S1A).

Table 2.

RIPK3 mutation identified in Still disease–like autoinflammatory syndrome family^*

Gene	Chromosome	Position	Consequence	MAF a	MAF_eas	CADD	Polyphen	FATHMM	ChinaMAP	MTR
RIPK3	14	24,339,086	c.400C>A:p.Gln134Lys	0.00000062	0	12.2	Probably damaging	Deleterious	0	0.93

^*CADD, Combined Annotation Dependent Depletion; FATHMM, Functional Analysis through Hidden Markov Models; MAF, minor allele frequency; MAF_eas, minor allele frequency in the East Asian subpopulation; MTR, missense tolerance ratio; RIPK3, receptor‐interacting protein kinase 3.

^a MAF in gnomAD v4.1.0.

The p.Q134K variant localizes to the kinase domain of RIPK3 (Figure 1C). Evolutionary conservation analysis revealed that residue Q134 is relatively conserved across species but not across the RIPK family, suggesting a RIPK3‐specific regulatory role (Figure 1D; Supplementary Figure S1B). AlphaFold3 modeling predicted that the Q134K mutation alters the local ATP‐binding environment and hydrogen‐bonding network, indicating a potential effect on ATP‐dependent kinase activity (Figure 1E).

The RIPK3 Q134K mutation enhances RIPK3 kinase activity and sensitizes HeLa cells to TSZ‐mediated necroptosis

To investigate the functional impact of the RIPK3 Q134K mutation, we transfected HeLa cells—which lack endogenous RIPK3 expression—with plasmids encoding WT, kinase‐dead mutant K50A, or Q134K RIPK3. Considering the critical role of the RIPK3 kinase domain in necroptosis, we induced necroptosis in HeLa cells by cotreatment with TNF‐α, second mitochondrial‐derived activator of caspases mimetic SM‐164, and the pancaspase inhibitor Z‐VAD‐FMK (TSZ). Notably, compared with the WT, the Q134K mutant exhibited significantly elevated phosphorylated RIPK3 (p‐RIPK3) and phosphorylated MLKL (p‐MLKL, a biomarker for necroptosis) levels in HeLa cells (Figure 2A and B). Furthermore, these elevated phosphorylation levels were effectively suppressed by the low concentrations of RIPK3 inhibitor GSK872 (Figure 2C and D), confirming RIPK3 dependency. Importantly, ADP‐Glo assay demonstrated that Q134K RIPK3 exhibited higher kinase activity in the presence of the MLKL substrate compared with WT RIPK3 (Figure 2E), indicating a gain‐of‐function effect at the kinase level.

Figure 2.

Mutant RIPK3 plasmids induce increased cell death in transfected Hela cells. (A) Western blots of Hela cells transfected with an EV WT RIPK3, kinase‐dead K50A, or Q134K RIPK3 following TSZ stimulation for the indicated times. (B) Densitometric analysis of p‐RIPK3/total RIPK3 and p‐MLKL/total MLKL shown in Figure 2A. Statistical comparisons were made between Q134K and WT groups at the indicated time points following TSZ stimulation. (C) Western blots of Hela cells transfected with the indicated RIPK3 constructs treated with TSZ, with or without GSK872 for 1 hour. (D) Densitometric analysis of p‐RIPK3/total RIPK3 and p‐MLKL/total MLKL in Figure 2C. (E) In vitro RIPK3 kinase activity measured by ADP‐Glo assay. (F) Blue native PAGE analysis showing RIPK3 oligomerization after TSZ stimulation. (G) Co‐immunoprecipitation analysis of RIPK3‐RIPK1 interaction following TSZ stimulation. (H) Flow cytometric analysis of Annexin V⁺/PI⁺ populations in WT‐ and Q134K RIPK3–expressing HeLa cells. (I) LDH release was measured in supernatants of Hela cells transfected with either WT RIPK3 or the Q134K mutant RIPK3 and stimulated with TSZ and/or GSK872 for 6 hours. (J) IL‐6 and TNF‐α levels in the supernatants of HeLa cells 4 hours after TSZ stimulation. *p < 0.05; ***p < 0.001; ****p < 0.001. EV, empty vector; IL‐6, interleukin‐6; LDH, lactate dehydrogenase; MLKL, mixed‐lineage kinase domain‐like protein; ns, not significant; p‐, phosphorylated; RIPK3, receptor‐interacting protein kinase 3; TNF‐α, tumor necrosis factor α; WT, wild‐type. Color figure can be viewed in the online issue, which is available at http://onlinelibrary.wiley.com/doi/10.1002/art.70146/abstract.

