Creation of mouse TNFR2-selective agonistic TNF mutants using a phage display technique

Tumor necrosis factor-α (TNF), which is an immuno-modulatory cytokine, has been suggested to cause inflammatory responses as well as protection against tissue dysfunction by binding two types of TNF receptor (TNFR1/TNFR2). However, the physiological effects of TNFR2-specific activation remain unclear. We therefore aimed to generate a TNF mutant with full TNFR2-selective agonist activity as a functional analytical tool. In this study, we utilized a phage display technique to create mouse TNFR2 (mTNFR2)-selective TNF mutants that bind specifically to mTNFR2 and show full bioactivity compared with wild-type TNF. A new phage library displaying TNF mutants was created, in which nine amino acid residues at the predicted receptor-binding site were randomized. From this library, an agonistic TNF mutant exhibiting high binding selectivity and bioactivity to mTNFR2 was isolated. We propose that this TNF mutant would be a powerful tool with which to elucidate the functional roles of mTNFR2.     

Membrane Tumor Necrosis Factor (TNF) Induces p100 Processing via TNF Receptor-2 (TNFR2)

Tumor necrosis factor (TNF) elicits its biological activities by stimulation of two receptors, TNFR1 and TNFR2, both belonging to the TNF receptor superfamily. Whereas TNFR1-mediated signal transduction has been intensively studied and is understood in detail, especially with respect to activation of the classical NFκB pathway, cell death induction, and MAP kinase signaling, TNFR2-associated signal transduction is poorly defined. Here, we demonstrate in various tumor cell lines and primary T-cells that TNFR2, but not TNFR1, induces activation of the alternative NFκB pathway. In accord with earlier findings demonstrating that only membrane TNF, but not soluble TNF, properly activates TNFR2, we further show by use of TNFR1- and TNFR2-specific mutants of soluble TNF and membrane TNF that soluble ligand trimers fail to activate the alternative NFκB pathway. In accord with the known inhibitory role of TRAF2 in the alternative NFκB pathway, TNFR2-, but not TNFR1-specific TNF induced depletion of cytosolic TRAF2. Thus, we identified activation of the alternative NFκB pathway as a TNF signaling effect that can be specifically assigned to TNFR2 and membrane TNF.     

A TNF receptor 2 selective agonist rescues human neurons from oxidative stress-induced cell death

Tumor necrosis factor (TNF) plays a dual role in neurodegenerative diseases. Whereas TNF receptor (TNFR) 1 is predominantly associated with neurodegeneration, TNFR2 is involved in tissue regeneration and neuroprotection. Accordingly, the availability of TNFR2-selective agonists could allow the development of new therapeutic treatments of neurodegenerative diseases. We constructed a soluble, human TNFR2 agonist (TNC-scTNF(R2)) by genetic fusion of the trimerization domain of tenascin C to a TNFR2-selective single-chain TNF molecule, which is comprised of three TNF domains connected by short peptide linkers. TNC-scTNF(R2) specifically activated TNFR2 and possessed membrane-TNF mimetic activity, resulting in TNFR2 signaling complex formation and activation of downstream signaling pathways. Protection from neurodegeneration was assessed using the human dopaminergic neuronal cell line LUHMES. First we show that TNC-scTNF(R2) interfered with cell death pathways subsequent to H(2)O(2) exposure. Protection from cell death was dependent on TNFR2 activation of the PI3K-PKB/Akt pathway, evident from restoration of H(2)O(2) sensitivity in the presence of PI3K inhibitor LY294002. Second, in an in vitro model of Parkinson disease, TNC-scTNF(R2) rescues neurons after induction of cell death by 6-OHDA. Since TNFR2 is not only promoting anti-apoptotic responses but also plays an important role in tissue regeneration, activation of TNFR2 signaling by TNC-scTNF(R2) appears a promising strategy to ameliorate neurodegenerative processes.     

Novel protein engineering strategy for creating highly receptor-selective mutant TNFs

Tumor necrosis factor (TNF) plays important roles in host defense and in preventing tumor formation by acting via its receptors, TNFR1 and TNFR2, functions of which are less understood. To this end, we have been isolating TNF receptor-selective mutants using phage display technique. However, generation of a phage library with large repertoire (>10⁸) is impeded by the limited transformation efficiency of Escherichia coli. Therefore, it is currently difficult to create a mutant library containing amino acid substitutions in more than seven residues. To overcome this problem, here we have used two different TNF mutant libraries, each containing random substitutions at six selected amino acid residues, and utilized a gene shuffling method to construct a randomized mutant library containing substitutions at 12 different amino acid residues of TNF. Consequently, using this library, we identified TNF mutants with greater receptor-selectivity and enhanced receptor-specific bioactivity than the existing mutants.     

