The tumor necrosis factor receptor stalk regions define responsiveness to soluble versus membrane-bound ligand

Abstract

The family of tumor necrosis factor receptors (TNFRs) and their ligands form a regulatory signaling network that controls immune responses. Various members of this receptor family respond differently to the soluble and membrane-bound forms of their respective ligands. However, the determining factors and underlying molecular mechanisms of this diversity are not yet understood. Using an established system of chimeric TNFRs and novel ligand variants mimicking the bioactivity of membrane-bound TNF (mTNF), we demonstrate that the membrane-proximal extracellular stalk regions of TNFR1 and TNFR2 are crucial in controlling responsiveness to soluble TNF (sTNF). We show that the stalk region of TNFR2, in contrast to the corresponding part of TNFR1, efficiently inhibits both the receptor's enrichment/clustering in particular cell membrane regions and ligand-independent homotypic receptor preassembly, thereby preventing sTNF-induced, but not mTNF-induced, signaling. Thus, the stalk regions of the two TNFRs not only have implications for additional TNFR family members, but also provide potential targets for therapeutic intervention.

Fig 1

(A) Schematic representation of wild-type TNFR and TNFR-Fas chimeras. The chimeric receptors TNFR1-Fas and TNFR2-Fas consist of the extracellular and transmembrane domains of TNFR1 (aa 1 to 236) and TNFR2 (aa 1 to 301), respectively, and the intracellular domain of Fas (aa 191 to 335). The characteristic four cysteine-rich domains (CRD1 to -4), the stalk region (S), the transmembrane region (T), and the amino acids in the fused regions are indicated. In the TNFR1-S2T2-Fas and TNFR2-S1T1-Fas chimeras, aa 197 to 236 of TNFR1 and aa 202 to 301 of TNFR2 have been exchanged between the chimeric receptors. TNFR1-S2_Δ42T2-Fas and TNFR2-S2_Δ42T2-Fas represent chimeras in which aa 202 to 243 of the stalk region of TNFR2 have been deleted. (B) Amino acid sequences of the TNFR1 stalk region (aa 197 to 211), the TNFR2 stalk region (aa 202 to 257), and the stalk region in TNFR2-S_Δ42T2-Fas after deletion of 42 aa. Underlining indicates additional amino acids introduced through the applied cloning strategy (see the supplemental material).

Fig 2

The TNFR2 stalk region inhibits TNFR-Fas responsiveness to sTNF. MF generated from tnfr1^−/− tnfr2^−/− double-knockout mice were stably transfected with TNFR1-Fas (A), TNFR2-Fas (B), TNFR1-S2T2-Fas (C), TNFR2-S1T1-Fas (D), TNFR1-S2_Δ42T2-Fas (E), and TNFR2-S2_Δ42T2-Fas (F) expression constructs. Cells remained untreated or were treated with sTNF and CysTNF (0.015 to 100 ng/ml each). Cell viability was assessed by crystal violet staining. The means and standard deviations of three replicates are shown. The data are representative of more than four independent experiments.

Fig 3

The stalk region prevents sTNF-mediated signaling complex formation of wild-type TNFR2. HeLa cells stably expressing TNFR2, TNFR2-S1T1-R2, and TNFR2-S2_Δ42T2-R2 remained untreated (−) or were treated with sTNF or CysTNF (10 ng/ml each) for 5, 10, and 15 min at 37°C. TNFR signaling complexes were immunoprecipitated and analyzed for TRAF2. cIgG, goat anti-TNFR2 (2 μg). The data shown are representative of at least three independent experiments. (A) TRAF2 Western blot analysis. (B) Quantification of TRAF2 Western blots. TRAF2 levels are expressed as percentages of relative recruitment after 5 min of CysTNF stimulation. The values have been corrected for protein loading as determined from β-actin levels in supernatants after immunoprecipitation. Standard deviations determined from at least three independent experiments are indicated. Statistical differences between samples were assessed by the nonparametric Mann-Whitney U test. *, P ≤ 0.05.

Fig 4

The stalk region determines sTNF-mediated downstream signaling of wild-type TNFR2. MF stably expressing TNFR2 (A to C and J), TNFR2-S1T1-R2 (D to F and J), and TNFR2-S2_Δ42T2-R2 (G to I and J) were left untreated (−) or treated with 100 ng/ml of sTNF or CysTNF. Cells were stained for endogenous p65 and analyzed by fluorescence microscopy. (A to I) Representative images of three independent experiments. (J) Quantification of p65 nucleus-positive cells. The percentages represent proportions of p65 nucleus-positive cells. The standard deviations were calculated from three independent experiments; more than 200 cells/experiment were assessed. Relevant pairs were assessed for their significance by unpaired Student t tests (n.s., not significant; *, P ≤ 0.05; **, P ≤ 0.01).

