A common human MLKL polymorphism confers resistance to negative regulation by phosphorylation

Across the globe, 2-3% of humans carry the p.Ser132Pro single nucleotide polymorphism in MLKL, the terminal effector protein of the inflammatory form of programmed cell death, necroptosis. Here we show that this substitution confers a gain in necroptotic function in human cells, with more rapid accumulation of activated MLKLS132P in biological membranes and MLKLS132P overriding pharmacological and endogenous inhibition of MLKL. In mouse cells, the equivalent Mlkl S131P mutation confers a gene dosage dependent reduction in sensitivity to TNF-induced necroptosis in both hematopoietic and non-hematopoietic cells, but enhanced sensitivity to IFN-β induced death in non-hematopoietic cells. In vivo, MlklS131P homozygosity reduces the capacity to clear Salmonella from major organs and retards recovery of hematopoietic stem cells. Thus, by dysregulating necroptosis, the S131P substitution impairs the return to homeostasis after systemic challenge. Present day carriers of the MLKL S132P polymorphism may be the key to understanding how MLKL and necroptosis modulate the progression of complex polygenic human disease.

oligomers with biological membranes can be inhibited by the synthetic compound necrosulfonamide 16,22,28 or inhibitory phosphorylation of MLKL at Serine 83 29 .Pharmacological or mutation driven disruption at any major necroptotic signaling checkpoint compromises a cell's capacity to execute necroptosis.
In mouse studies, MLKL-mediated cell death has been implicated as a driver or suppressor of diseases spanning almost all physiological systems depending on the pathological context.The generation of Mlkl gene knockout (Mlkl −/− ), knock-in and conditional knockout mouse models have enabled the role of necroptosis in infectious and non-infectious challenges to be dissected in physiological detail 30 .Interestingly, genetic deletion of Mlkl has no overt developmental or homeostatic effects, with the exception of a reduction in age-related sterile inflammation in female mice 18,31,32 .This is in direct contrast with two mouse models harboring Mlkl point mutations that dysregulate MLKL activation, Mlkl D139V and Mlkl S83G , which exhibit early neonatal death and severe inflammatory phenotypes 29,33 .Altogether, these observations suggest that while constitutive absence of MLKLmediated death is benign, imbalanced execution of necroptotic cell death is deleterious.
Consistent with this notion, more than 20 unique diseaseassociated human germline gene variants in the core necroptotic machinery, encompassing RIPK1, RIPK3, MLKL, have been identified 34,35 .In one family, a haplotype including a rare MLKL loss-of-function gene variant (p.Asp369GlufsTer22, rs561839347) is associated with a severe and progressive novel neurogenerative spectrum disorder characterized by global brain atrophy 36 .A more frequent MLKL loss-of-function gene variant (p.Gln48Ter, rs763812068) was found to be >20 fold enriched in a cohort of Hong Kong Chinese patients suffering from Alzheimer's disease 37 and common variants that cluster around the MLKL brace region were shown to be enriched in trans in a cohort of Chronic Recurrent Multifocal Osteomyelitis patients 33 .More recently, a hypomorphic MLKL missense gene variant (p.G316D, rs375490660) was reported to be associated with Maturity Onset Diabetes of the Young 38 .
Here we present the cellular and physiological characterization of a serine to proline missense polymorphism at MLKL amino acid 132 (p.Ser132Pro; S132P).The S132P polymorphism is the third most frequent human MLKL missense coding variant in the gnomAD database, a large repository of whole genome and exome sequence data from humans of diverse ancestry 39 .To examine the potential human disease-causing effects of this MLKL variant, we exogenously expressed MLKL S132P in human cell lines and introduced the mouse counterpart variant (Mlkl S131P ) into a genetically modified mouse model, revealing that this polymorphism confers MLKL gain-of-function in a cell-and stimulus-dependent manner.This MLKL gain-of-function manifests in in vivo changes to the immune response, impaired bacterial clearance, and defective emergency hematopoiesis.These observed phenotypes provide important insights into how this highly frequent human MLKL S132P polymorphism may contribute to the progression of complex disease.

Carriers of the p.Ser132Pro polymorphism exhibit diverse inflammatory disease profiles
With a global minor allele frequency (MAF) of 0.0138, the MLKL S132P (rs35589326) polymorphism is predicted to be carried by 2-3% of the human population.It has not been detected in individuals assigned East Asian ancestry, is rare in individuals of African or Latino/Admixed American ancestry, and is carried by an estimated 6-7% of individuals of Ashkenazi Jewish ancestry (MAF 0.0315) (www.gnomAD.com,February 2023) (Fig. 1a).Notably, Ser132 is highly conserved across species (Fig. 1b).
Two heterozygous carriers of the MLKL S132P polymorphism were identified in an Australian registry of patients suffering from immune related disease.Patient 1, a female of South American heritage, was diagnosed with SAPHO (synovitis, acne, pustulosis, hyperostosis, osteitis) syndrome (inheritance chart unavailable).Patient 2, a female of European heritage, was diagnosed with systemic IgG4 disease (Supplementary Fig. 1A).Both patients have one or more immediate family members carrying the MLKL S132P polymorphism who were also diagnosed with inflammatory diseases in early adulthood.Following whole genome sequencing, Patient 1 was found to exhibit a region of loss of heterozygosity (5:96031569-5:96364063) which covers CAST, ERAP1, ERAP2 and LNPEP.These genes have been previously associated with inflammatory disease [40][41][42][43][44] .Patient 2 does not carry any other predicted pathogenic gene variants with a MAF < 0.0005 (IUIS human Inborn Errors of Immunity).Primary peripheral blood mononuclear cells (PBMC) isolated from patient 2 showed reduced MLKL protein levels, accompanied by a nominal increase in pro-inflammatory cytokine production in response to the Toll-like receptor (TLR) agonists, lipopolysaccharide and poly I:C, relative to age and sex matched healthy donor PBMCs not carrying the MLKL S132P polymorphism (Supplementary Fig. 1B, C).

