We have identified a residue necessary for the cleavage of human p75 TNF-R from the cell surface by deletion and mutagenesis analysis of the membrane-proximal domain between amino acids 147 and 221. Deletion analysis of this area showed that residues between amino acids 207 and 216 are required for shedding. Site-directed mutagenesis of proline 211 to glycine reduced PMA-induced shedding of human p75 TNF-R from COS-7 or Raw 264.7 cells. Mutation of glycine 210 to aspartic acid did not affect receptor shedding. Mutation of serine 212 to leucine did not affect the PMA-induced shedding from the surface of COS-7 cells, but reduced the efficiency of shedding from the surface of Raw 264.7 macrophages by PMA or LPS. Deletion of putative elastase cleavage sites at amino acids 151 to 153, 162 to 163, and 177 to 178 (also a putative metalloprotease site similar to the cleavage site of TNF-α) or mutation of a serine-threonine-serine motif resembling a similar motif at the site of L-selectin cleavage at amino acids 202 to 204 did not reduce shedding of p75 TNF-R after treatment of cells with PMA. This work shows that a single amino acid mutation at proline 211 of human p75 TNF-R can prevent shedding from the cell surface, and that deletion of other previously proposed putative cleavage sites of the human p75 TNF-R does not prevent its shedding.

Proteolysis of membrane-bound proteins has often been regarded simply as a means to remove and degrade them as part of their normal turnover. However, many membrane-bound proteins resident on a wide variety of cell types have been shown to be specifically released in a directed and regulated manner to produce their soluble forms, including cytokines, cytokine receptors, cell adhesion molecules, ectoenzymes, and other proteins important in normal development and disease (for reviews, see Refs. 1–3).

TNF-α has previously been described as a pleiotropic cytokine produced mainly by macrophages (4). Two receptors able to bind both TNF-α and lymphotoxin (TNF-β) have been isolated and molecularly cloned. They have molecular masses of 55 kDa (p55 TNF-R) (5, 6, 7) and 75 kDa (p75 TNF-R) (8, 9) and share significant homologies in their extracellular domains, which contain cysteine-rich repeat sequences, defining them as members of the nerve growth factor/TNF-R family (8). The TNF-Rs are coexpressed by most cell lines and primary tissues, and they have been found in soluble form both in body fluids of patients with various conditions (10, 11) and in supernatants of cells in culture (12, 13, 14). Soluble TNF-Rs are able to bind TNF-α in a reversible manner; they neutralize TNF-α and restrict its availability to cell surface TNF-Rs (15, 16). Both receptors have been shown to be proteolytically cleaved from the cell surface, and no alternative splice sites have been identified in the mRNA, as have been identified for other cytokine receptors with soluble counterparts, e.g., IL-4 (17) and IL-7 (18) receptors.

TNF-α is synthesized as a membrane-bound molecule of 26 kDa that is cleaved from the cell surface, yielding the soluble mature 17-kDa protein (19). The cleavage sites of both TNF-α and p55 TNF-R have been identified by sequencing the soluble proteins and by mutagenesis of the proposed cleavage sites (16, 20, 21, 22). However, a cleavage site has not been identified for the p75 TNF-R, although several have been suggested. For example, there are several potential elastase cleavage sites (23), and a potential site similar to the cleavage site of TNF-α has been put forward (24). Mutational analysis of p55 TNF-R has shown that only a limited sequence is required for cleavage from the cell surface and that the cytoplasmic portion does not influence shedding (16), whereas the cytoplasmic domain of p75 TNF-R does modulate its cleavage. A phosphorylation event is required before p75 TNF-R is cleaved from the cell surface (25), and it has recently been shown that the 26-kDa TNF-α molecule is also phosphorylated (26), although it is not known whether this phosphorylation is necessary for cleavage or if it is required for certain signaling mechanisms of membrane-bound TNF-α.