Creation and X-ray structure analysis of the tumor necrosis factor receptor-1-selective mutant of a tumor necrosis factor-alpha antagonist

Tumor necrosis factor-alpha (TNF) induces inflammatory response predominantly through the TNF receptor-1 (TNFR1). Thus, blocking the binding of TNF to TNFR1 is an important strategy for the treatment of many inflammatory diseases, such as hepatitis and rheumatoid arthritis. In this study, we identified a TNFR1-selective antagonistic mutant TNF from a phage library displaying structural human TNF variants in which each one of the six amino acid residues at the receptor-binding site (amino acids at positions 84-89) was replaced with other amino acids. Consequently, a TNFR1-selective antagonistic mutant TNF (R1antTNF), containing mutations A84S, V85T, S86T, Y87H, Q88N, and T89Q, was isolated from the library. The R1antTNF did not activate TNFR1-mediated responses, although its affinity for the TNFR1 was almost similar to that of the human wild-type TNF (wtTNF). Additionally, the R1antTNF neutralized the TNFR1-mediated bioactivity of wtTNF without influencing its TNFR2-mediated bioactivity and inhibited hepatic injury in an experimental hepatitis model. To understand the mechanism underlying the antagonistic activity of R1antTNF, we analyzed this mutant using the surface plasmon resonance spectroscopy and x-ray crystallography. Kinetic association/dissociation parameters of the R1antTNF were higher than those of the wtTNF, indicating very fast bond dissociation. Furthermore, x-ray crystallographic analysis of R1antTNF suggested that the mutation Y87H changed the binding mode from the hydrophobic to the electrostatic interaction, which may be one of the reasons why R1antTNF behaved as an antagonist. Our studies demonstrate the feasibility of generating TNF receptor subtype-specific antagonist by extensive substitution of amino acids of the wild-type ligand protein.     

Dissociation of TNF-alpha cytotoxic and proinflammatory activities by p55 receptor- and p75 receptor-selective TNF-alpha mutants.

Human tumour necrosis factor alpha (TNF-alpha) is a pleiotropic cytokine capable of killing mammalian tumour cells in vitro and in vivo, and of enhancing the proinflammatory activity of leucocytes and endothelium, the latter effects limiting its usage as an antitumour agent in humans. Using TNF-alpha mutants with a selective capacity to bind to the TNF p55 receptor (TNFR55) or to the p75 receptor (TNFR75) we show here that these two major activities of TNF-alpha can be dissociated. The TNFR55-selective mutants (R32W, E146K and R32W-S86T) which bind poorly to TNFR75 displayed similar potency to wild-type TNF in causing cytotoxicity of a human laryngeal carcinoma-derived cell line (HEp-2) and cytostasis in a human leukaemic cell line (U937). However, these TNFR55-selective mutants exhibited lower proinflammatory activity than wild-type TNF. Specifically, TNF-alpha's priming of human neutrophils for superoxide production and antibody-dependent cell-mediated cytotoxicity, platelet-activating factor synthesis and adhesion to endothelium were reduced by up to 170-fold. Activation of human endothelial cell functions represented by human umbilical venular endothelial cell (HUVEC) adhesiveness for neutrophils, E-selectin expression, neutrophil transmigration and IL-8 secretion were also reduced by up to 280-fold. On the other hand, D143F, a TNFR75-selective mutant tested either alone or in combination with TNFR55-selective mutants, did not stimulate these activities despite being able to cause cytokine production in TNFR75-transfected PC60 cells.(ABSTRACT TRUNCATED AT 250 WORDS)     

Ligand-induced internalization of TNF receptor 2 mediated by a di-leucin motif is dispensable for activation of the NFκB pathway