MLKL S132P confers resistance to chemical and natural regulatory inhibition
To examine any changes to MLKL function conferred by this common polymorphism, we stably transduced parental wild-type (WT) and MLKL −/− versions of the human colonic epithelial cell line HT29 with doxycycline inducible MLKL WT and MLKL S132P gene expression constructs.We also included the T357/S358 phosphosite mutant MLKL TSEE , previously shown to be inactive 20,45 , as a negative control for necroptosis induction.Cells expressing MLKL S132P died with similar kinetics to MLKL WT following necroptotic stimulation (TNF, Smac mimetic compound A and pan-caspase inhibitor IDN-665; TSI) and, as expected, MLKL TSEE reconstituted cells were resistant to necroptotic cell death (Fig. 1c, Supplementary Fig. 1D, E).Interestingly, in the presence of the MLKL inhibitor necrosulfonamide (NSA), TSI-stimulated cells expressing MLKL S132P exhibited higher levels of cell death than their MLKL WT counterparts at later timepoints (Fig. 1c, Supplementary Fig. 1D).We termed this override of MLKL inhibition 'breakthrough' cell death.Breakthrough death also occurred when MLKL S132P was co-expressed with endogenous MLKL and increased when MLKL S132P expression was augmented by higher doses of doxycycline (Fig. 1c, Supplementary Fig. 1D, E).We also observed breakthrough death in both wild-type and MLKL −/− forms of the human monocytic cell line U937, indicating this is not a cell type specific phenomenon (Supplementary Fig. 1F, G).Breakthrough death was neither due to differences in MLKL expression, nor changes in MLKL phosphorylation (T357/S358), because both were equivalent between MLKL WT and MLKL S132P reconstituted cells (Fig. 1d, Supplementary Fig. 1H, I).Instead, cellular fractionation experiments suggest that breakthrough death was likely due to enhanced association of MLKL S132P with biological membranes (Fig. 1e, Supplementary Fig. 1J).
The small molecule inhibitor, NSA, blocks human MLKL activity through covalent modification of Cysteine 86, located in the MLKL four helix-bundle, executioner domain (Fig. 1f) 16,28 .Recently, it was discovered that MLKL can be endogenously phosphorylated at the proximal residue, Serine 83, and this phosphorylation event plays a species conserved inhibitory role in both human (S83) and mouse (S82) (Fig. 1f) 29 .Like NSA, phosphorylation of MLKL S83 does not alter the capacity of RIPK3 to phosphorylate MLKL at S357/T358 but prevents necroptosis by blocking the association of MLKL with cellular membranes 29 .To test whether phosphorylation at S83 is, like NSA, less effective at inhibiting MLKL S132P induced death, we created stable HT29 (MLKL −/− and wild-type) cell lines that exogenously express gene constructs encoding MLKL WT , MLKL S132P , MLKL S83D (phosphomimetic) and double mutant MLKL S83D,S132P .Consistent with recent studies 29 , the phosphomimetic mutation of S83, MLKL S83D , ablated the capacity of MLKL to execute necroptotic cell death (Fig. 1g).Also, as previously reported, the inhibitory effect of MLKL S83D overrode the activating effects of endogenous MLKL phosphorylation at Ser357/Thr358 (Fig. 1g, Supplementary Fig. 1K, L).Exogenous expression of MLKL S83D also reduced total levels of necroptotic cell death in the presence of endogenous MLKL, indicating that this S83 phosphomimetic acts in a dominant negative manner (Supplementary Fig. 1L).Strikingly, combining the MLKL S132P and MLKL S83D substitutions to create the compound mutant MLKL S83D, S132P restored necroptotic cell death responsiveness, albeit with reduced kinetics and reduced maximal cell death when compared to MLKL S132P alone (Fig. 1g).The MLKL S132P variant also partially overcomes the dominant negative effect of S83  hMLKL WT hMLKL S83D hMLKL S83D,S132P Buffer alone phosphomimetic mutation in cells that endogenously express wildtype MLKL (Supplementary Fig. 1L).
To further dissect how the MLKL S132P substitution facilitates gainof-function, we used liposome dye release assays to test the membrane-damaging capacity of recombinant full-length MLKL WT , MLKL S132P , MLKL S83D and MLKL S83D, S132P expressed and purified from insect cells.We found that the membrane-damaging capacity of recombinant MLKL S132P was increased, and MLKL S83D reduced, relative to MLKL WT (Fig. 1h).Consistent with our observations in cells, the combination mutant MLKL S83D, S132P displayed a membrane-damaging capacity greater than MLKL S83D alone but reduced in comparison to MLKL WT (Fig. 1h).Our results suggest S132P promotes membrane association, and this is likely to contribute to the gain-of-function we observed in vitro under pharmacological and natural inhibition.
Given the clear gain-of-function conferred to MLKL in the context of simulated inhibitory phosphorylation (MLKL S83D, S132P ), we questioned whether mutations that resemble this phosphomimetic mutant may occur naturally.According to the gnomAD database (www.gnomAD.com,Jan 2023), no individuals have been recorded with a polymorphism that encodes the p.Ser83Asp (S83D) replacement.However, there are individuals that carry closely related changes, p.Ser83Cys (MAF 3.54e10 −5 ) and p.Arg82Ser (MAF 1.57e10 −5 ).We created stable MLKL −/− HT29 cell lines that exogenously express gene constructs encoding MLKL S83C and MLKL R82S and assessed their capacity to execute necroptosis.Cells expressing MLKL S83C died with similar kinetics to MLKL WT following necroptotic stimulation however, MLKL R82S reconstituted cells were resistant to cell death (Supplementary Fig. 1M).Consistent with our observations for MLKL S83D , MLKL R82S was phosphorylated at Ser357/Thr358, and the MLKL R82S,S132P compound mutant restored necroptotic killing function (Supplementary Fig. 1M, N).Whilst gene variants at or adjacent to the S83 inhibitory phosphorylation site are less frequent than the MLKL S132P polymorphism in humans, they nonetheless strengthen the precedent for genetically encoded diversity in human MLKL function, and the potential for functionally synergistic or neutralizing combinations of MLKL gene variants.