Missense variants at residue 134 (Q134H and Q134R) are present in population databases without reported disease associations. To assess the functional sensitivity of this site, we generated Q134H and Q134R mutants. Upon TSZ stimulation, both mutants induced p‐RIPK3 and p‐MLKL more strongly than WT but less than Q134K, indicating an intermediate activation phenotype (Supplementary Figure S2A). These differences likely reflect distinct physicochemical properties of the substituted residues that differentially modulate local kinase activation.

To further investigate the mechanism of the RIPK3 Q134K variant, we examined RIPK3 dimerization. Blue native PAGE revealed enhanced RIPK3 homodimerization in Q134K‐expressing cells compared with WT, particularly upon TSZ stimulation (Figure 2F). In contrast, co‐immunoprecipitation showed comparable RIPK3–RIPK1 interactions between WT and Q134K (Figure 2G). Consistently, inhibition of RIPK1 with necrostatin‐1s only modestly reduced the levels of p‐RIPK3 and p‐MLKL in Q134K‐expressing cells (Supplementary Figure S2B). Together, these data indicate that the Q134K mutation preferentially enhances RIPK3 self‐association, thereby promoting RIPK3 activation independently of RIPK1.

Subsequently, Annexin V/PI staining was performed to examine cell death. Following TSZ treatment, the Q134K mutant group exhibited a significant increase in Annexin/PI double‐positive late apoptotic/necrotic cells compared with WT, which was markedly reduced by GSK872 (Figure 2H). In contrast, the Annexin V⁺PI⁻ (early apoptotic) population showed variable changes across conditions (Supplementary Figure S2D). This variability may reflect the transient nature of early apoptosis under strong necroptotic stimulation. To further assess cell necroptosis, we conducted an LDH release assay. Similarly, the mutant group showed significantly higher LDH release, which was markedly reduced by GSK872 (Figure 2I). Because activation of necroptosis is known to induce a strong inflammatory response, including the release of pro‐inflammatory cytokines such as IL‐6, ¹⁹ we measured IL‐6 levels in the supernatant of HeLa cells. The mutant group exhibited significantly higher IL‐6 secretion in response to TSZ stimulation compared with the WT group, which was effectively inhibited by GSK872 (Figure 2J). In addition, we also measured the levels of TNF‐α in the supernatant of HeLa cells to explore whether the Q134K mutation leads to increased TNF‐α release following necroptosis, thereby establishing a positive feedback loop that further promotes necroptosis. Consistent with our hypothesis, the Q134K mutant group exhibited significantly higher TNF‐α levels in the supernatant after TSZ stimulation in comparison with the WT group (Figure 2J). In summary, the reduction of inflammatory signaling through RIPK3 inhibition further supports that the Q134K mutation enhances the sensitivity of HeLa cells to TSZ‐induced necroptosis, thereby exacerbating inflammation.