Endocytosis is an important mechanism to regulate tumor necrosis factor (TNF) signaling. In contrast to TNF receptor 1 (TNFR1; CD120a), the relevance of receptor internalization for signaling as well as the fate and route of internalized TNF receptor 2 (TNFR2; CD120b) is poorly understood. To analyze the dynamics of TNFR2 signaling and turnover at the plasma membrane we established a human TNFR2 expressing mouse embryonic fibroblast cell line in a TNFR1^−/−/TNFR2^−/− background. TNF stimulation resulted in a decrease of constitutive TNFR2 ectodomain shedding. We hypothesized that reduced ectodomain release is a result of TNF/TNFR2 complex internalization. Indeed, we could demonstrate that TNFR2 was internalized together with its ligand and cytoplasmic binding partners. Upon endocytosis the TNFR2 signaling complex colocalized with late endosome/lysosome marker Rab7 and entered the lysosomal degradation pathway. Furthermore, we identified a di-leucin motif in the cytoplasmic part of TNFR2 suggesting clathrin-dependent internalization of TNFR2. Internalization defective TNFR2 mutants are capable to signal, i.e. activate NFκB, demonstrating that the di-leucin motif dependent internalization is dispensable for this response. We therefore propose that receptor internalization primarily serves as a negative feed-back to limit TNF responses via TNFR2.     

IAPs limit activation of RIP kinases by TNF receptor 1 during development

Inhibitor of apoptosis (IAP) proteins cIAP1, cIAP2, and XIAP (X-linked IAP) regulate apoptosis and cytokine receptor signalling, but their overlapping functions make it difficult to distinguish their individual roles. To do so, we deleted the genes for IAPs separately and in combination. While lack of any one of the IAPs produced no overt phenotype in mice, deletion of cIap1 with cIap2 or Xiap resulted in mid-embryonic lethality. In contrast, Xiap(-/-)cIap2(-/-) mice were viable. The death of cIap2(-/-)cIap1(-/-) double mutants was rescued to birth by deletion of tumour necrosis factor (TNF) receptor 1, but not TNFR2 genes. Remarkably, hemizygosity for receptor-interacting protein kinase 1 (Ripk1) allowed Xiap(-/-)cIap1(-/-) double mutants to survive past birth, and prolonged cIap2(-/-)cIap1(-/-) embryonic survival. Similarly, deletion of Ripk3 was able to rescue the mid-gestation defect of cIap2(-/-)cIap1(-/-) embryos, as these embryos survived to E15.5. cIAPs are therefore required during development to limit activity of RIP kinases in the TNF receptor 1 signalling pathway.     

Characterisation of receptor-specific TNFalpha functions in adipocyte cell lines lacking type 1 and 2 TNF receptors

Tumour necrosis factor-alpha (TNFalpha) is a multifunctional cytokine that exerts a myriad of biological actions in numerous different tissues including adipocytes through its two distinct cell surface receptors. To address the role of each TNF receptor in the biological actions of TNFalpha in adipocytes, we have developed four new preadipocyte cell lines. These were established from wild type controls (TNFR1(+/+)R2(+/+)) and from mice lacking TNFR1 (TNFR1(-/-)), TNFR2 (TNFR2(-/-)) or both (TNFR1(-/-)R2(-/-)). All four new cell lines can fully differentiate to form mature adipocytes, under appropriate culture conditions, as judged by cell morphology, expression of multiple adipogenic markers and the ability to mediate agonist-stimulated lipolysis and insulin-stimulated glucose transport. In wild type (TNFR1(+/+)R2(+/+)) and TNFR2(-/-) adipocytes, TNFalpha stimulated lipolysis and inhibited insulin-stimulated glucose transport as well as insulin receptor autophosphorylation. In contrast, these activities were completely lost in the TNFR1(-/-)R2(-/-) and TNFR1(-/-) cells. Taken together, these studies demonstrate that TNFalpha-induced lipolysis, as well as inhibition of insulin-stimulated glucose transport are predominantly mediated by TNFR1 and that the presence of TNFR2 is not necessary for these functions. This new experimental system promises to be useful in dissecting the molecular pathways activated by each TNF receptor in mediating the biological functions of TNFalpha in differentiated adipocytes.     