MLKL S131P mice exhibit differences in steady state immune cell populations
To address if the MLKL S132P polymorphism contributes to immunoinflammatory disease, we performed detailed histological, immunophenotypic, and experimental analyses of a CRISPR generated knockin mouse which carried the orthologous mutation, Mlkl S131P .Mlkl S131P heterozygotes and homozygotes were born according to the expected Mendelian ratios and had normal lifespans (Fig. 2a, b).The healthy presentation of Mlkl S131P mice contrasts with the lethal phenotypes of mice engineered to encode Mlkl D139V and Mlkl S83G mutations 29,33 .Wild-type, Mlkl S131P heterozygote and homozygote mice had comparable body weight and no gross histological differences up to 9 months of age (Fig. 2c, Supplementary Fig. 2A).Notably, distinct from Mlkl D139V or Mlkl S83G homozygotes, no inflammation was detected in the salivary glands, mediastinum, liver or lungs (Supplementary Fig. 2A) 29,33 .No differences in the number of circulating platelets, red blood cells, or white blood cells were observed between wild-type, Mlkl S131P heterozygote and homozygote mice across age (Supplementary Fig. 2B-D).However, reduced numbers of classical Ly6C hi 'inflammatory' monocytes in the bone marrow, but not in the secondary lymphoid organs (spleen and inguinal lymph nodes), were observed in Mlkl S131P homozygotes and heterozygotes relative to wild-type littermate controls (Fig. 2d-i, Supplementary Fig. 2E, J, O).In the secondary lymphoid organs, Mlkl S131P homozygotes had a small but significant increase in splenic CD4 + T cells and B cells relative to wild-type littermate controls (Fig. 2g, Supplementary Fig. 2M, N).Compared to wild-type controls, Mlkl S131P heterozygotes had significant increases in splenic and lymph node B cell populations, as well as splenic CD8 + T cells and Ly6C lo monocytes (Fig. 2g-i, Supplementary Fig. 2J-S).All other innate and adaptive immune cell populations were comparable between genotypes in the bone marrow, spleen, and inguinal lymph nodes (Fig. 2d-i, Supplementary Fig. 2E-S).Further, wild-type, heterozygote and homozygote mice had comparable levels of plasma cytokines at steady state (Fig. 2j).
To investigate the underlying cause of reduced Ly6C hi monocytes in the bone marrow of Mlkl S131P homozygote mice, we examined both the abundance of different myeloid stem cell populations, and their capacity to differentiate and form colonies ex vivo.Mlkl WT/WT and Mlkl S131P/S131P hematopoietic stem cells exhibited equivalent capacity to generate blast, eosinophil, granulocyte, granulocyte-macrophage, and megakaryocyte colonies under all stimulations investigated (Fig. 2k).While the number and composition of myeloid progenitor cell populations in the bone marrow were not significantly different (Supplementary Fig. 2t), compared to Mlkl WT/WT , Mlkl S131P/S131P bone marrow gave rise to an increased number of macrophage colonies under SCF, IL-3 and EPO combined stimulation (Fig. 2k).
inducer of Mlkl gene expression in mice (Fig. 2l, Supplementary Fig. 2U, V) 46 .This sensitivity was further enhanced by TNF and was refractory to inhibitors of RIPK1 (Nec-1s) or RIPK3 (GSK′872) (Fig. 2l, Supplementary Fig. 2U, V).Furthermore, Mlkl S131P forms high molecular weight oligomers following combined stimulation with IFNβ and TNF that are present in the crude membrane fraction of cells (Supplementary Fig. 2X).These results are analogous to cells encoding the constitutively-active Mlkl D139V mutant and cells exogenously expressing inducible Mlkl S131P33 .This supports the notion that MLKL S131P has constitutive RIPK3-independent activity when endogenously expressed in mouse cells 33 .These endogenous Mlkl S131P observations were not limited to fibroblasts.In bone marrow derived macrophages (BMDM), Mlkl S131P/S131P cells also exhibited diminished sensitivity to TSI-induced necroptosis and endogenously produced MLKL S131P was present at reduced levels in comparison to MLKL WT (Supplementary Fig. 2Y, Z).In contrast to immortalized dermal fibroblasts, BMDMs derived from Mlkl S131P homozygotes did not undergo cell death in the presence of IFN-β alone, despite clear IFN-β induced upregulation of MLKL protein

Number of colonies
Bone marrow relative to untreated cells (Supplementary Fig. 2Y, Z).Overall, these data show that, across different cell types, MLKL S131P is not toxic at steady state, but upon different stimuli, both gain-and loss-ofsensitivity to necroptotic cell death is evident.This suggests that any functional deficits or enhancements in carriers of the MLKL S132P polymorphism are likely to be diverse, with cell-and/or contextspecific manifestations.
Emergency hematopoiesis is defective in Mlkl S131P/S131P mice Differences in steady state immune cell populations suggest that overt phenotypes may develop in Mlkl S131P homozygotes following experimental challenge.Specifically, reduced numbers of steady state Ly6C hi monocytes in the bone marrow of Mlkl S131P homozygotes could indicate a defect in hematopoiesis.To test this, we examined recovery from myelosuppressive irradiation as an assessment of hematopoietic function in mice carrying the Mlkl S131P allele.Following myelosuppressive irradiation, recovery of hematopoietic cell numbers and circulating peripheral blood cells was significantly delayed in Mlkl S131P/S131P mice compared with wild-type controls (Fig. 3a-f).In Mlkl S131P/S131P mice, peripheral red blood cell and platelet numbers were significantly reduced at 14 days post irradiation, with the former also decreased at 21 days (Fig. 3a, b).Despite equivalent total white blood cell numbers at 21 days post irradiation, a significant reduction in peripheral monocyte numbers and a non-significant decrease trend in neutrophils was observed in Mlkl S131P/S131P mice compared to wild-type controls (Fig. 3c, Supplementary Fig. 3A, B).These reductions in circulating blood cell numbers were accompanied by significant decreases in the nucleated viable, progenitor, and LSK cell populations in the bone marrow of Mlkl S131P/S131P mice (Fig. 3d-f).Contrastingly, at 21 days post irradiation, Mlkl S131P heterozygotes displayed a significant increase in their total nucleated viable cells in comparison to wild-type controls.This was driven predominantly by significant increases in the number of LSK cells (Fig. 3d, e).Impaired recovery of Mlkl S131P/S131P LSK and progenitor cell numbers was characterized by increased expression of ROS and Annexin V at 21 days post irradiation (Supplementary Fig. 3C, D).Mlkl S131P homozygotes had increased plasma G-CSF levels at 14 and 21 days post irradiation when compared to Mlkl WT/WT controls (Fig. 3g).At 21 days, all other plasma cytokines, with exception of IL-1β and RANTES, were equivalent between genotypes (Supplementary Fig. 3E).Mlkl WT/S131P and Mlkl S131P/S131P mice had decreased plasma IL-1β levels, whilst Mlkl S131P/S131P mice alone had reduced RANTES levels compared to Mlkl WT/WT controls (Supplementary Fig. 3E).
Recruitment of Ly6C hi monocytes to sites of sterile inflammation is reduced in Mlkl S131P/S131P mice To examine if the defects in emergency hematopoiesis displayed by Mlkl S131P/S131P mice could result in defective immune cell recruitment to sites of inflammation or infection, we employed a model of localized sterile inflammation induced by zymosan.Intra-peritoneal injection of zymosan results in a rapid influx of immune cells, predominantly neutrophils, into the peritoneal cavity over 72 hours 47 .The cellular content of the peritoneal cavity and numbers of circulating peripheral blood cells was examined at 4 and 24 hours post-zymosan injection.Mlkl S131P/S131P mice displayed equivalent early recruitment of neutrophils and other immune cells to the peritoneum, with exception of reduced quantities of Ly6C hi monocytes (Fig. 4a, b, Supplementary Fig. 4A-H).At 24 hours post-zymosan injection, this significant reduction in the recruitment of Ly6C hi monocytes to the peritoneum of Mlkl S131P/S131P mice was more pronounced (Fig. 4c, d, Supplementary Fig. 4I-P).In the peripheral blood, monocytes were significantly elevated in Mlkl S131P/S131P mice in comparison to wild-type controls at 4 hours post-zymosan (Fig. 4e, Supplementary Fig. 4Q-S).However, by 24 hours, the number of peripheral blood monocytes was equivalent between genotypes, with only a non-significant increase in neutrophils observed in Mlkl S131P/ S131P mice (Fig. 4f, Supplementary Fig. 4T-V).Mlkl WT/S131P mice exhibited increases in the peripheral blood lymphocyte and neutrophil numbers when compared to wild-type controls at 24 hours (Fig. 4f).We also measured cytokines present in the peritoneum at 4 and 24 hours postzymosan injection (Fig. 4g, h, Supplementary Fig. 4W-Y).At 4 hours, Mlkl S131P/S131P mice had increased quantities of IL-13 and MCP-1 when compared to wild-type controls (Fig. 4g, h).No statistically significant differences were observed in the quantities of peritoneal cytokines between any genotypes at 24 hours post-injection (Supplementary Fig. 4Y).
We previously observed that Mlkl S131P/S131P cells exhibited reduced response to TSI-induced necroptosis.To assess whether this defect in necroptotic function was observed in activated immune cells recruited to sites of inflammation, we isolated neutrophils from the peritoneum at 4 hours post-zymosan injection.Consistent with our previous in vitro findings, no differences were observed in the capacity of Mlkl S131P/ S131P inflammatory neutrophils to undergo spontaneous apoptosis (Supplementary Fig. 4Z).However, the percentage of Mlkl S131P/S131P inflammatory neutrophils dying following necroptotic stimulation with TSI was decreased in comparison to wild-type neutrophils (Fig. 4i).