Fig 5

A Gly-Ser stalk region renders TNFR2 responsive to sTNF. (A) Amino acid sequences of the stalk regions of wild-type (wt) TNFR2 and various TNFR2-Fas chimeras in which overlapping amino acid sequences had been replaced by Gly-Ser linkers (Ex1, aa 202 to 219; Ex2, aa 215 to 232; Ex3, aa 228 to 249; Ex4, aa 241 to 257). Linker sequences are underlined. (B to D) MF stably expressing the TNFR2-Fas stalk variants (Ex1, Ex2, Ex3, and Ex4) or TNFR-Fas chimeras in which the complete stalk regions had been replaced by a 56-aa Gly-Ser linker (C and D) were left untreated or treated with sTNF and CysTNF (0.015 to 100 ng/ml). The data shown are representative of at least three independent experiments. (B) Cell viability at 100 ng/ml is expressed as a percentage relative to the viability of untreated cells. The error bars indicate standard deviations from three independent experiments.

Fig 6

TNF dissociation kinetics at 37°C. MF stably expressing TNFR1-Fas (A), TNFR2-Fas (B), TNFR1-S2T2-Fas (C), and TNFR2-S1T1-Fas (D) were incubated with ¹²⁵I-labeled TNF at 4°C. Dissociation of ¹²⁵I-labeled TNF was measured at 37°C in the presence of unlabeled sTNF. Nonspecific binding was determined in the presence of a 200-fold excess of unlabeled sTNF (less than 5% of the total binding) and was subtracted. TNFR complex half-lives were calculated from the one-phase exponential decay curves. The data shown are representative of at least three independent experiments.

Fig 7

The TNFR2 stalk region prevents homotypic ligand-independent receptor interactions. MF stably expressing TNFR2-Fas, TNFR2-S1T1-Fas, and TNFR2-S2_Δ42T2-Fas were analyzed by flow cytometry (A and C) (anti-TNFR2, black line; secondary antibodies only, gray shaded; percentages, cells gated positive for the chimeras), and homotypic receptor interactions were assessed in chemical cross-linking studies (B and D). (B and D) Cells were incubated with or without the membrane-impermeable chemical cross-linker BS³ (33 μM to 500 μM). The two images in each panel refer to a single gel each. In order to obtain identical sample sequences, the gel image in panel D was rearranged. Monomeric and dimeric receptor species are indicated. The data shown are representative of three independent experiments.

Fig 8

Higher-order receptor cluster formation of wild-type TNFR2 is prevented by the stalk region. HEK 293 Flp-In T-Rex cells inducibly expressing TNFR2 (A, D, and G), TNFR2-S1T1-R2 (B, E, and H), and TNFR2-S2_Δ42T2-R2 (C, F, and I) were left untreated or induced with doxycycline. The data shown are representative of three independent experiments. (A to C) Receptor chimera cell surface expression as analyzed by flow cytometry (anti-TNFR2, black lines; secondary antibodies only, gray shaded; percentages, cells gated positive for the chimeras). (D to I) Higher-order receptor cluster formation. Cells were stained with Alexa Fluor 546-labeled sTNF and analyzed by confocal microscopy. Optical layers from the vertical z axis were stacked. The layer images closest to the glass surface of representative cells are shown. The boxes indicate areas that are shown at ×8.4 magnification in panels G, H, and I. Higher-order receptor clusters are indicated by arrows.

Fig 9

Higher-order receptor cluster formation and activation of TNFR-Fas chimeras are controlled by the stalk region. HEK 293 Flp-In T-Rex cells inducibly expressing TNFR1-Fas (A, E, and I), TNFR2-Fas (B, F, and J), TNFR2-S1T1-Fas (C, G, and K), and TNFR2-S2_Δ42T2-Fas chimeras (D, H, and L) were induced with doxycycline. The data shown are representative of three independent experiments. (A to D) Analyses of the formation of higher-order receptor clusters. Cells grown in the presence of zVAD-fmk were stained on ice using Alexa Fluor 546-labeled sTNF. After fixation, the cells were analyzed by confocal microscopy. (E to H) Analyses of cell surface expression of TNFR-Fas chimeras by flow cytometry. Cells grown in the presence of zVAD-fmk were stained using TNFR1-specific antibodies (black line in panel E) or anti-TNFR2-specific antibodies (black lines in panels F to H). Also shown are unstained cells (gray shaded) and noninduced anti-TNFR-stained cells (dashed lines). (I to L) Analyses of spontaneous caspase 3 activation upon TNFR-Fas chimera overexpression. Cells were stained with anti-active caspase 3 antibodies (black lines) and analyzed by flow cytometry. Also shown are noninduced anti-active caspase 3-stained cells (dashed lines) and induced unstained cells (gray shaded).

Fig 10

Quantitative analysis of TNFR clusters and distribution of TNFR cluster sizes. Confocal images obtained from the experiments shown in Fig. 8 and 9 (HEK 293 Flp-In T-Rex cells inducibly expressing TNFR2, TNFR2-S1T1-R2, and TNFR2-S2_Δ42T2-R2 [A] and TNFR2-Fas, TNFR2-S1T1-Fas, and TNFR2-S2_Δ42T2-Fas [B] chimeras) were quantified for cluster size distribution (0.04 to 9.72 μm²) per cell. The data are representative of 36 to 57 cells (A) and 25 to 57 cells (B).