CYLD and the NEMO Zinc Finger Regulate Tumor Necrosis Factor Signaling and Early Embryogenesis

NF-κB essential modulator (NEMO) and cylindromatosis protein (CYLD) are intracellular proteins that regulate the NF-κB signaling pathway. Although mice with either CYLD deficiency or an alteration in the zinc finger domain of NEMO (K392R) are born healthy, we found that the combination of these two gene defects in double mutant (DM) mice is early embryonic lethal but can be rescued by the absence of TNF receptor 1 (TNFR1). Notably, NEMO was not recruited into the TNFR1 complex of DM cells, and consequently NF-κB induction by TNF was severely impaired and DM cells were sensitized to TNF-induced cell death. Interestingly, the TNF signaling defects can be fully rescued by reconstitution of DM cells with CYLD lacking ubiquitin hydrolase activity but not with CYLD mutated in TNF receptor-associated factor 2 (TRAF2) or NEMO binding sites. Therefore, our data demonstrate an unexpected non-catalytic function for CYLD as an adapter protein between TRAF2 and the NEMO zinc finger that is important for TNF-induced NF-κB signaling during embryogenesis.     

Down-regulation of occludin expression in astrocytes by tumour necrosis factor (TNF) is mediated via TNF type-1 receptor and nuclear factor-κB activation

Tight junctions form the diffusion barrier of brain microcapillary endothelial cells and support cell polarity. Also astrocytes express tight junction components such as occludin, claudin-1, ZO-1 and ZO-2, but do not establish a permeability barrier. However, little is known about the function and regulation of these molecules in astrocytes. We studied the impact of tumour necrosis factor (TNF) on occludin and ZO-1 expression in astrocytes. TNF decreased occludin, but not ZO-1 expression. In brain microcapillary endothelial cells, as well as in epithelial cells, occludin expression was not influenced by TNF. Removal of TNF from astrocytes restored the basal level of occludin. Down-regulation was inhibited by caffeic acid phenethyl ester, a specific inhibitor of nuclear factor-kappaB (NF-kappaB) activation. Exposure of astrocytes isolated from mice deficient in either TNF type-1 receptor (TNFR1), TNF type-2 receptor (TNFR2), or both, respectively, revealed that down-regulation was mediated entirely by TNFR1. ZO-1, which can interact with occludin, was found to co-precipitate connexin43, but not occludin. These findings demonstrate that TNF selectively down-regulates occludin in astrocytes, but not in cells forming established tight junctions, through TNFR1 and suggest that NF-kappaB is involved as a negative regulator.     

Fluorescence correlation spectroscopy reveals topological segregation of the two tumor necrosis factor membrane receptors

The proinflammatory cytokine tumor necrosis factor (TNF) binds two distinct plasma membrane receptors, TNFR1 and TNFR2. We have produced different receptor mutants fused with enhanced green fluorescent protein to study their membrane dynamics by fluorescence correlation spectroscopy (FCS). TNFR1 mutants show diffusion constants of approximately 1.2 × 10^− 9 cm²/s and a broad distribution of diffusion times, which is hardly affected by ligand binding. However, cholesterol depletion enhances their diffusion, suggesting a constitutive affinity to cholesterol rich membrane microdomains. In contrast, TNFR2 and mutants thereof diffuse rather fast (D? = 3.1 × 10^− 9 cm²/s) with a marked reduction after 30 min of TNF treatment (D? = 0.9 × 10^− 9 cm²/s). This reduction cannot be explained by the formation of higher ordered receptor clusters, since the fluorescence intensity of TNF treated receptors indicate the presence of a few receptor molecules per complex only. Together, these data point to a topological segregation of the two TNF receptors in different microcompartments of the plasma membrane independent of the cytoplasmic signaling domains of the receptors.     

ST6Gal-I regulates macrophage apoptosis via α2-6 sialylation of the TNFR1 death receptor