Mlkl S131P homozygotes exhibit reduced capacity to clear Salmonella infection
Reductions in the recruitment of Ly6C hi monocytes to sites of inflammation raise important questions as to whether the Mlkl S131P mutation impacts defense against pathogens.Salmonella enterica has long been an important pathogenic selective pressure in humans 48 .For non-typhoidal Salmonella serovars, such as S. enterica serovar Typhimurium, human infection is typically limited to the gastric mucosa 49,50 .In mice and immunocompromised humans, S. Typhimurium can cause severe systemic disease after dispersal from the gut by dendritic cells and macrophages to peripheral organs including the spleen and liver 51 .Mlkl S131P homozygote, heterozygote and wild-type mice were infected with a metabolically growth-attenuated Salmonella Typhimurium strain BRD509 (from here referred to as Salmonella), via oral gavage (1 × 10 7 colony forming units (CFU)).Daily monitoring for the 14 day infection period showed no differences in core body temperature or body mass between genotypes (Supplementary Fig. 5A, B).Consistent with normal weights, there were no obvious differences in the integrity of the intestinal epithelial barrier or intestinal monocyte and macrophage populations at infection endpoint (Fig. 5a).Despite this, Mlkl S131P/S131P mice had increased Salmonella burden in both the spleen and liver compared to wild-type controls (Fig. 5b, Supplementary Fig. 5C, D).Bacterial colonization in the feces was increased in female Mlkl S131P/S131P mice relative to wild-type female controls (Fig. 5b, Supplementary Fig. 5E).In the peripheral blood, Mlkl S131P/S131P mice had significantly reduced numbers of circulating lymphocytes and monocytes, as well as a trend towards decreased numbers of circulating neutrophils (Fig. 5c, Supplementary Fig. 5F-H).Infected Mlkl S131P/S131P mice also exhibited significant decreases in the quantity of splenic Ly6C hi monocytes in comparison to infected wild-type controls (Fig. 5d, e, Supplementary Fig. 5I-O).At 14 days post-infection, Mlkl WT/ S131P and Mlkl S131P/S131P mice both had significant increases in plasma concentrations of MCP-1, when compared to wild-type controls (Fig. 5f).All other plasma cytokines were equivalent between genotypes at 14 days post-infection (Fig. 5g).
We did not observe any differences in the kinetics of Salmonellainduced cell death in ex vivo BMDM infection assays, nor inflammasome activation, measured by GSDMD cleavage (Supplementary Fig. 5P, Q).Together these data show a hindered pathogen defense in Mlkl S131P homozygotes accompanied by widespread immunophenotypic deficiencies.Combined, our investigations of Mlkl S131P homozygotes under challenge highlight a defect in emergency hematopoiesis that manifests in a disruption to integral inflammatory and immune responses.This provides important insights into the potential modulation of immunoinflammatory disorders in MLKL S132P carriers.