Enhanced necroptosis response of PBMCs in the patient

Subsequently, we examined the inflammatory response in the patient. A marked increase in the levels of pro‐inflammatory cytokines, including TNF‐α, IL‐1β, IL‐6, and IL‐8 was detected in the patients’ serum, especially in P3 (Figure 3A; Supplementary Figure S3). Consistent with the findings observed in HeLa cells, PBMCs from P3 treated with TSZ showed significantly elevated levels of p‐RIPK3 and p‐MLKL compared with those of healthy controls, both of which were effectively suppressed by the RIPK3 inhibitor GSK872 (Figure 3B and C). Annexin V/PI staining and LDH release assays were subsequently conducted to assess cell death in PBMCs. Upon TSZ stimulation, PBMCs from P3 showed a marked increase in Annexin V⁺PI⁺ late apoptotic/secondary necrotic cells compared with healthy controls, whereas the Annexin V⁺PI⁻ (early apoptotic) population did not show a consistent increase (Figure 3D; Supplementary Figure S4B). Quantification of total Annexin V–positive cells revealed a concordant increase following TSZ stimulation, which was effectively suppressed by GSK872 (Supplementary Figure S4C). In parallel, LDH release was significantly elevated in P3 PBMCs and reduced by GSK872 (Figure 3E). Additionally, we quantified the levels of TNF‐α, IL‐1β, and IL‐6 in the supernatants of PBMCs stimulated with TSZ. The results revealed that PBMCs from the patient secreted significantly higher levels of TNF‐α, IL‐1β, and IL‐6 when contrasted with those of healthy individuals. Importantly, GSK872 effectively suppressed the secretion of these cytokines (Figure 3F). These findings provide additional evidence that RIPK3 activity is enhanced in the patient's immune cells, with the Q134K variant potentially contributing to RIPK3‐associated necroptosis.

Figure 3.

Patient PBMCs are sensitive to TSZ‐induced necroptosis. (A) Serum levels of cytokines TNF‐α, IL‐1β, IL‐6, and IL‐8 in three patients and six HCs. Plasma samples from P1 and P2 were collected under treatment, whereas plasma from P3 was collected at baseline. (B) Western blots of PBMCs from P3 and HCs stimulated with TSZ for 2 hours in the presence or absence of GSK872. (C) Densitometric analysis of p‐RIPK3/total RIPK3 and p‐MLKL/total MLKL in Figure 3B. (D) Flow cytometric analysis of Annexin V/PI staining PBMCs treated as indicated, showing Annexin V⁺/PI⁺ populations. (E) LDH release was measured in PBMC supernatants stimulated with TSZ for 6 hours, with or without GSK872. (F) The concentrations of IL‐1β, IL‐6, and TNF‐α in the supernatants of PBMCs at 4 hours after specific stimulation. *p < 0.05; ****p < 0.0001. HC, healthy control; IL, interleukin; LDH, lactate dehydrogenase; MLKL, mixed‐lineage kinase domain‐like protein; ns, not significant; p‐, phosphorylated; P, patient; PBMC, peripheral blood mononuclear cell; RIPK3, receptor‐interacting protein kinase 3; TNF‐α, tumor necrosis factor α. Color figure can be viewed in the online issue, which is available at http://onlinelibrary.wiley.com/doi/10.1002/art.70146/abstract.

RIPK3 variant promotes inflammation through the NF‐κB and MAPK signaling pathways

To gain additional insight into the molecular mechanisms underlying the inflammatory phenotype observed in the patient's PBMCs, RNA sequencing was performed on PBMCs from P3 and four healthy individuals. KEGG analysis showed significant enrichment of inflammatory pathways, most prominently the MAPK signaling pathway, along with the IL‐17 signaling pathway, Toll‐like receptor signaling pathway, and TNF signaling pathway (Figure 4A). Notably, IL‐17 pathway enrichment was driven mainly by downstream NF‐κB/MAPK target genes (eg, FOS, FOSB, JUN, JUND, IL1B, TNF, and TRAF4), whereas canonical Th17 differentiation markers (RORC, IL17A, and IL21) were not differentially expressed. Considering the pivotal role of the NF‐κB pathway in necroptosis, GSEA was performed. Notably, a significant up‐regulation of the NF‐κB signaling pathway was observed in the patient's PBMCs (Figure 4B). Consistently, volcano plot analysis highlighted marked up‐regulation of NF‐κB– and MAPK‐related genes (Figure 4C). In contrast, IFN‐stimulated genes were not globally up‐regulated, and no IFN‐related pathways were enriched (Supplementary Figure S5). Together, these results suggest that the Q134K mutation may trigger activation of the NF‐κB and MAPK pathways, thereby amplifying inflammatory responses.