In this study we have identified by deletion analysis and site-directed mutagenesis a single amino acid that is required for cleavage of the human p75 TNF-R from the surface of COS-7 cells and Raw 264.7 macrophages. We show that deletion of previously described potential cleavage sites for TNF-R such as the three potential elastase cleavage sites (23) at amino acids 151 to 153, 162 to 163, and 177 to 178 (also a putative metalloprotease site similar to the cleavage site of TNF-α) (16) or mutation of a serine threonine serine motif resembling a similar motif at the L-selectin cleavage site (27) at amino acids 202 to 204 did not prevent shedding of p75 TNF-R either spontaneously or after treatment of cells with PMA. A Clostridial collagenase cleavage site (28) at amino acids 209 to 210 was also shown not to be involved in the cleavage of p75 TNF-R from the cell surface by site-directed mutagenesis of the amino acid at the potential p1′ site of the sissile bond.

All chemicals were purchased from BDH (Poole, U.K.), Fisons (Loughborough, U.K.), or Sigma Chemcial Co. (Poole, U.K.). Cell culture reagents, G418, and FCS were purchased from Life Technologies, Inc. (Paisley, Scotland). The mAb UTR-1 was a gift from Hoffmann-La Roche. The components of the human soluble p75 TNF-R ELISA were provided by Dr. W. Buurman (Maastricht, The Netherlands) (29). Goat anti-mouse biotin conjugate and streptavidin-phycoerythrin were purchased from Southern Biotechnology, Inc. (Birmingham, AL). Goat anti-mouse horseradish peroxidase conjugate, enhanced chemiluminescence reagents, and Sequenase version 2.0 were purchased from Amersham International (Aylesbury, U.K.) Recombinant human TNF-α was a gift from Zeman et al. (30) and from Genentech, Inc. (San Francisco, CA). Restriction enzymes were purchased from Boehringer Mannheim (Lewes, U.K.), and DNA-modifying enzymes were obtained from New England Biolabs (Hitchin, U.K.). An initial stock of M13KO7 helper phage was purchased from Pharmacia (St. Albans, U.K.). Initial stocks of Escherichia coli CJ236 (dut-, ung-, thi-, relA-; pJC105 (Cm′)) were purchased from Bio-Rad (Hemel Hempstead, U.K.).

The p75 receptor mutants were constructed by site-directed mutagenesis using single-stranded uracil containing DNA as the template for the reactions. The ssDNA was prepared as follows. The p75 TNF-R cDNA expressed from a CMV promotor in a plasmid carrying the f1 phage origin of replication and an ampicillin resistance gene was transformed into E. coli CJ236. Single colonies were picked and grown with agitation for 5 h at 37°C in 2× YT broth (16 g/l bactotryptone, 5 g/l yeast extract, 10 g/l NaCl, and 25 mM Tris-HCl, pH 7.5) supplemented with 100 μg/ml ampicillin, 15 mg/ml chloramphenicol, and 25 mg/ml uridine. The cultures were then infected with M13KO7 at an multiplicity of infection of 1 × 1011/ml and allowed to stand for 30 min at 37°C followed by incubation at 37°C with agitation for 1 h. Kanamycin (70 μg/ml) was added to the cultures, and they were grown for an additional 5 h before being expanded sixfold in 2× YT with 100 μg/ml ampicillin, 15 μg/ml chloramphenicol, 70 μg/ml kanamycin, and 25 μg/ml uridine. The ssDNA was collected, and site-directed mutagenesis was performed as described by Kunkel (31). Positive clones were identified by sequencing with Sequenase version 2.0.

The SV40-transformed African Green Monkey cell line, COS-7 (American Type Culture Collection, Rockville, MD; CRL 1651), was maintained in DMEM supplemented with 10% FCS, penicillin (100 U/ml)/streptomycin (100 μg/ml), and 2 mM glutamine. Twenty-four hours before transfection cells were plated at 2.5 × 105 on 60-mm dishes (for ELISA and ligand binding) or at 5 × 105 on 90-mm dishes (for Western blotting). Ten or twenty micrograms of plasmid DNA (for 60- or 90-mm dishes, respectively) was used to transfect cells by the calcium phosphate coprecipitation method as described previously (32). To normalize results for the efficiency of transfection of each plate, cells were cotransfected with 2 μg of pSV2L (33), a luciferase expression plasmid. Cells were allowed to recover at 37°C for 48 h in DMEM with 10% FCS before harvesting.

