Facts

  • Necroptosis is closely associated with the pathogenesis of many human diseases.

  • The in vivo appearance of necroptosis indicates that both caspase-independent and caspase-dependent mechanisms control this cell death pathway.

  • More than 70 human molecules play a role in the regulation of necroptosis.

  • More than 20 approved drugs have the potential to regulate necroptosis.

Open Questions

  • How can we monitor and regulate necroptosis in human diseases?

  • What are the main molecular targets in caspase independent regulatory mechanisms of necroptosis?

  • How effective can the off-label use of already approved drugs in necroptosis-driven diseases be?

Introduction

The development and homeostasis of multicellular organisms depends on the balance between cell proliferation and cell death. In the past few years new regulated cell death pathways have been discovered and classified1. One of these tightly controlled inflammatory cell death pathways – necroptosis – has come to the center of attention because of its known contribution to the pathogenesis of many diseases1,2.

Many death-, pattern recognition-, DNA binding-, adhesion, and dependence-receptors, immune reactions, pathogens and various drugs have been identified as necroptosis triggers1,3. Necroptosis utilizes a signaling pathway requiring the involvement of receptor interacting protein kinase 3 (RIPK3)4, mixed lineage kinase domain-like protein (MLKL)5 and upon stimulation of death receptors (DR)2 RIPK1. RIPK3 oligomerization and its subsequent phosphorylation allows the RIPK3-MLKL interaction and the double phosphorylation of MLKL by RIPK36. After this step, MLKL forms oligomers and translocates to the plasma membrane to execute necroptosis (Fig. 1). Generally, necroptosis requires inhibition of caspases3,7 or the absence of the pro-caspase-8-activating adaptor Fas-associated protein with death domain (FADD)8, demonstrating the crucial role of the apoptotic platform in the negative regulation of necroptosis. Active caspases block necroptosis2 preferentially through the cleavage of RIPK19, RIPK33,10, and cylindromatosis (CYLD) protein11 which acts as the de-ubiqutinase enzyme of RIPK1. During DR-mediated signaling, inhibitors of apoptosis proteins (IAPs) initiate the ubiquitination of RIPK1 and this process favors cell survival12. Blockage of IAPs or the subsequent events of IAP-induced signaling strongly support necroptosis13. Various molecular pathways have been documented as regulators of downstream necroptotic events beside MLKL-mediated membrane rupture, but the complexity of the signaling and regulation network of necroptosis are still not fully understood.

Fig. 1: Backbone of necroptosis signaling.
figure 1

Various extra - or intracellular signals activates the RIPK3 protein directly or through RIPK1. RIPK3-mediated phosphorylation induces MLKL membrane translocation and consequently, ion influx results in necroptosis147. Survival signals through upregulation of IAPs or activation of TAK1 kinase pathway blocks RIPK1-induced signaling and protects cells from unwanted necroptosis. Caspase-8-mediated cleavage of pro-necroptotic RIPK1 and RIPK3 ensures the dominance of immunologically silent apoptosis to immune stimulant necroptosis

The immunological outcome of cell death can be classified as anti-inflammatory or pro-inflammatory and tolerogenic or immunogenic1. Dominance of apoptosis ensures the tolerogenic outcome of cell death under physiological conditions. When apoptosis signaling is blocked, necroptotic pathways are activated and the dying cells have the potential to initiate innate immune responses via production of damage associated molecules (DAMPs) resulting in an inflammatory response14. Signaling in necroptotic cells also supports the cross priming capacity of dendritic cells (DCs)15.

In this review our goal was to understand the molecular background of necroptosis in humans and to find potential points of clinical intervention. We summarized how the expression, posttranslational modification, and localization of necroptotic molecules are regulated and what the interaction partners of the necrosome complex are. Finally, we provide an overview of drugs, which are already used in the clinic and have been shown to affect necroptosis.

Necroptosis involved in human diseases

Currently, necroptosis is mainly documented in various in vivo mice models16,17, but regulated necrosis contributes to the pathogenesis of many human diseases (Table 1). Both up and down-regulation of necroptosis and misregulation of the apoptosis-necroptosis transition which modifies the immunological outcome of cell death contribute to the evolution of degenerative disorders, autoimmune processes, inflammatory diseases or the immune surveillance of tumors.