Macrophages play a central role in innate immunity, however mechanisms regulating macrophage survival are not fully understood. Herein we describe a novel apoptotic pathway involving α2-6 sialylation of the TNFR1 death receptor by the ST6Gal-I sialyltransferase. Variant glycosylation of TNFR1 has not previously been implicated in TNFR1 function, and little is known regarding the TNFR1 glycan composition. To study sialylation in macrophages, we treated U937 monocytic cells with PMA, which stimulates both macrophage differentiation and apoptosis. Interestingly, macrophage differentiation induces ST6Gal-I down-regulation, leading to reduced α2-6 sialylation of selected receptors. To prevent loss of α2-6 sialylation, we forced constitutive expression of ST6Gal-I, and found that this strongly inhibited PMA-induced apoptosis. Given that PMA-mediated apoptosis is thought to result from up-regulation of TNFα, which then activates TNFR1, we next evaluated the α2-6 sialylation of TNFR1. U937 cells with forced ST6Gal-I displayed TNFR1 with elevated α2-6 sialylation, and this was associated with diminished TNFα-stimulated apoptosis. Correspondingly, removal of α2-6 sialylation from TNFR1 through either neuraminidase treatment or expression of ST6Gal-I shRNA markedly enhanced TNFα-mediated apoptosis. To confirm the physiologic importance of TNFR1 sialylation, we generated overexpressing ST6Gal-I transgenic mice. Peritoneal macrophages from transgenic lines displayed TNFR1 with elevated α2-6 sialylation, and these cells were significantly protected against TNFα-stimulated apoptosis. Moreover, greater numbers of thioglycollate-induced peritoneal cells were observed in transgenic mice. These collective results highlight a new mechanism of TNFR1 regulation, and further intimate that loss of α2-6 sialylation during macrophage differentiation may limit macrophage lifespan by sensitizing cells to TNFα-stimulated apoptosis.     

TNF receptor 2 protects oligodendrocyte progenitor cells against oxidative stress

The neuroprotective role of TNF receptor 2 (TNFR2) has been shown in various studies. However, a direct role of TNFR2 in oligodendrocyte function has not yet been demonstrated. Using primary oligodendrocytes of transgenic mice expressing human TNFR2, we show here that TNFR2 is primarily expressed on oligodendrocyte progenitor cells. Interestingly, preconditioning with a TNFR2 agonist protects these cells from oxidative stress, presumably by increasing the gene expression of distinct anti-apoptotic and detoxifying proteins, thereby providing a potential mechanism for the neuroprotective role of TNFR2 in oligodendrocyte progenitor cells.     

Modulation of allergic inflammation in mice deficient in TNF receptors

Tumor necrosis factor-alpha (TNF) is implicated as an important proinflammatory cytokine in asthma. We evaluated mice deficient in TNF receptor 1 (TNFR1) and TNFR2 [TNFR(-/-) mice] in a murine model of allergic inflammation and found that TNFR(-/-) mice had comparable or accentuated responses compared with wild-type [TNFR(+/+)] mice. The responses were consistent among multiple end points. Airway responsiveness after methacholine challenge and bronchoalveolar lavage (BAL) fluid leukocyte and eosinophil numbers in TNFR(-/-) mice were equivalent or greater than those observed in TNFR(+/+) mice. Likewise, serum and BAL fluid IgE; lung interleukin (IL)-2, IL-4, and IL-5 levels; and lung histological lesion scores were comparable or greater in TNFR(-/-) mice compared with those in TNFR(+/+) mice. TNFR(+/+) mice chronically treated with anti-murine TNF antibody had BAL fluid leukocyte numbers and lung lesion scores comparable to control antibody-treated mice. These results suggest that, by itself, TNF does not have a critical proinflammatory role in the development of allergic inflammation in this mouse model and that the production of other cytokines associated with allergic disease may compensate for the loss of TNF bioactivity in the TNFR(-/-) mouse.     

Modulation of endotoxin-induced NF-kappa B activation in lung and liver through TNF type 1 and IL-1 receptors

We investigated the requirement for tumor necrosis factor-alpha (TNF-alpha) and interleukin (IL)-1 receptors in the pathogenesis of the pulmonary and hepatic responses to Escherichia coli lipopolysaccharide (LPS) by studying wild-type mice and mice deficient in TNF type 1 receptor [TNFR1 knockout (KO)] or both TNF type 1 and IL-1 receptors (TNFR1/IL-1R KO). In lung tissue, NF-kappaB activation was similar among the groups after exposure to aerosolized LPS. After intraperitoneal injection of LPS, NF-kappaB activation in liver was attenuated in TNFR1 KO mice and further diminished in TNFR1/IL-1R KO mice; however, in lung tissue, no impairment in NF-kappaB activation was found in TNFR1 KO mice and only a modest decrease was found in TNFR1/IL-1R KO mice. Lung concentrations of KC and macrophage-inflammatory peptide 2 were lower in TNFR1 KO and TNFR1/IL-1R KO mice after aerosolized and intraperitoneal LPS. We conclude that LPS-induced NF-kappaB activation in liver is mediated through TNF-alpha- and IL-1 receptor-dependent pathways, but, in the lung, LPS-induced NF-kappaB activation is largely independent of these receptors.     