Figure 4.

RIPK3 Q134K variant induces inflammation through the NF‐κB and MAPK signaling pathways. (A) KEGG enrichment analysis of up‐regulated DEGs in P3 PBMCs compared with HCs. (B) Gene set enrichment analysis plot showing enrichment of the NF‐κB pathway in P3 versus HCs. (C) Volcano plot showing DEGs in PBMCs from P3 compared with HCs. (D) NF‐κB and AP‐1 luciferase reporter activity in HeLa cells expressing EV, WT, or Q134K RIPK3 following TSZ stimulation. (E) Luciferase reporter activity in HeLa cells treated with TSZ in the presence or absence of GSK872. (F) Western blots of Hela cells transfected with the indicated RIPK3 constructs treated with TSZ for the indicated times. (G) Western blots of Hela cells transfected with the indicated RIPK3 constructs treated with TSZ with or without GSK872 at 0.5 hours after specific stimulation. (H and I) The concentrations of IL‐6 and TNF‐α in the supernatants of Hela cells at 4 hours after specific stimulation. Statistical comparisons were performed relative to the TSZ group. *p < 0.05; **p < 0.01; ****p < 0.0001. DEG, differentially expressed gene; EV, empty vector; HC, healthy control; IL, interleukin; NES, normalized enrichment score; ns, not significant; P, patient; PBMC, peripheral blood mononuclear cell; RIPK3, receptor‐interacting protein kinase 3; TNF‐α, tumor necrosis factor α; WT, wild‐type. Color figure can be viewed in the online issue, which is available at http://onlinelibrary.wiley.com/doi/10.1002/art.70146/abstract.

Consistent with the transcriptomic findings, NF‐κB– and AP‐1–dependent luciferase reporter assays showed that the Q134K variant induced significantly higher NF‐κB and AP‐1 reporter activity compared with WT RIPK3, which was markedly suppressed by the GSK872 (Figure 4D and E). Immunoblot analysis further revealed increased phosphorylation of p65, p38, ERK, and JNK in Q134K‐expressing cells following TSZ stimulation, and this effect was attenuated by GSK872 (Figure 4F and G). Moreover, inhibition of MAPK signaling (p38, ERK, or JNK) or treatment with prednisolone or dexamethasone significantly reduced IL‐6 and TNF‐α secretion in HeLa cells (Figure 4H and I). These findings not only further corroborate the role of the RIPK3 Q134K mutation in the activation of the MAPK and NF‐κB pathways but also align with the clinical efficacy of glucocorticoid therapy in the patient. This therapeutic effect may be attributed to the inhibitory action of prednisolone and dexamethasone on the NF‐κB signaling pathway. ²⁰ , ²¹

DISCUSSION

In this study, we report a family case of a rare RIPK3 variant associated with Still disease–like autoinflammatory syndrome and demonstrate that this variant exerts a gain‐of‐function effect on RIPK3 kinase activity. Functional analyses revealed that this mutation enhances RIPK3 homodimerization, MLKL activation, and necroptosis, subsequently activating the MAPK and NF‐κB signaling pathways. This, in turn, promotes the production of proinflammatory cytokines, such as IL‐1β, IL‐6, and TNF‐α, thereby establishing a positive feedback loop of cell death and inflammation. Our findings provide the first evidence linking a human RIPK3 genetic variant to Still disease–like manifestations, suggesting a potential pathogenic mechanism involving enhanced RIPK3‐driven necroptotic signaling and amplified inflammatory responses.