Table 1 Necroptosis related diseases in human

Some physiological processes such as alteration of glucose level, oxygen deprivation or immune reactions resulted in elevated RIPK3 expression allowing in vivo emergence of necroptosis. Hyperglycemia (35–40 mM glucose) markedly enhanced the expression of RIPK3 in various cell lines and primed cells for necroptosis18,19. Similarly, upregulated expression of RIPK1, RIPK3 and MLKL, and increased RIPK1/3 complex formation have been observed in hypoxic cells20,21,22. At the same time caspase-8 mRNA, functioning as a negative regulator of necroptosis, was reported to be transiently decreased following the deprivation of oxygen and glucose (OGD)23. These processes are also involved in brain injury caused by hypoxia-ischemia and OGD-induced necroptosis24,25. Type I26,27,28 and type II27,29 interferons have been published to induce increased expression of RIPK3, while constitutive IFNβ signaling was demonstrated to increase the intracellular level of MLKL28. CD8+T lymphocytes can trigger both apoptosis and necroptosis, which make these cells capable of killing tumor cells, even those that escaped apoptosis30. T cell-mediated necroptotic cytolysis also plays a role in activation induced cell death, and can be critical in the development of autoimmune reactions31.

Upregulation of necroptosis in human diseases

Necroptosis takes part in the pathogenesis of human neurodegenerative disorders, such as Multiple Sclerosis (MS)32, Alzheimer’s disease (AD)33,34, and Amyotrophic Lateral Sclerosis (ALS)35,36. Defects in the activation of caspase-8 were demonstrated in the pathologic process of MS. Additionally, activated forms of RIPK1, RIPK3 and MLKL were detected in the cortical lesions of human MS samples32. Activated RIPK1 as a marker of necroptosis was also observed in human AD brains correlating positively with Braak stage and negatively with brain mass and cognition33,34. In ALS samples, multiple biochemical hallmarks of necroptosis including increased levels of RIPK1, RIPK3 and MLKL and elevated pRIPK1 and pMLKL were detected in both microglia and oligodendrocytes. Importantly, pMLKL was primarily localized in the white matter, where demyelination was found35. In spinal cord injury strong RIPK3 expression and MLKL phosphorylation were detected37.

In certain cardiovascular diseases, such as chronic heart failure (HF) cell loss and subsequent deterioration of contractile function is associated with elevated expression of RIPK1, RIPK3, and pRIPK3. On the other hand, the expression of caspase-8 was downregulated suggesting activation of necroptosis signaling. MLKL expression did not differ among the control and HF groups; however, pMLKL were present in all HF samples, which is in contrast to the controls where this was almost undetectable38. A genetic variant in the RIP3 promoter region was associated with increased RIPK3 transcription, which contributed to the poor prognosis of HF patients39.

In humans with unstable carotid atherosclerosis, expression of RIPK3 and MLKL was increased, while the phosphorylation of MLKL was detected in advanced atheromas40. In patients with abdominal aorta aneurysm, the tissue showed elevated levels of RIPK1 and RIPK3 proteins41,42. In coronary artery disease higher plasma RIPK3 levels were detected than in controls43.

Regarding gastrointestinal diseases, increased RIPK3 expression was detected in liver biopsies from patients with alcoholic liver disease17, while both RIPK3 and MLKL expression was increased in non-alcoholic fatty liver diseases44,45, as well as elevated MLKL phosphorylation in drug-induced liver injury46. High levels of RIPK3 and MLKL phosphorylation were also detected in the liver biopsies of patients with primary biliary cholangitis, in contrast with its low hepatic expression in healthy controls47. Similarly, increased levels of RIPK3 were documented in the terminal ileum of patients with Crohn’s disease17 and elevated RIPK3 and MLKL levels were observed in inflamed tissues of inflammatory bowel disease (IBD) and allergic colitis patients, whereas the expression of caspase-8 in these tissues was reduced48. The migration of human neutrophils to sites of inflammation was found to activate the RIPK3-MLKL pathway: a strong pMLKL signal was observed in infiltrating tissue neutrophils in samples collected from patients with cutaneous vasculitis, ulcerative colitis, and psoriasis49,50.

Phosphorylation of MLKL molecules was also detected in human acute kidney injury biopsies51, in biopsies taken immediately after excision for transplantation52 and in autosomal dominant polycystic kidney disease53 representing involvement of necroptosis in renal disorders. Antineutrophil cytoplasmic antibody (ANCA) induces neutrophil extracellular traps via necroptosis and causes subsequent endothelial cell damage. ANCA-associated vasculitis exhibited a specific p-MLKL staining in glomerular neutrophils in human kidney biopsies54.