The murine macrophage-like cell line, Raw 264.7 (American Type Culture Collection, TIB71), was maintained in DMEM supplemented with 10% FCS, penicillin (100 U/ml)/streptomycin (100 μg/ml), and 2 mM glutamine. Twenty-four hours before transfection, cells were plated at 5 × 105 on 90-mm dishes. Twenty micrograms of plasmid DNA together with 2 μg pSV2Neo, containing a neomycin resistance gene, were used to transfect cells by a modification of the calcium phosphate coprecipitation method as described above. Chloroquine (25 μM) was added to cells before addition of the DNA precipitate to inhibit lysosomal function. Cells were maintained with chloroquine for 48 h until selection in medium containing between 100 and 500 μg/ml G418. The medium and selection drug were replaced twice a week until clones of resistant cells were visible. Cell clones were either pooled and maintained as a population followed by cloning by limiting dilution, or they were individually removed from dishes using cloning rings. Positive clones were identified by FACS analysis using a FACScan (Becton Dickinson, Mountain View, CA) after surface labeling with UTR-1 followed by goat anti-mouse IgG-biotin and streptavidin-phycoerythrin.

Cells were collected by scraping and were lysed with 10 mM Tris-HCl (pH 7.4), 10 mM NaCl, 1 mM MgCl2, and 1% Nonidet P-40, and the nuclear debris was removed by centrifuging at 2000 rpm for 5 min. The cell lysates were mixed with an equal volume of prewarmed SDS sample buffer (100 mM Tris-HCl (pH 6.8), 4% SDS, 20% glycerol, and 0.001% bromophenol blue) without β-ME and boiled for 5 min. After being resolved on a 9% SDS-polyacrylamide gel, proteins were electrophoretically transferred to a nitrocellulose membrane, and the following steps were performed at room temperature. Membranes were blocked for 2 h in 20 mM Tris-HCl (pH 7.5), 200 mM NaCl, 0.1% Tween-20, and 5% nonfat dry milk powder followed by incubation with 1 μg/ml UTR-1 diluted in blocking solution for 1 h. After washing three times with 20 mM Tris-HCl (pH 7.5), 500 mM NaCl, and 0.1% Tween-20, membranes were incubated for 30 min with 20 ml of the original blocking solution together with sheep anti-mouse IgG- horseradish peroxidase conjugate, diluted 1/1000. After washing three times in PBS/0.1% Tween-20, detection was conducted using ECL reagents. Membranes were wrapped in Saran wrap and exposed to x-ray film from 30 s to 30 min.

Recombinant human TNF-α (from Genentech) was radiolabeled with 125I by the chloramine-T method as previously described (34). Cells were washed twice with cold PBS/1% BSA and incubated with 4 nM 125I-labeled TNF-α in 1 ml of DMEM/2% BSA for 1 h at 4°C with agitation, with or without addition of a 25-fold excess of nonradiolabeled recombinant human TNF-α. Cells were washed three times with cold PBS/1% BSA and then incubated in 1 ml of solubilization buffer (1% Triton X-100, 10% glycerol, and 25 mM HEPES, pH 7.4) for 30 min at 37°C. Supernatants from lysed cells were counted in an LKB 1282 Compugamma gamma counter (LKB, Rockville, MD) for 60 s.