RIPK1 and RIPK3, both members of the RIPK family, share critical roles in pathways mediating inflammation and necroptosis through similar signaling cascades. During necroptosis, RIPK3 acts as a crucial downstream effector of RIPK1. Given their close functional relationship, we hypothesize that diseases caused by pathogenic variants in RIPK1 and RIPK3 may exhibit some overlapping features. Currently, two major categories of disorders associated with RIPK1 pathogenic variants have been reported: RIPK1 deficiency with autoinflammation ²² , ²³ and cleavage‐resistant RIPK1‐induced autoinflammatory syndrome. ²⁴ , ²⁵ These disorders share certain clinical features with the condition described here, such as recurrent fever and systemic inflammation, suggesting partial phenotypic overlap among RIPK family–associated diseases. However, our mechanistic data indicate that the RIPK3 Q134K variant acts through a pathway distinct from RIPK1‐driven disease mechanisms. We found that the Q134K variant neither enhanced RIPK3‐RIPK1 heterodimerization nor rendered RIPK3 or MLKL phosphorylation sensitive to RIPK1 inhibition. Instead, it preferentially promoted RIPK3 self‐association and kinase activation, indicating a RIPK3‐intrinsic mechanism of signal amplification. Together, our results refine the current framework of RIPK‐associated autoinflammatory diseases and highlight that perturbations at different nodes of the necroptotic pathway can give rise to overlapping yet mechanistically distinct clinical phenotypes.

Hitherto, pathogenic mutations in >50 genes have been implicated in the development of SAIDs. ⁷ Several SAIDs and inborn errors of immunity (IEIs) have been linked to defects in necroptotic signaling, including RIPK1 deficiency, ²² RIPK1 cleavage resistance, ²⁴ , ²⁵ linear ubiquitin chain assembly complex deficiency, ²⁶ , ²⁷ , ²⁸ ovarian tumor deubiquitinase with linear linkage specificity deficiency, ²⁹ and TBK1 deficiency, ³⁰ giving rise to a broad clinical spectrum ranging from early‐onset arthritis and systemic vasculitis to organ‐specific inflammation and immune dysregulation. Increasing evidence also supports a genetic contribution to Still disease, suggesting a shared genetic basis with Still disease–like autoinflammatory syndromes. ³¹ In this study, analysis based on the updated 2024 International Union of Immunological Societies classification did not identify additional variants in known IEI‐ or SAID‐associated genes beyond RIPK3 p.Q134K, supporting a primary role for aberrant RIPK3 signaling. In terms of treatment, it remains individualized because of the absence of standardized guidelines. High‐dose glucocorticoids are often effective, and biologic therapies targeting IL‐1β, IL‐6, and TNF‐α pathways are increasingly used. Additionally, hematopoietic stem cell transplantation has demonstrated partial efficacy in certain cases. ³² In our study, the patients exhibited systemic inflammation with elevated cytokine production, consistent with excessive cytokine release downstream of RIPK3‐driven necroptotic signaling and supporting the idea that cytokine‐targeted therapies may represent rational therapeutic options.

Previous studies of RIPK3 mutations predominantly investigate their functional impacts and pro‐inflammatory mechanisms, where these genetic alterations are typically introduced experimentally in cellular or animal models. To date, only one report has associated RIPK3 mutations (RIPK3 R422* and P493fs9*) with human disease, which impair cell death‐dependent regulation of herpes simplex virus‐1 in cortical neurons independent of type I IFN immunity, thereby increasing susceptibility to herpes simplex encephalitis. ³³ In contrast, our study associates a RIPK3 variant with Still disease–like autoinflammatory phenotypes. Although circulating IFN‐α and IFN‐γ levels showed a modest upward trend, they remained within the normal range, and transcriptomic analysis revealed no global induction of IFN‐stimulated genes, indicating that the p.Q134K variant primarily promotes necroptotic and inflammatory signaling rather than IFN dysregulation. Together, our findings provide novel genetic and mechanistic insights into the pathogenesis of SAIDs and highlight the context‐dependent effects of distinct RIPK3 variants on immune signaling pathways.