Concerning skin diseases, human biopsy samples obtained from patients with Lichen Planus (LP) and Systemic lupus erythematosus (SLE) confirm the role of necroptosis in their development. RIPK3 and MLKL activation was demonstrated in podocytes in renal biopsies from patients with lupus nephritis55. LP and SLE tissue sections showed enhanced epidermal expression of phosphorylated RIPK356. B cells from SLE patients also significantly displayed high expression levels of necroptosis-related genes57. As we already mentioned, phosphorylation of MLKL in the infiltrated human neutrophils was also found in cutaneous vasculitis and psoriasis49,50. Upregulation of RIPK3, and elevated MLKL phosphorylation were observed in the skin samples from patients with toxic epidermal necrolysis in correlation with unwanted necroptosis and subsequent inflammation58.

Expression of RIPK3 and dynamin-related protein 1 (Drp1) was increased in lung tissue homogenates collected from patients suffering from chronic obstructive pulmonary disease, proving the role of necroptotic cell death in pulmonary diseases59. In Kashin–Beck disease (KBD) necroptosis dominates as a cell death mechanism in the middle zone of cartilage from KBD children60. Necroptotic cell death is involved in the progression of chronic periodontitis, as gingival tissue in patients showed increased levels of RIPK1, RIPK3, and MLKL, as well as increased phosphorylation of MLKL61.

Although RIPK1 is one of the key molecules required for execution of necroptosis, patients with its complete deficiency due to homozygous mutations suffered from recurrent infections, early-onset of IBD and progressive polyarthritis. In vitro, cells with RIPK1 deficiency showed impaired mitogen-activated protein kinase activation and cytokine secretion and were prone to necroptosis62,63.

Role of necroptosis in cancers

An increasing number of studies have been published about the importance of necroptotic cell death in anti-cancer therapies, which have been extensively reviewed in recent papers64,65.

Briefly, both pro- and anti-tumoral effects have been demonstrated following necroptosis in cancer development and progression. The anti-tumoral effect of necroptosis has been shown in many types of cancer in which the expression of RIPK366,67 or MLKL68 was silenced or polymorphisms in their coding genes lead to modified expression of necrosomal components66,69. In general, necroptosis resistance of cancer cells is a common process, and escape from necroptosis was suggested to be a potential hallmark of cancer, similar to the escape from apoptosis64. Additionally, effective anti-cancer agents trigger immunogenic cell death, inducing the killing of the transformed cells and provoking the members of innate and adaptive immune system to attack. Beside the massive release of DAMPs, necroptotic cells create a great possibility to trigger the activation of CD8 + T cells via cross presentation15,70. The dual ability of necroptosis to activate innate and adaptive immunity simultaneously makes this cell death pathway a promising therapeutic target.

However, the tumor-promoting outcome of necroptosis has also been shown. RIPK3 and MLKL expression seems to vary among tissue samples from different subtypes and stages of cancer, and downregulation of necroptosis mediators has also been published in various cancers71,72,73. Upregulated RIPK3 expression is a general phenomenon in tumor necrotic areas playing a critical role in tumor growth and metastasis74. Necroptosis-induced inflammation contributes to tumorigenesis and necroptosis can also lead to an immunosuppressive tumor microenvironment75. The immune-suppressing environment was associated with necroptosis-induced expression of the chemokine attractant CXCL171. It has also been shown that tumor cells induce necroptosis of endothelial cells, which promotes tumor cell extravasation and metastasis76. Thus, we can conclude that necroptosis occurs in different phases during tumorigenesis and plays an ambivalent role in tumor formation.

Molecular mechanisms in the regulation of necroptosis

To understand the molecular background of necroptosis and to find potential points of clinical intervention we summarize below how the expression, the posttranslational modification, and the localization of key necroptotic molecules (RIPK1, RIPK3 and MLKL) are regulated, while also highlighting the interaction partners of the necrosome complex.

Regulation the expression level of necroptotic proteins

RIPK3-RIPK3 homodimerization is sufficient to induce necroptosis; after which, its kinase domain stimulates the activation of RIPK3 through cis-autophosphorylation; a prerequisite step for the recruitment of MLKL77,78,79. Thus, RIPK3 dimerization is probably the most critical point of necroptosis induction. Several lines of evidence support the idea that increased expression of RIPK3 can induce its oligomerization and can initiate necroptosis42,80. RIPK1 dimerization, and accordingly upregulation of RIPK1, facilitates RIPK3 oligomerization, mainly upon death receptor stimuli.