Twenty-four hours after osmotic shock, the medium was removed from transfected COS-7 cells and replaced with 1 ml of DMEM without serum. Cells were left in medium alone and incubated for 18 h with 50 ng/ml recombinant human TNF-α (30) or incubated for 3 h the following morning with 10 ng/ml PMA. The supernatants were collected, and the cells were washed twice with PBS. Cells were lysed for 10 min at room temperature in 300 ml of lysis buffer (25 mM Tris-phosphate (pH 7.8), 2 mM DTT, 2 mM 1,2-diaminocyclohexane-N,N,N′,N′-tetraacetic acid, 10% glycerol, and 1% Triton X-100) with occasional gentle rocking and then removed from petri dishes using a sterile cell scraper. The lysed cells were centrifuged at 13,000 rpm for 1 min to pellet the cell debris. Supernatants and cell extracts were diluted in PBS containing 1% BSA and quantitated by human p75 TNF-R ELISA as described previously (29). Cell extracts were also assayed for luciferase activity using a Labsystems Luminoskan luminometer. One hundred milliliters of luciferase assay reagent (20 mM tricine (pH 7.8), 1.07 mM (MgCO3)4 Mg(OH)2·5H2O, 6.67 mM MgSO4, 0.1 mM EDTA, 33.3 mM DTT, 270 μM coenzyme A (lithium salt), 470 μM luciferin, and 530 μM ATP) was injected onto 20 ml of sample (held in Labsystems white opaque 96-well plates) by the luminometer. After a 2-s delay, an integral light measurement was made for 10 s.

Permanently transfected Raw 264.7 cells were seeded at 1 × 106 cells/60-mm dish. The following day the medium was replaced with 1 ml of DMEM containing 10% FCS. Cells were incubated in medium alone, with 10 ng/ml LPS (Sigma), or with 10 ng/ml PMA for 3 h at 37°C. Supernatants were quantitated by human p75 TNF-R ELISA as described above.

Levels of soluble and membrane-bound human p75 TNF-R were determined in supernatants and cell extracts of treated and untreated transfected cells by ELISA and then normalized for the efficiency of transfection (in the case of transiently transfected COS-7 cells) by luciferase assay of cell extracts according to the equation: [soluble membrane-bound p75 TNF-R (pg/ml)]/luciferase activity. The fold increase in shedding above similarly transfected untreated controls was then calculated for cells treated with TNF, PMA, or LPS, thus normalizing any differences in overall expression of the different mutants in the two cell types transfected. The values from at least four independent experiments were used to calculate statistical differences. Values were plotted ±1 SD.

Statistical analysis was performed using Minitab Release 8, Macintosh version (Minitab, State College, PA). The nonparametric Mann-Whitney test was used for all statistical data shown.

To determine whether the cell types used in this study were able to cleave wild-type human p75 TNF-R, the shed forms of transfected human p75 TNF-R from both monkey COS-7 and mouse Raw 264.7 cells were compared with soluble p75 TNF-R shed from the surface of human Jijoye lymphoma cells after treatment with PMA, by Western blotting and immunodetection (Fig. 1). Human p75 TNF-R shed from the surface of either COS-7 cells or Raw 264.7 cells had a similar mobility after SDS-PAGE as human p75 TNF-R shed from the surface of Jijoye cells. This suggests that the mechanism of cleavage of membrane-bound p75 TNF-R in all three cell types may be the same and that major differences in post-translational modifications, such as glycosylation, are negligible.

FIGURE 1.

Western blotting of soluble human p75 TNF-R shed from the surface of different cell types. COS-7 cells, transiently transfected with human p75 TNF-R, stably transfected Raw 264.7 cells, or Jijoyes were stimulated with 10 ng/ml PMA for 3 h at 37°C in serum-free medium. The supernatants were separated by nonreducing SDS-PAGE followed by electroblotting and immunodetection with UTR-1.

FIGURE 1.

Western blotting of soluble human p75 TNF-R shed from the surface of different cell types. COS-7 cells, transiently transfected with human p75 TNF-R, stably transfected Raw 264.7 cells, or Jijoyes were stimulated with 10 ng/ml PMA for 3 h at 37°C in serum-free medium. The supernatants were separated by nonreducing SDS-PAGE followed by electroblotting and immunodetection with UTR-1.