Discussion
A high frequency MLKL S132P polymorphism present in 2-3% of the global population confers a gain-of-function to MLKL resulting in hematopoietic dysfunction and immune cell defects in a genetically modified mouse model.In human cells, MLKL S132P was resistant to chemical inhibition by necrosulfonamide treatment and endogenous inhibitory phosphorylation at Serine 83.A gain in necroptotic function was also observed when the murine equivalent, Mlkl S131P , was examined in situ.Fibroblasts isolated from Mlkl S131P homozygotes exhibit RIPK3independent cell death in the presence of IFN-β, a strong inducer of Mlkl gene expression.Under regular culture conditions, MLKL S131P protein levels are reduced relative to MLKL WT , manifesting in a reduced capacity for necroptosis in fibroblasts and immune cells stimulated with TSI.While the reduction in MLKL S132P protein levels was also evident in PBMCs isolated from one human individual heterozygous for this polymorphism, our capacity to fully compare between mouse and human systems is limited by the lack of suitable human cell lines that express MLKL S132P from its endogenous gene locus.The low number of patients identified in our study limits our current conclusions regarding the disease implications of the MLKL S132P polymorphism however, the contribution of MLKL S132P to human inflammatory disease remains a tantalizing and important avenue for future study.
Extensive characterization of the Mlkl S131P mouse unveiled steady state decreases in the bone marrow pool of inflammatory Ly6C hi monocytes, that were also reflected in reduced numbers at sites of sterile inflammation and bacterial infection.During zymosan-induced peritonitis, increased peripheral monocytes were observed at 4 hours however, Ly6C hi monocytes were significantly reduced in the peritoneum of homozygotes at 24 hours.Under infection challenge, an increased burden of Salmonella was present in the spleen and liver of homozygotes, accompanied by significant reductions in the circulating lymphocytes and monocytes.In both cases, monocyte deficits were accompanied by an increase in plasma MCP-1 levels.These deficits are not attributed to any inherent variations in Mlkl S131P mouse monocyte/macrophage progenitor cell populations at steady state however, whether these deficits are the result of Ly6C hi monocyte recruitment to an unidentified site of inflammation in Mlkl S131P/S131P mice at steady state remains an important avenue of open investigation.Interestingly, Mlkl S131P stem cells did exhibit an increased capacity to form macrophage colonies ex vivo under IL-3, EPO and SCF combination treatment.These data led us to hypothesize that Mlkl S131P/S131P Ly6C hi monocytes may be more prone to differentiation than their Mlkl WT/WT counterparts under specific conditions.Hematopoietic dysfunction in Mlkl S131P homozygotes is evident following radio-ablation, with a severely reduced capacity for stem cells to repopulate in situ, or even Mlkl WT recipients.Hematopoietic stem cells expressing the constitutively active Mlkl S131P are characterized by increased ROS and Annexin V positivity following irradiation, marking their enhanced propensity for cell death.Taken together, the capacity for monocyte generation, activation and peripheral differentiation within the Mlkl S131P mouse under conditions of stress has emerged as the most enticing area for future investigation.
Prior to this work two other single point mutant Mlkl mouse models, Mlkl D139V and Mlkl S83G , had been reported, both of which exhibited full or partial homozygous postnatal lethality characterized by severe inflammation 29,33 .In stark contrast to these models, the Mlkl S131P homozygotes were born normal and with no signs of inflammatory disease at steady state.However, upon challenge, similarities between Mlkl S131P and Mlkl D139V mice were revealed.Hematopoietic dysfunction is observed following myelosuppressive irradiation in homozygous or heterozygous mice harboring the Mlkl S131P or Mlkl D139V allele, respectively 33 .
Despite the ostensible basal state phenotypes of Mlkl D139V and recently described Mlkl S83G homozygotes differing from the Mlkl S131P homozygotes 29,33 , there were similarities observed at the molecular level.As for MLKL S131P , RIPK3-independent gain-of-function was reported for MLKL D139V , but not MLKL S83G29, 33 .This major similarity in molecular function is not unexpected considering the proximity of the D139 and S131 residues within mouse MLKL.Constitutive activity of the D139V brace helix mutants is restrained by cellular mechanisms that limited protein level, with decreases in endogenous MLKL observed in cells generated from Mlkl D139V homozygotes.Interestingly, ubiquitin-Fig. 3 | Mlkl S131P mice show delayed recovery from myelosuppressive irradiation.a Peripheral red blood cells (n = 9, 8, 8, 15, 13, 13, 8, 9, 3, 3, 3, 3, 9, 7, 7), b platelets (n = 5, 7, 9, 15, 13, 13, 8, 9, 3, 3, 3, 3, 9, 7, 7) and c monocytes (n = 9, 8, 9, 15, 13, 13, 8, 9, 3, 3, 3, 3, 9, 6, 7) in Mlkl WT/WT , Mlkl WT/S131P , and Mlkl S131P/S131P mice following treatment with 5.5 Gy radiation.Quantified nucleated viable cells (d), progenitor (e), and LSK (f) populations in the bone marrow of mice after myelosuppressive radiation (d-f n = 10, 6, 7, 3, 3, 3, 3, 3, 3, 9, 7, 7).g Multiplex measurement of plasma G-CSF levels at 7, 14 and 21 days post-myelosuppressive radiation.Each symbol represents one individual mouse sampled, with mean ± SEM of n = 2, 3, 3, 3, 3, 3, 5, 3, 4 biologically independent mice from two separate experiments.Bone marrow from Mlkl WT/WT , Mlkl WT/S131P or Mlkl S131P/S131P mice on CD45 Ly5.2 background was mixed with wild-type GFP + competitor bone marrow on a CD45 Ly5.2 background and transplanted into irradiated CD45 Ly5.1 recipients.h Relative donor contribution to PBMCs was assessed at 6 weeks post-transplantation. Mean ± SEM shown for n = 11, 6, 9 biologically independent recipient mice.Host contribution (CD45 Ly5.1 ) depicted in pink, GFP competitor in green and test (Mlkl WT/WT , Mlkl WT/S131P or Mlkl S131P/S131P ) in purple.P value calculated using an unpaired, two-tailed Students t test (a-g).Source data are provided as a Source Data file.mediated targeting of MLKL to the lysosome was identified to be mechanistically involved in the restraint of necroptosis in cells expressing MLKL D139V , and as a means of clearing endosomal Listeria monocytogenes and Yersinia enterocolita 33,52,53 .Exploring if the reduction in cellular MLKL levels we observed in the PBMCs of a human carrier of the MLKL S132P polymorphism, or cells isolated from the MLKL S131P mouse, is likewise mediated by ubiquitination and lysosomal targeting will be an important next step, particularly as it relates to the clearance of these intracellular bacteria from human cells.Ablation of the inhibitory phosphorylation site in Mlkl S83G mice similarly subjects MLKL to downregulation in mouse cells 29 .In sum, our analysis of MLKL S131P further highlights the importance of cellular mechanisms The finding that MLKL S132P retains necroptotic killing activity despite simulated inhibitory phosphorylation at Ser83 in human cells is highly notable.The inflammatory phenotypes that result from homozygosity of the phosphoablating (S83G) mutation provides insight into the potential disease development that may occur in carriers of the MLKL S132P polymorphism under environmental and cellular scenarios where Ser83 inhibitory phosphorylation is deployed.Identifying the kinase(s) responsible for phosphorylation at MLKL Ser83 and understanding which cell types are primed to deploy this kinase(s) is an important next step in determining if this polymorphism promotes clinically relevant changes to homeostasis or disease outcomes in humans.
Evidence for positive selection has been found for over 300 immune-related gene loci and many of these have been found to be associated with the incidence of autoimmune and autoinflammatory disease in modern humans 54,55 .Many of these variants have also been mechanistically linked to pathogen defense (Karlsson et al., 2014,  Ramos et al., 2015), with pathogenic microbes a major driver of genetic selection over human history.While a diminished capacity for mice to clear disseminated Salmonella would argue against the hypothesis that the MLKL S132P polymorphism has been positively selected for in human populations, it is important to note that only 1.4% of human carriers are homozygotes 39 .MLKL S132P heterozygosity is by far the most prevalent, and evolutionarily relevant, human scenario.Mlkl S131P heterozygote mice displayed gene dosage phenotypes consistent with homozygotes, with one exception.Following sublethal myelosuppressive irradiation, a gain in bone marrow hematopoietic stem cell numbers in heterozygotes was observed at day 21 whilst a catastrophic drop occurred in homozygotes.This increased stem cell capacity does not persist long-term.In competitive bone marrow transplants, Mlkl WT/ S131P stem cells compete similarly to Mlkl WT/WT stem cells at 6 weeks posttransplantation.However, the increased fitness conferred to heterozygous mice at this early timepoint provides intriguing insights into other selective pressures that may have promoted accumulation of this polymorphism in humans 56 .Sepsis, where emergency hematopoiesis is an essential determinant in survival, is an interesting and highly evolutionarily important avenue of exploration for the study of MLKL S132P polymorphism frequency 57,58 .Since nutrition is another important driver of genetic adaptation in humans, exploring the role of MLKL S132P in metabolic disease is also of interest.An important precedent for this is a recent report of an association between human RIPK1 promoter polymorphisms and diet-induced obesity 59 .The potential for negative selection of this polymorphism over time is also an important avenue of exploration in light of the recent discovery of a common TYK2 variant and its role in enhancing susceptibility to severe infection by historically important human pathogen mycobacterium tuberculosis 60,61 .
By many, MLKL is viewed as a potential therapeutic target for drug discovery due to its established involvement in multiple human diseases, especially inflammatory pathologies 33,[62][63][64] .To date no human clinical trials have been conducted on MLKL-targeted small molecules, although several inhibitors have been reported in the literature, including human MLKL inhibitor NSA, which all function by targeting Cys86 16,65 .Intriguingly, we show that in the presence of NSA human MLKL S132P protein displays a gain-of-function that results in non-inhibitable cell death.While NSA has been a fundamental tool for in vitro studies of necroptosis, our findings raise interesting questions about the suitability of Cys86-targetted MLKL inhibitors for the 2-3% of the population that carry the MLKL S132P polymorphism.