All aspects of necroptotic protein expression are intensely regulated, including their transcriptional activity, the stability of the expressed molecules and their degradation. Specificity protein 1 (Sp1), a zinc-finger transcription factor, directly regulates RIPK3 expression in cancer cells. Knockdown of endogenous Sp1 significantly decreases the transcription of RIPK3, while re-expression of Sp1 restores necroptotic response in vitro81. Induction of necroptosis by interferon gamma (IFN-γ) resulted in elevated levels of RIPK327 and MLKL28,29,82. This effect was found to depend on janus kinase 1 (JAK1) and its substrates: the signal transducer and activator of transcription 1 (STAT1) and interferon regulatory factor (IRF) transcription factors, pinpointing interferon-stimulated gene factor 3 (ISGF3) as a critical promoter83. Bromodomain-containing protein 4 (BRD4), a member of the bromodomain and extraterminal domain (BET) family, has been shown to interact IRF1 and to upregulate MLKL transcription84. Oncogenes such as BRAF and AXL have also been implicated in the regulation of RIPK3 expression67. The activity of RIPK3 promoter is tightly controlled by methylation67,85,86,87 (Fig. 2a). Ubiquitin-like PHD and RING finger domain-containing protein 1 (UHRF1) is essential for the maintenance of the hypermethylation of the RIPK3 promoter and thus contributes to the silencing of RIPK3 expression in quiescent cells.

Fig. 2: Direct interacting partners of main necroptotic signaling molecules.
figure 2

Sp1 transcription factor increases RIPK3 expression. INFγ-mediated up-regulation of RIPK3 and MKLK level depend on JAK1 kinase, and STAT1 and IRF transcription factors. BRD4 cooperating with IRF1 also increase MLKL transcription. Hypermethylation of the RIPK3 promoter by UHRF1 results in silenced RIPK3 expression. The stability of all RIPK1, RIPK3 and MLKL proteins are increased by HSP90 and CDC37 co-chaperone complex and by FKBP12. The level of both RIPK1 and RIPK3 are down-regulated by caspase-8-mediated cleavage. Cathepsins are also capable of processing RIPK1. A20, CHIP, Optn, PELI1 and Triad3a ubiquitin-ligases mediate K48-linked polyubiquitylation and the subsequent proteasome dependent degradation of: RIPK1, RIPK3 and/or MLKL Upon necroptosis human RIPK1 is autophosphorylated at ser14, ser15, ser161, ser166 and RIPK3 at ser199 and ser227 and ser277. The transient phosphorylation of RIPK1 at ser321 is phosphorylated transiently by TAK1 leads to RIPK1-independent apoptosis and the sustained phosphorylation of RIPK1 by TAK1 at ser321, ser332, ser334 and ser336 induces RIPK1 kinase activation106. IKKα/IKKβ also phosphorylate RIPK1 at ser25 and thereby block RIPK1 activity108,214,215. Mitogen-activated protein kinase-activated protein kinase 2 (MK2) mediates phosphorylation of RIPK1 at ser321 and ser336 and restrains integration of RIPK1 into the cytosolic death complex107,216,217. The phosphorylation at ser89 by a currently unknown kinase inhibits the RIPK1 kinase activity218. Ubiquitylation of RIPK1 at Lys115 by PELI219 or Lys377 by cIAP1, cIAP2 and Parkin220 promotes necroptosis. LUBAC complex and the deubiquitinase CYLD regulates M1 ubiquitination of RIPK1221. Lys363 ubiquitylation of RIPK3 leads to its proteasomal degradation. RIPK3 is responsible for the phosphorylation of MLKL at thr357 and ser358. TAM (Tyro3, Axl, and Mer) family of receptor tyrosine kinases phosphorylate MLKL on Tyr376 to facilitate MLKL oligomerization145. MLKL is also phosphorylated on Ser441 by a still unidentified kinase222. Caspase-8 mediates the cleavage and inactivation of RIPK1 at asp324 and RIPK3 at asp328. O-GlcNAcylation of the RIPK3 at thr467 by OGT prevents necroptosis223. Red names indicate interaction partners of RIPK1, RIPK3, MLKL which activate necroptosis, blue marks necroptosis inhibitors

Following transcriptional regulation multiple processes control the protein level of necrosome components. The heat shock protein 90 (HSP90) and CDC37 co-chaperone complex increases the stability of all RIPK188, RIPK389, and MLKL90 proteins. Consequently, inhibitors of HSP90 facilitated the degradation of these necroptotic components and potently blocked necroptosis91. Protein levels of RIPK1 and RIPK3 also decreased in FK506-binding protein 12 (FKBP12) knockdown cells92.

On the contrary, cells treated with Hsp70 inhibitors underwent cell death, because Hsp70 enhances the stability of necroptosis antagonists, the RIPK1 regulators: cIAP1/2, x-linked inhibitor of apoptosis protein (XIAP), and the cellular FLICE-like inhibitor protein (cFLIP)93.