Close modal

Four deletions were made, each of 25 amino acids, covering the region of the extracellular domain of p75 TNF-R from the end of the fourth cysteine-rich repeat (amino acid 147) to the start of the transmembrane domain (amino acid 221). These deletion mutants were named Δ1 (removal of amino acids 147–171), Δ2 (removal of amino acids 172–196), Δ3 (removal of amino acids 197–221), and Δ123 (removal of amino acids 147–221; see Tables I and II). Ligand binding analysis of the mutants revealed that all four were expressed on the surface of COS-7 cells and were able to bind 125I-labeled recombinant human TNF-α, although binding of Δ1 and Δ123 was reduced (Table III). This can be explained by the fact that in both mutants part of the fourth cysteine-rich repeat has been deleted, which has previously been shown to be required for TNF-α binding to p75 TNF-R (35). Western blot analysis of extracts from COS-7 cells transfected with wild-type or mutant p75 TNF-R cDNA showed that all were expressed and were recognized by a human p75 TNF-R-specific Ab (Fig. 2 A). Again Δ123 was expressed to a much lesser extent than the wild-type p75 TNF-R and the other deletion mutants. It is possible that the large deletion of 75 amino acids had affected its recognition by the Ab UTR-1. Δ1, Δ2, and Δ3 have a greater mobility than the wild-type p75 TNF-R, as would be expected after deletion of 25 amino acids from each molecule. The mobility of Δ123, however, is similar to that of the wild-type p75 TNF-R, and this may be due to some unfolding of the molecule due to the large deletion within it.

Table III.

Ligand-binding analysis of wild-type p75 TNF-R and mutantsa

MutantBinding of 125I TNF-α (cpm)
WT p75 TNF-R 112,919 
Δ1 25,948 
Δ2 43,124 
Δ3 70,749 
Δ123 7,742 
Δ3b 127,405 
Δ3c 8,918 
Δ3d 10,609 
SRS 6,787 
PMDP 52,158 
GGS 58,388 
GPL 23,434 
Mock 600 
MutantBinding of 125I TNF-α (cpm)
WT p75 TNF-R 112,919 
Δ1 25,948 
Δ2 43,124 
Δ3 70,749 
Δ123 7,742 
Δ3b 127,405 
Δ3c 8,918 
Δ3d 10,609 
SRS 6,787 
PMDP 52,158 
GGS 58,388 
GPL 23,434 
Mock 600 
a

Transfected Cos 7 cells were incubated with 4 nM 125I TNF with or without the addition of 1 mM unlabeled recombinant human TNF. Binding of 125I TNF to p75 TNF-R was assessed by counting solubilized samples on a gamma counter.

FIGURE 2.

Western blotting of wild-type and mutant p75 TNF-R. Cell lysates from COS-7 cells were separated by SDS-PAGE, transferred to nitrocellulose membrane, and incubated with UTR-1 followed by detection with ECL. A, Cells were transfected with wild-type p75 TNF-R, Δ1, Δ2, Δ3, or Δ123 cDNAs. B, Cells were transfected with wild-type p75 TNF-R, Δ3b, Δ3c, or Δ3d cDNAs. C, Cells were transfected with wild-type p75 TNF-R, SRS, or PMDP cDNAs. D, Cells were transfected with wild-type p75 TNF-R, GGS, or GPL cDNAs.

FIGURE 2.

Western blotting of wild-type and mutant p75 TNF-R. Cell lysates from COS-7 cells were separated by SDS-PAGE, transferred to nitrocellulose membrane, and incubated with UTR-1 followed by detection with ECL. A, Cells were transfected with wild-type p75 TNF-R, Δ1, Δ2, Δ3, or Δ123 cDNAs. B, Cells were transfected with wild-type p75 TNF-R, Δ3b, Δ3c, or Δ3d cDNAs. C, Cells were transfected with wild-type p75 TNF-R, SRS, or PMDP cDNAs. D, Cells were transfected with wild-type p75 TNF-R, GGS, or GPL cDNAs.