Patient recruitment, ethics, and informed consent
Patients and their relatives were recruited from the Department of Clinical Immunology and Allergy, Royal Melbourne Hospital, Victoria, Australia, and the Centre for Personalized Immunology, Australian National University, Canberra, Australia.Unrelated, age and sex matched healthy controls that did not carry the MLKL p.Ser132Pro polymorphism were recruited via the Volunteer Blood Donor Registry, Parkville.Patient #1 is a (self-reported) female, aged in their 30's, diagnosed with SAPHO syndrome and patient #2 is a (self-reported) female, aged in their 50's, diagnosed with systemic IgG4 disease.Healthy control was a (self-reported) female that was age matched to patient #2.Selection of patients from this registry was based upon genotype and their geographical location (Royal Melbourne Hospital), and no other patient characteristics, including sex, was considered.No suitable male carriers of the p.Ser132Pro polymorphism were available in this study.Informed written consent was obtained from all participants for genomic analysis and immunological studies.All procedures were performed and are reported here with the approval of human ethics review boards of all Institutes that participated in human genetics studies; Australian National University, The Walter and Eliza Hall Institute of Medical Research (approved projects 2009.162,10/02) and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Genomic analysis
Whole exome sequencing was performed by the Canberra Clinical Genomics service.Libraries were prepared and enriched using the SureSelect Clinical Research Exome v2 kit (Agilent Technologies), and targeted regions were sequenced using an Illumina sequencing system with 100 bp paired-end reads, with an average depth of >35.Reads were aligned to the human genome reference sequence (GRCh38) using the Burrows-Wheeler Aligner (BWA-MEM), and variant calls made using the Genomic Analysis Tool Kit (GATK).Patient #1 and #2 MLKL sequence data can be directly accessed from the Sequence Read Archive (SRA BioProject accession number PRJNA1007397) via the following link; https://www.ncbi.nlm.nih.gov/sra/?term=PRJNA1007397.

Mice
Mlkl S131P CRISPR mouse strain 33 was bred and maintained at WEHI (Parkville) Animal Facility.On limited occasions, C57BL/6 J mice bred at WEHI Kew facility were used as non-littermate WT controls (where indicated).C57BL/6-CD45 ly5.1 mice were bred and imported from WEHI Kew facility.These are temperature (21 °C ± 3 °C) and humidity (40-70%) controlled specific pathogen free facilities with a 14 h:10 h day:night cycle.All mice were euthanised by CO 2 asphyxiation utilizing Quietek automated CO 2 induction units.Experiments were performed with both male and female Mlkl S131P colony mice aged from 6-52 weeks old (specific age ranges denoted in experimental methods).Data for both sexes were presented combined or separately as indicated.
Cell treatments were completed with agonists/antagonists at the following concentrations: 100 ng/ml recombinant hTNF-Fc (produced in house as in 67 ), 500 nM Smac mimetic Compound A (provided by Tetralogic Pharmaceuticals; as in 68
Human blood (patient and healthy donor) was collected by collaborators via venipuncture at the Royal Melbourne Hospital.Collected blood was diluted with PBS and layered on an equal volume of Histopaque (density 1.077 g/ml) and centrifuged for 30 minutes, 700 × g at 20 °C.The layer containing peripheral blood mononuclear cells (PBMC) was harvested and washed with PBS, then frozen in FCS + 10% DMSO and stored in liquid nitrogen.
CRISPR-edited derivative HT29 and U937 cells lacking MLKL were reported previously 45 and were produced in-house using unmodified HT29 cells supplied by Mark Hampton (originally from ATCC) and U937 cells supplied by ATCC.HEK293T cells were sourced from the laboratory of John Silke and were originally purchased from the ATCC.
Western blot U937 cells were seeded into 48-well plates at 60,000 cells/well and induced for 3 hours with doxycycline (20 ng/mL, 100 ng/mL or 500 ng/ mL) to stimulate MLKL expression.HT29 cells were seeded into 48-well plates at 45,000 cells/well and following 12-14 hours of cell adherence, cells were induced overnight with doxycycline (20 ng/mL, 100 ng/mL or 500 ng/mL) for stimulation of MLKL expression.BMDMs were plated at 400,000 cells/well in a 24-well plate and MDFs (primary and immortalised) were plated at 30,000 cells/well in a 48-well plate.Cells were stimulated as indicated for 6 hours, except for BMDMs stimulated with LPS for 2 hours before addition of Smac mimetic Compound A for a further 4 hours.Human primary PBMCs were plated at 45,000 cells/well and stimulated for 4 hours.All cells were harvested in SDS Laemmli's lysis buffer, boiled at 100 °C for 10-15 min, and then resolved by 4-15% Tris-Glycine SDS-PAGE (Bio-Rad).Proteins were transferred to nitrocellulose or PVDF membrane and probed with antibodies as indicated.