The expression of necroptotic molecules are downregulated by cleavage and proteosomal degradation. The most well-known inhibitor of necroptosis, caspase-8 cleaves both RIPK19, RIPK394, and the necroptosis promoting deubiquitinase CYLD proteins11. In macrophages, cathepsins were also reported to be capable of processing RIPK1, which resulted in significant decrease in necroptotic cell death95.

Several ubiquitin-ligases mediate K48-linked polyubiquitylation and the subsequent proteasome dependent degradation of necroptotic molecules: RIPK1 is regulated by A2096, carboxyl terminus of Hsp70-interacting protein (CHIP; also known as STUB1)97, optineurin (Optn)35, Triad3a98, RIPK3 by CHIP97, Optn35, E3 ubiquitin ligase Pellino 1 (PELI1)99, and MLKL by Optn (Table 2)35. Knock down of any of these K48 ubiquitin-ligases increased the sensitivity of necroptosis in both in vitro and in vivo studies. (Fig. 2b).

Table 2 Molecules in necroptotic signaling

Posttranslational modifications in the regulation of necroptosis

Accumulating evidence suggests that cell death pathways are finely tuned by posttranslational modifications, such as ubiquitination and phosphorylation. Multiple excellent recent reviews go into extensive detail about the role of these processes in necroptosis100, therefore we only provide a brief overview of these processes below. These pathways are mentioned in the tables and figures of this manuscript in the interest of providing a comprehensive visual guide to these processes as well (Fig. 2c).

The necrosome is formed due to the phosphorylation driven assembly of RIPK1, RIPK3, and MLKL4,80,101. However several phosphorylation steps have been published to inhibit necroptosis, chief among them the transforming growth factor beta-activated kinase 1 (TAK1) complex, which is the most important hub for these necroptosis-dampening signals102,103. Various protein complexes are assembled along TNFR signaling; namely the survival (complex I), the apoptotic (complex IIa and IIb) and the necroptosis inducer (complex IIc) complexes. Upon activation TNFR recruits TRADD, RIPK1, TRAF2, TRAF5 proteins. The gathered E3 ubiquitin ligases, cIAP-1 and cIAP-2 molecules, and the linear ubiquitin chain assembly complex LUBAC (consisting of HOIP, HOIL-1L and Sharpin)104 polyubiquitinates RIPK1, and modified RIPK1 can now act as a scaffold for TAK1 and the IKK complex105 which molecules in many ways block RIPK1-mediated cell death pathways, and thus the formation of complex II:106,107,108 These mechanism are: (1) By inducing the activation of NFκB and MAPK signaling pathways and thereby increasing the transcription of several survival molecules such as cIAP1/2109 and FLIP110 (2) by blocking the binding of cell death related molecules to RIPK1111 and (3) by phosphorylating RIPK1106,108.

Interaction partners of necrosome components

The activity of necrosome components are also mediated by molecular interactions (Fig. 2d). Three molecules, aurora kinase A (AURKA), PPM1b, and HSP90 have been recently identified as binding partners of RIPK390,91,112,113 and/ or RIPK191,112 in resting cells. AURKA112 and PPM1b113 act as local inhibitors against spontaneous necroptosis, since their silencing induces necroptosis. PPM1b as a phosphatase prevents RIPK3 autophosphorylation in resting cells113. AURKA together with its downstream target, Glycogen synthase kinase 3β (GSK3β) regulates the formation of RIPK1-RIPK3 and RIPK3-MLKL complexes112. Silencing or blocking of AURKA, or inhibitors of GSK3β result in necroptosis without any other stimuli. Phosphorylation of GSK3β at Ser9 suppresses necroptosis through interfering with the formation of RIPK3-MLKL complex, however the direct targets of GSK3β still have not been identified. The third molecule which associates with RIPK3 in resting cells, HSP90, is required for proper activation of necroptosis. Formation of the HSP90–CDC37 complex is necessary for RIPK1–RIPK3 interaction, thus it mediates RIPK3 activation during necroptosis. Unsurprisingly HSP90 inhibitors can block TNF-induced systemic inflammatory response syndrome (SIRS) in rats91. Additionally, membrane tethered mucins have been shown to interact with RIPK1 to block necroptosis in human bronchial epithelial cells in vitro114.

The nuclear retinoic acid receptor gamma (RARγ) is released from the nucleus to initiate the formation of cell death signaling complexes by mediating RIPK1 dissociation from TNFR when cIAP activity is blocked. In vitro silencing of RARγ inhibited necroptosis and in vivo results also confirmed that RARγ was essential for TNF-induced RIPK1-initiated apoptosis and necroptosis (Table 2)115.