RIPK3 is broadly expressed across various immune cell types, exhibiting low immune cell specificity. Beyond immune cells, it is also expressed in a variety of tissues and organs, including the gastrointestinal tract, spleen, kidneys, bone marrow, lymphoid tissues, skin, and soft tissues (https://proteinatlas.org/). Given its ubiquitous expression, RIPK3 variants may affect multiple organ systems throughout the body, potentially accounting for the systemic manifestations observed in our patients. Moreover, the infiltration of activated immune cells into different tissues may further lead to the heterogeneity of clinical presentations.

A limitation of this study is that comprehensive segregation analysis within the extended family could not be completed, as the genotype of the proband's healthy sister was unavailable. Therefore, incomplete penetrance or variable expressivity associated with the RIPK3 p.Q134K variant cannot be formally excluded. Additional cases will be required to further define the penetrance and phenotypic spectrum of RIPK3‐associated autoinflammatory disease.

In summary, our study identifies a familial clustering of a Still disease–like autoinflammatory syndrome associated with a rare variant in the RIPK3 kinase domain. This variant enhances RIPK3 kinase activity, leading to aberrant necroptotic and inflammatory signaling, thereby providing mechanistic insight into the pathogenesis of Still disease–like inflammation. Accordingly, targeting RIPK3 and its downstream MAPK and NF‐κB pathways, together with cytokine blockade, may represent a rational therapeutic strategy for RIPK3‐associated autoinflammatory diseases.

AUTHOR CONTRIBUTIONS

All authors contributed to at least one of the following manuscript preparation roles: conceptualization AND/OR methodology, software, investigation, formal analysis, data curation, visualization, and validation AND drafting or reviewing/editing the final draft. As corresponding authors, Drs Yang, Jia, and Hu confirm that all authors have provided the final approval of the version to be published and takes responsibility for the affirmations regarding article submission (eg, not under consideration by another journal), the integrity of the data presented, and the statements regarding compliance with institutional review board/Declaration of Helsinki requirements.

Supporting information

Disclosure Form:

Supplementary Figure S1. Sanger sequencing and conservation analysis of the RIPK3 p.Gln134Lys variant. (A) Confirmation of RIPK3 variants at residue Gln134 for patients P1‐P3 and their family members by Sanger sequencing. (B) Conservation analysis of the p.Gln134 residue across the RIPK protein family, with amino acids color‐coded by physicochemical properties.

Supplementary Figure S2: Mechanistic characterization of RIPK3 Q134 variants. (A) Immunoblot analysis of HeLa cells expressing EV, WT RIPK3, or the indicated Q134 variants (Q134K, Q134H, Q134R) following TSZ stimulation. (B) Immunoblot analysis of HeLa cells treated with TSZ in the presence or absence of the RIPK1 inhibitor Nec1s. (C) Flow cytometry gating strategy in Hela cells. (D) Quantification of Annexin V⁺PI^‐ (early apoptotic) cells in Hela cells under the indicated treatment conditions. Data are shown as mean ± SEM.

Supplementary Figure S3: Serum levels of cytokines IFN‐α, IFN‐γ, IL‐2, IL‐4, IL‐5, IL‐10, IL‐12p70, and IL‐17 in three patients and 6 healthy controls (HCs).

Supplementary Figure S4. Flow cytometric analysis of Annexin V/PI staining in human blood. (A) Flow cytometry gating strategy in human blood. (B) Quantification of Annexin V⁺PI^‐ (early apoptotic) cells in PBMCs from P3 and healthy controls (HCs) under the indicated treatment conditions. Data are shown as mean ± SEM. (C) Quantification of total Annexin V‐positive cells in HeLa cells expressing WT or Q134K RIPK3 and in PBMCs from P3 and HCs under the indicated conditions. Data are shown as mean ± SEM.

Supplementary Figure S5: IFN‐related gene expression heatmap in PBMCs from patient P3 compared with HCs.

Supplementary Table S1. List of candidate variants in Still disease‐like Autoinflammatory Syndrome family.