BN-PAGE
Immortalised MDFs were plated at 1 × 10 6 cells/well in a 6-well plate, left to settle for 4 hours and stimulated with TNF + IFN-β for 22 hours or TSI for 6 hours.MLKL −/− HT29 cells were plated at 0.25 × 10 6 cells/ well in a 12-well plate and left to settle overnight.HT29 cells were then stimulated with 100 ng/ml doxycycline overnight followed by TSI or TSI + NSA treatment for 2, 4 or 6 hours.All cells were permeabilised in MELB buffer (20 mM HEPES pH 7.5, 100 mM sucrose, 100 mM KCl, 2.5 mM MgCl 2 . 2 µM N-ethyl maleimide) supplemented with 0.025% digitonin, protease and phosphatase inhibitors.Using centrifugation (5 min, 11,000 × g), cells were fractionated into cytoplasmic and crude membrane fractions and solubilised in 1% digitonin.Samples were mixed with BN-PAGE loading buffer and resolved using 4-16% Nati-vePAGE (Invitrogen) gel.Proteins were transferred to PVDF membrane that was subsequently destained (50% v/v Methanol, 25% v/v acetic Fig. 5 | Mlkl S131P homozygote mice exhibit bacterial clearance defects following oral Salmonella infection.a ZO-1 (yellow) and CD68 (purple) staining in epithelial barrier of intestinal sections taken at 14 days post-Salmonella infection.Images are representative of n = 5 biologically independent mice per genotype.b Increased bacterial burden observed in spleen and liver, but not feces in Mlkl S131P/S131P mice at infection endpoint (n = 13, 9, 16).c Quantification of circulating white blood cells (lymphocytes, neutrophils and monocytes) using ADVIA hematology (n = 13, 8, 16) and (d, e) splenic adaptive (CD4 + T cells, CD8 + T cells and B cells) and innate (macrophages, Ly6C hi monocytes, Ly6C lo monocytes and neutrophils) immune cells using flow cytometry (n = 14, 9, 16; with exception of macrophages n = 11, 7, 11).f, g Multiplex measurement of plasma cytokine levels at 14-days post Salmonella infection (n = 10, 5, 10).Salmonella infection was performed on three independent occasions.Error bars represent mean ± SEM with each symbol representing an individual mouse.P value calculated using a two-tailed Mann-Whitney test b or an unpaired, two-tailed Students t test (c-g).Source data are provided as a Source Data file.acid), denatured (6 M Guanidine hydrochloride, 5 mM β-mercaptoethanol, 10 mM Tris pH 6.8), blocked (1% BSA or 2% skim milk) and probed with denoted antibodies.Samples of fractions were reduced in SDS Laemmli's lysis buffer and then immunoblotted for loading controls.

IncuCyte analysis
Primary MDFs were plated at 8000 cells per well in a 96-well plate.BMDMs were plated at 150,000 cells per well on day 6 of culture in a 48-well plate.Primary MDFs and BMDMs were left to settle overnight before stimulation.The next day MDFs and BMDMs were stimulated in culture media supplemented with propidium iodide.Immortalised MDFs were plated at 10,000 cells/well in a 96-well plate and left to settle for 4-6 hours.Cells were then stimulated in DMEM + 1% FCS supplemented with SYTOX Green (1:10,000).HT29 cells were plated at 45,000 cells per well in a 48-well plate and left to settle for 6 hours before overnight doxycycline pre-stimulation (20 ng/mL, 100 ng/mL or 500 ng/mL).HT29 cells were stimulated in Phenol Red-free media supplemented with 2% FCS, 1 mM Na pyruvate, 1 mM L-GlutaMAX, SYTOX Green (Invitrogen, S7020) and either DRAQ5 (Thermofisher, #62251) (MLKL −/− HT29) or SPY_620 (Spirochrome, SC401) (WT HT29).
Neutrophils isolated from the peritoneum at 4 hours postzymosan injection were counted and plated at 60,000 cells/well in a 48-well plate.Plating media (RPMI + 8% FCS) was supplemented with SYTOX Green and DRAQ5 dyes.
Images were taken every hour using IncuCyte SX5 or S3 imager and analysed using Sartorius Incucyte Imager Software v2022B, v2021B, v2020C & v2018A.Percentage values were quantified by number of dead cells (SYTOX Green or propidium iodide positive) as a proportion of total cell number (DRAQ5 or SPY620 positive).TNF ELISA 100,000 cells were stimulated with LPS (10 ng/mL) or Poly I:C (2.5 µg/ mL) for 3 hours.PBMC supernatant cytokine content was measured by ELISA (R&D: STA00C) according to the manufacturer's instructions.The measurements were performed in technical triplicates.

Mouse histopathology
Organs from male and female Mlkl S131P mice between 7-9-months of age were fixed in 10% buffered formalin and examined by histopathologists and veterinary histopathologists at the Australian Phenomics Network, Melbourne.

Hematological analysis
Cardiac, submandibular or retro-orbital blood collected from male and female Mlkl S131P colony mice at steady state (8-52 weeks old) or following challenge was placed into EDTA-coated tubes.Blood cells were left undiluted or diluted 2-to 11-fold in DPBS for automated blood cell quantification using an ADVIA 2120i hematological analyser on the same day as harvest.

Cytokine quantification
All plasma was collected by centrifugation (10,000 × g, 5 min) and stored at −80 °C.Lavage fluid and plasma cytokine quantities were measured by Bioplex Pro mouse cytokine 23-plex assay (Bio-Rad #M60009RDPD) according to manufacturer's instructions.When samples were denoted as <OOR, below reference range, for a particular cytokine they were assigned the lowest recorded value for that cytokine across all samples.When samples were denoted as >OOR, above reference range, for a particular cytokine they were assigned the highest recorded value for that cytokine across all samples.