Although RIPK1 initiates RIPK3 activation during death receptor driven necroptosis, it plays an ambivalent role in the regulation of RIPK3 aggregation. Under special circumstances instead of activation, RIPK1 acts to suppress the spontaneous activation of RIPK3 by TIR-domain-containing adapter-inducing interferon-β (TRIF)116 or DNA-dependent activator of IFN-regulatory factors (DAI; also known as ZBP1)78,117. RIPK3 oligomerization is able to seed a RHIM dependent oligomer and this process is both sufficient and a necessary step in necroptosis. RHIM domains of RIPK1 intrinsically inhibit RHIM-mediated RIPK3 aggregation by competing with the RHIM domain of TRIF or DAI; conversely death domain-driven RIPK1 oligomerization results in RIPK3 aggregation and necroptosis. In vivo results also reveal a kinase-independent function for RIPK1 in inhibiting necroptosis. Caspase-8/RIPK1 double-knockout animals die shortly after birth, however, additional ablation of RIPK3 to make caspase-8/RIPK1/RIPK3 triple knockouts rescues the viability of these animals117,118,119,120. These data undoubtedly prove the anti-necroptotic activity of RIPK1 under special conditions78.

MLKL association with RIPK3 is also suppressed by a constitutive interaction of MLKL with a competitive inhibitor, TRAF2, in resting cells. TRAF2 deubiquitination by CYLD promotes the dissociation of TRAF2 from MLKL and allows necroptosis121. Two other molecules inhibit cell death by blocking MLKL association with pro-necroptotic components: Repulsive guidance molecule b (RGMb) inhibits MLKL membrane translocation or membrane binding122 and Redox regulator thioredoxin-1 (TRX1) blocks MLKL disulfide bond formation, and through it the critical polymerization of MLKL123.

Various molecules have been published to act as downstream targets of RIPK3 and others to regulate MLKL localization and/or activation. RIPK3 constitutes an important upstream kinase of death associated protein (Daxx), triggering its nuclear export. The Akt/mTOR pathway124,125,126, and Ca2+/calmodulin-dependent protein kinase II (CaMKII)127 are also active effectors of downstream necroptotic signaling. Accordingly, several models suggest that effects on these signaling routes modify necroptotic intensity. Poly [ADP-ribose] polymerase 1 (PARP-1)128 (debated in ref. 129,130) and phosphoglycerate mutase family member 5 (PGAM5)131 (debated in ref. 132) have been documented as cell type specific regulators of downstream necroptotic events (Table 2).

Glucose metabolism and ROS production in necroptosis

Reactive oxygen species (ROS) have long been considered to contribute to necroptosis49,133,134,135. Oxidation of specific cysteine residues in RIPK1 by ROS activates RIPK1 autophosphorylation. A positive feedback loop is generated because silencing of RIPK1 or RIPK3 reduces ROS production. RIPK1 autophosphorylation is also promoted by mitochondrial ROS and is essential for RIPK3 recruitment into the necrosome. However, necroptosis could occur without ROS induction in some cell lines135,136.

Metabolic enzymes − human liver glycogen phosphorylase (PYGL), glutamate-ammonia ligase (GLUL), glutamate dehydrogenase 1 (GLUD1) − increase pyruvate production from glycogen or play a role in glutamine catabolism. These enzymes are activated by RIPK3, resulting in enhancement of aerobic respiration and thus likely contribute to TNF-induced ROS production80. Pyruvate dehydrogenase complex (PDC) converts pyruvate to acetyl-CoA, and triggers the entrance of metabolic flux into the tricarboxylic acid cycle. Activated RIPK3 in the necrosome enhances PDC activity by phosphorylating the PDC E3 at T135 and plays a major role in increasing aerobic respiration. Based on in vitro studies, activation of these enzymes has additive effects to aerobic respiration and ROS production (Table 2)80,134.

Intracellular localization of necrosome components

The intracellular localization of necrosome components seems to be crucial in the regulation of necroptosis. The RHIM domain of RIPK1 and RIPK3 mediates the assembly of heterodimeric filamentous structures, and the amyloid-like aggregation of RIPK1/RIPK3 complexes79. Compromised cluster formation correlated with decreased programmed necrosis. MLKL has also been reported to form SDS-resistant, disulfide bond-dependent polymers during necroptosis and it has been shown that these MLKL polymers were independent of RIPK1/RIPK3 fibers137.