TNF/TNFR1 signaling mediates doxorubicin-induced diaphragm weakness

Doxorubicin, a common chemotherapeutic agent, causes respiratory muscle weakness in both patients and rodents. Tumor necrosis factor-α (TNF), a proinflammatory cytokine that depresses diaphragm force, is elevated following doxorubicin chemotherapy. TNF-induced diaphragm weakness is mediated through TNF type 1 receptor (TNFR1). These findings lead us to hypothesize that TNF/TNFR1 signaling mediates doxorubicin-induced diaphragm muscle weakness. We tested this hypothesis by treating C57BL/6 mice with a clinical dose of doxorubicin (20 mg/kg) via intravenous injection. Three days later, we measured contractile properties of muscle fiber bundles isolated from the diaphragm. We tested the involvement of TNF/TNFR1 signaling using pharmaceutical and genetic interventions. Etanercept, a soluble TNF receptor, and TNFR1 deficiency protected against the depression in diaphragm-specific force caused by doxorubicin. Doxorubicin stimulated an increase in TNFR1 mRNA and protein (P < 0.05) in the diaphragm, along with colocalization of TNFR1 to the plasma membrane. These results suggest that doxorubicin increases diaphragm sensitivity to TNF by upregulating TNFR1, thereby causing respiratory muscle weakness.     

Mutants of lymphotoxin-α with augmented cytotoxic activity via TNFR1 for use in cancer therapy

The cytokine lymphotoxin-α (LTα) is a promising candidate for use in cancer therapy. However, the instability of LTα in vivo and the insufficient levels of tumor necrosis factor receptor 1 (TNFR1)-mediated bioactivity of LTα limit its therapeutic potential. Here, we created LTα mutants with increased TNFR1-mediated bioactivity by using a phage display technique. We constructed a phage library displaying lysine-deficient structural variants of LTα with randomized amino acid residues. After affinity panning, we screened three clones of lysine-deficient LTα mutant, and identified a LTα mutant with TNFR1-mediated bioactivity that was 32 times that of the wild-type LTα (wtLTα). When compared with wtLTα, the selected clone showed augmented affinity to TNFR1 due to slow dissociation rather than rapid association. In contrast, the mutant showed only 4 times the TNFR2-mediated activity of wtLTα. In addition, the LTα mutant strongly and rapidly activated caspases that induce TNFR1-mediated cell death, whereas the mutant and wtLTα activated nuclear factor-kappa B to a similar extent. Our data suggest that the kinetics of LTα binding to TNFR1 play an important role in signal transduction patterns, and a TNFR1-selective LTα mutant with augmented bioactivity would be a superior candidate for cancer therapy.     

Single chain TNF derivatives with individually mutated receptor binding sites reveal differential stoichiometry of ligand receptor complex formation for TNFR1 and TNFR2

Most members of the tumor necrosis factor ligand family form noncovalently linked homotrimers, capable to bind up to three molecules of the respective membrane receptors. For several receptors a membrane distal homophilic interaction domain has been identified, called pre-ligand binding assembly domain. Accordingly, affinity values determined by typical equilibrium binding studies are likely to be influenced by avidity effects. Using our recently introduced covalently stabilized TNF (single chain TNF, scTNF), we have here investigated receptor–ligand binding stoichiometry in our well characterized system of TNFR–Fas chimeras. We produced scTNF derivatives with functionally deleted individual receptor binding sites, resulting in TNF mutants capable to only bind to one or two receptor molecules, rather than three. Equilibrium binding affinity studies on ice with these molecules revealed no significant changes after a single receptor binding site had been functionally deleted. In contrast, functional abrogation of two receptor binding sites showed a strong decrease in both, affinity and bioactivity on TNFR2–Fas. In contrast, TNFR1–Fas ligand binding and receptor activation was only affected after functional deletion of all three receptor binding sites. Our data demonstrate pivotal differences in ligand/receptor interactions between TNFR1–Fas and TNFR2–Fas, arguing for avidity effects important for TNF binding and downstream signaling of TNFR2, but to a lesser extent of TNFR1. These results are supported by data revealed from chemical crosslinking experiments suggesting the existence of preformed TNFR–Fas homodimers.