Mouse model of Salmonella infection
Male and female Mlkl S131P colony mice used in this experiment were a mix of littermates and non-littermates aged 6-12 weeks and wild-type mice that were littermates behaved equivalently to non-littermates.Mice were infected with Salmonella enterica serovar Typhimurium strain BRD509 69 at 10 7 colony forming units (CFU) by oral gavage.Mice were euthanized by CO 2 asphyxiation at 14 days post-infection.Cardiac bleeds were taken, and blood populations analyzed using an ADVIA hematology analyser.Liver, spleen, and feces were harvested for enumeration of viable bacteria on nutrient agar.Organs from infected mice were weighed and homogenized in 2 mL (spleens), 5 mL (livers) or 1 mL (feces) of PBS.Homogenates were serially diluted (in duplicate) in PBS and 10 µl drops plated out in duplicate onto LB agar ( + streptomycin) and incubated overnight at 37 °C.CFU/mL was calculated per organ for each mouse and then standardized to CFU/organ based upon organ weight.A portion of the spleen was processed for flow cytometry analysis.

In vitro Salmonella infection
Salmonella Typhimurium strain SL1344 was grown shaking at 37 °C overnight in LB broth supplemented with 50 µg/mL streptomycin.OD 600 was determined using a spectrophotometer to calculate multiplicity of infection (MOI).BMDMs on day 6 of differentiation were plated at 4 × 10 5 cells/well in a 6-well plate for western blot analyses or 5 × 10 4 cells/well in a 96-well plate for LDH assays.Cells were infected at MOI:25 for western blot or MOI:10 or 50 for LDH assays in antibioticand serum-free DMEM for denoted incubation times.For all experimental analyses, following 30 minute incubation, cells were washed and replaced in DMEM media supplemented with 50 µg/ml gentamycin to ensure growth inhibition of extracellular bacteria.BMDM cell death levels were measured as a percentage of LDH release using the Promega CytoTox 96 Non-Radioactive Cytotoxicity Assay (G1780), according to manufacturer's instructions.

Sublethal irradiation
Male and female Mlkl S131P colony mice used in this experiment were a mix of littermates and non-littermates aged 7-17 weeks, and wild-type mice that were littermates behaved equivalently to non-littermates.Mice were irradiated with a 5.5 Gy sub-lethal dose of γ-irradiation and received neomycin (2 mg/mL) in the drinking water for 3 weeks.At 4 days post irradiation, a mandible bleed was analyzed via ADVIA hematology to confirm successful irradiation.Mice had submandibular bleeds analyzed by ADVIA hematology and had long bones harvested for flow cytometry analysis at either 7, 14 or 21 days postirradiation to assess stem cell capacity.

Hematopoietic stem cell transplants
Donor bone marrow were injected intravenously into recipient female C57BL/6-CD45 ly5.1 mice (WEHI Kew) aged 6-10 weeks, following 11 Gy of γ-irradiation split over two equal doses.Recipient mice received neomycin (2 mg/mL) in the drinking water for 3 weeks.Long term capacity of stem cells was assessed by flow cytometric analysis of donor contribution to recipient mouse peripheral blood at 6 weeks.

Zymosan-induced peritonitis
Male and female Mlkl S131P colony mice used in this experiment were a mix of littermates and non-littermates aged 7-17 weeks, and wild-type mice that were littermates behaved equivalently to non-littermates. 1 mg of zymosan A from Saccharomyces cerevisiae (Sigma-Aldrich) was intra-peritoneally injected into mice to induce sterile peritonitis.After 4 or 24 hours, mice were euthanized, cardiac bled and bone marrow collected.The peritoneal cavity was washed with 1 ml of cold PBS and cells within the lavage fluid were collected by centrifugation (300 × g, 5 minutes).Bone marrow and peritoneum immune cells were quantified by flow cytometry.
All samples were analyzed using the Aurora Cytex flow cytometer and FlowJo version 10.8.1.With exception of zymosan induced peritonitis and Salmonella experiments that used the Aurora Cytex automated volume calculator, cells were mixed with counting beads to quantify absolute cell numbers.All flow cytometry gating strategies are shown in Supplementary Fig. 7 and antibody details are outlined in Supplementary Table 1.

Liposome dye release assays
Recombinant full-length human MLKL (residue 2-471) proteins were expressed in Sf21 insect cells using bacmids prepared in DH10Mul-tiBac E.coli (ATG Biosynthetics) from pFastBac Htb vectors using established procedures 70 .Briefly, full-length GST-tagged human MLKL proteins were purified using glutathione agarose resin (UBPBio) 45 followed by size-exclusion chromatography using HiLoad 16/600 Superdex 200 pg column (Cytivia).Fractions corresponding to full-length human MLKL tetramers (elution volume 55-63 ml) were pooled for liposome assays.Large Unilamellar Vesicles (LUV) 100 nm in diameter were prepared using a plasma membrane-like lipid mix (20% POPE, 40% POPC, 10% PI/PI(4,5)P 2 , 20% POPS, 10% POPG) resuspended in chloroform at 20 mg/mL.To form dye filled liposomes, dried lipids were resuspended in LUV buffer (10 mM HEPES pH 7.5, 135 mM KCl) supplemented with 50 mM 5 (6)-Carboxyfluorescein dye (Sigma).Freeze-thaw cycles were completed five times prior to extrusion (21 times) of the lipid mixture through polycarbonate membranes of 100 nm size cutoff (Avanti Polar Lipids, AL USA) using a pre-warmed mini extruder (Avanti Polar Lipids, AL USA).Liposome stocks were stored at 4 °C in the dark.Recombinant human MLKL protein was diluted to 1 µM (2× desired final concentration) in LUV buffer and aliquoted into a 96-well flat-bottom plate (ThermoFisher Scientific).Liposomes were purified from excess dye using a PD-10 desalting column (Cytiva) and diluted to 20 µM in LUV buffer.Immediately following addition of the liposomes to the plate (1:1 ratio liposomes:protein) fluorescence (485 nm excitation, 535 nm emission) was measured every 2 minutes for 60 minutes total on the CLARIOstar plate reader (BMG Labtech; 5.70 R2).Baseline measurements were determined by incubation of liposomes with LUV buffer alone.All assays were performed in triplicate.Data plotted as mean ± SD of one independent repeat that is representative of three independent assays.

Statistical analyses
All data were graphed using Prism v9.Individual points plotted signify independent experimental repeats performed on different days or biologically independent data points.All P values were calculated in Prism v9 using the statistical test identified in figure legends.

Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.