MLKL translocation to the cell membrane is an obligatory step in necroptotic signaling. Phosphatidyl-inositol phosphates (PIPs) as critical binders of MLKL are required for plasma membrane targeting of MLKL and subsequent membrane permeabilization in necroptosis138,139. Highly phosphorylated inositol products, but not weakly phosphorylated precursors are able to displace the MLKL auto-inhibitory brace region, which is a necessary event for late plasma membrane breakdown and cell death140,141. Accordingly, necroptosis requires inositol polyphosphate-specific kinase activity and in cells containing mutant IP kinases, MLKL failed to oligomerize and localize to membranes despite proper RIPK3-dependent phosphorylation142. Deletion of inositol polyphosphate multikinase (IPMK), inositol-tetrakisphosphate 1-kinase (ITPK1)142 or inositol pentakisphosphate 2-kinase (IPPK)143 inhibited necroptosis. Connected to this, phosphatidylinositol transfer protein alpha (PITPα) interacts with MLKL which facilitates MLKL oligomerization and plasma membrane translocation144. Following membrane localization TAM (Tyro3, Axl, and Mer) family of receptor tyrosine kinases phosphorylate MLKL to protmote MLKL oligomerizatin and necroptosis145. Beside their direct pore forming ability, membrane-localized MLKL regulates transient receptor potential cation channel, subfamily M, member 7 (TRPM7), a non-voltage-sensitive ion channel, for the mediation of Ca2+ influx146,147.

Once MLKL is membrane associated, all the endosomal sorting complexes required for transport III machinery (ESCRT-III), flotillin-mediated endocytosis and ALIX-syntenin-1-mediated exocytosis act to sustain survival of the cell. The ESCRT-III-driven plasma membrane repair machinery limits the duration of the loss of plasma membrane integrity upon MLKL activation52,148, while endo- and exocytosis removes phospho-MLKL from the plasma mebrane149. MLKL also forms a complex with multiple membrane metalloproteinases upon necroptotic stimulus. A disintegrin and metalloproteinase (ADAM)-enzymes are activated to mediate the shedding of cell-surface proteins in response to necroptotic stimuli and through this process also play a key role in promoting necroptosis, but only in adherent cells (Table 2)150.

RIPK1151, RIPK3152,153, and MLKL154,155 have all been reported to localize to the nucleus and these translocations preceded necroptotic death154. RIPK3 and MLKL have been shown to became activated in the nucleus, and after their cooperative nuclear export, they contribute to cytosolic necrosome formation155. Following the interaction of RIPK3 and MLKL, the translocation of this complex to mitochondria-associated membranes has also been demonstrated and this relocation was found to be essential for necroptosis signaling156. The intracellular trafficking of necrosomes is regulated by the TNF-induced guanine nucleotide-binding protein γ 10 (Gγ10) – Src signaling pathway157, however, RIPK1/RIPK3 kinase activity has no direct interaction with Gγ10 or on Src kinase.

Drugs to regulate necroptosis intensity

In vitro studies prefer to use caspase inhibitors to activate necroptosis, however we still do not fully understand how necroptosis is activated under physiological conditions. The in vivo appearance of necroptosis indicates that in addition to caspase-mediated processes various caspase independent regulatory mechanisms control necroptosis. Drugs affecting either the expression or the activity of necroptosis mediators, or that modify the indirect regulators of necroptosis may have therapeutic potential (Tables 3 and 4).

Table 3 Available drugs to modify necroptosis intensity
Table 4 Components of traditional medicine as necroptosis regulators

Regulation the expression level of necrosome components

Drugs that control the promoters of RIPK3 or MLKL or modify the stability and degradation of these molecules can regulate necroptosis sensitivity. Interferons27,29, hypomethylating agents such as decitabine (5-aza-2′-deoxycytidine) and 5-azacytidine (used in Myelodysplastic syndromes and AML)87, histone deacetylase inhibitor valproic acid158 (VPA), anti-fungal miconazole159, traditional Chinese medicine drugs (shikonin160,161, resibufogenin162, bufalin163, youdujing164, emodin165), and components found in different plants (matrine166, genipine167, lycorine168, quercetin169, curcumol170, Bulnesia sarmientoi171) were all found to upregulate the expression of RIPK1 or RIPK3.

On the other hand, various inhibitors of the HSP90 have been documented to downregulate necroptosis (Kongensin A172, G-TPP173, geldanamycin174, gamitrinib10, DHQ3175 and 17-demethoxy-reblastatin175). Cyclosporine A176, Diacerein177 (Used in Europe and Asia to treat joint diseases), immunosuppressive and antiproliferative Rapamycin178 and traditional Chinese medicine such as patchouli alcohol179 have been also documented to reduce the expression of principal necroptotic mediators. Ex-527180 (which completed a phase II clinical trial in Huntington disease) regulates necroptosis through the inhibition of Sirt1 deacetylase.

Regulation the activity of necrosome components

Beside the expression of necrosome components, the activity of these enzymes is also modified by various drugs. Promising specific inhibitors are currently being developed for the central molecules of necroptosis. RIPK1, RIPK3, and MLKL (reviewed in refs. 181,182) which may interfere with unwanted cell death and subsequent inflammation. Multiple second mitochondria-derived activator of caspase (SMAC) mimetics and TAK-1 (reviewed in ref. 183,184) inhibitors are being tested in clinical trials to activate necroptosis for therapeutical intervention, by restoring the sensitivity of apoptosis-resistant tumors to cell death. Since these drugs are reviewed elsewhere, we focus on currently available necroptosis regulators.

Drugs currently used for the treatment of different forms of tumors display anti-necroptotic activity (Dabrafenib185,186, Sorafenib187,188, Pazopanib189, Ponatinib189, and Carfilzomib190) as does the anti-epilepsy drug Phenhydan191. Phenytoin51 (a clinically used anti-convulsant) or herbal components such us wogonin192 and aucubin193 inhibit RIPK1 activity. All these drugs provide immediate translational potential to dampen necroptosis-driven tissue degradation. Presumably, these drugs will be additive to the above-mentioned necroptosis inhibitors which downregulate the expression of necrosome components.

On the other hand, radiation194, or chemotherapeutic agents such as anthracyclines and oxaliplatin195, cisplatin196,197, 5-fluorouracil198 or the pan-BCL-2 inhibitor Obatoclax199 (several phase two trials have been completed), traditional Chinese medicines such as resibufogenin162 (also tested in phase II of a clinical trial on pancreatic cancer), aucubin193, tanshinone200 or neoalbaconol201 have been documented to upregulate necroptosis. Based on current results, these drugs regulate the activity, and not the expression of necroptotic component. As a mono-therapy these group of necroptosis regulators could be ineffective in tumors that downregulate the level of RIPK3 or MLKL, but these medicines may increase the effect of the above listed mediators in combination therapy following the restoration of RIPK1 or RIPK3 expression in cancer cells.

Regulation the signaling of necroptotic pathway

Some drugs regulate necroptosis by modulating the level or activity of partner molecules of the necrosome. For example, VPA induces the release of SMAC from mitochondria thereby upregulating necroptosis similarly to the widely tested SMAC mimetics. Dimethyl fumarate (DMF202, which is currently used in relapsing-remitting multiple sclerosis) induces necroptosis via downregulation of the negative regulators of necroptosis such as IAPs and cFLIPs. Aurora kinase inhibitors have been shown to directly induce necroptosis and stimulated intra-tumoral phosphorylation of MLKL203. Drugs antagonizing Trx1function as necroptosis inducers. PX-12123 (completed phase I of a clinical trial on advanced metastatic cancer) and DMF204 target TRX1 and have been shown to sensitize tumor cells to necroptosis.

Various drugs activate necroptosis via regulation of downstream components of necroptosis. Adiponectin receptor agonists205 (tested in various clinical trials), DMF202, neoablaconol206 induce ROS production. Lithium126 (clinically used for treating bipolar disorders) facilitates AKT-mTOR-mediated necroptosis, while dasatinib (used drug in CML) induces HMGB1-mediated necroptosis.

Necroptosis can be inactivated via the regulation of interacting partners of the necrosome or by downstream components, as well. The proteasome inhibitor Bortezomib207 (used in Multiple Myeloma treatment) and a HDAC inhibitor Vorinostat208 (approved for the treatment of Cutaneous T cell lymphoma) have been demonstrated to inhibit necroptosis through the upregulation of necroptosis inhibitors, sequentially stabilizing IAPs or increasing FLIP expression.

Various ROS scavengers have been implicated in the modulation of necroptosis209,210. Dexmedetomidine (used in moderate sedation) inhibits HMGB1 production211. Melatonin212 (used for jetlag sleep disorder) blocks PGAM5, while P110 is a selective inhibitor of Drp1213, therefore, these two drugs dampen the intensity of necroptosis via a well-documented PGAM-Drp1 pathway.

While there are no drugs on the market directly approved to regulate necroptosis, various medicines have the potential to both up and downregulate necroptosis, and to interact different levels of necroptosis signaling. Necroptosis has fundamental roles in various human diseases which makes it rational to try and apply the necroptosis regulator drugs in these syndromes.