Close modal

The spontaneous and induced shedding of the mutants from the surface of COS-7 cells was assessed. Supernatants and cell extracts were assayed for the presence of human p75 TNF-R by ELISA (29), and the fold increase in shedding compared to that in untreated, similarly transfected cells was calculated (Fig. 3 A). Shedding of wild-type p75 TNF-R was increased 1.7-fold above that in untreated transfectants after treatment of cells with TNF and 4.4-fold after treatment with PMA, while shedding of Δ1 and Δ2 was increased 1.7- and 2.9-fold, respectively, after treatment with TNF. A 3.3-fold increase in shedding of Δ1 and a 4.2-fold increase in shedding of Δ2 were observed after treatment of cells with PMA. In contrast, there was no increase in the secretion of Δ3 or Δ123 above that in untreated transfectants after treatment of cells with TNF-α, and there was a modest increase in shedding above that in untreated controls of 1.7-fold for Δ3 and 1.9-fold for Δ123 after treatment with PMA. This represents a significant reduction in induction of shedding compared with the increase in shedding above that in similarly transfected unstimulated controls observed after PMA treatment of COS-7 cells transfected with wild-type p75 TNF-R (p = 0.0373 for Δ3 and p = 0.0216 for Δ123).

FIGURE 3.

Fold increase in shedding of wild-type p75 TNF-R and deletion mutants from the surface of COS-7 cells. Transfected COS-7 cells were incubated with 50 ng/ml recombinant human TNF-α for 18 h or with 10 ng/ml PMA for 3 h, or were left in medium alone. Cell extracts and supernatants were collected and analyzed by ELISA for the presence of human p75 TNF-R. The fold increase in shedding above that in untreated transfectants was calculated as described in Materials and Methods. Values were plotted ±1 SD. A, COS-7 cells transfected with wild-type p75 TNF-R, Δ1, Δ2, Δ3, or Δ123 deletion mutants (* indicates p = 0.0373; ** indicates p = 0.0216). B, COS-7 cells transfected with wild-type p75 TNF-R, Δ3b, Δ3c, or Δ3d deletion mutants (* indicates p = 0.0454; ** indicates p = 0.0081). No human p75 TNF-R was detected from the supernatants or cell extracts of mock-transfected cells before or after incubation with recombinant human TNF-α and PMA (data not shown).

FIGURE 3.

Fold increase in shedding of wild-type p75 TNF-R and deletion mutants from the surface of COS-7 cells. Transfected COS-7 cells were incubated with 50 ng/ml recombinant human TNF-α for 18 h or with 10 ng/ml PMA for 3 h, or were left in medium alone. Cell extracts and supernatants were collected and analyzed by ELISA for the presence of human p75 TNF-R. The fold increase in shedding above that in untreated transfectants was calculated as described in Materials and Methods. Values were plotted ±1 SD. A, COS-7 cells transfected with wild-type p75 TNF-R, Δ1, Δ2, Δ3, or Δ123 deletion mutants (* indicates p = 0.0373; ** indicates p = 0.0216). B, COS-7 cells transfected with wild-type p75 TNF-R, Δ3b, Δ3c, or Δ3d deletion mutants (* indicates p = 0.0454; ** indicates p = 0.0081). No human p75 TNF-R was detected from the supernatants or cell extracts of mock-transfected cells before or after incubation with recombinant human TNF-α and PMA (data not shown).

Close modal

The gross deletions suggested that the region spanned by the Δ3 deletion was essential for cleavage of p75 TNF-R. Therefore, a second set of deletion mutants was made, each five amino acids in length, spanning the region deleted by Δ3, named Δ3b, -c, and -d (see Tables I and II). The three deletion mutants were expressed in COS-7 cells and were shown by Western blot analysis to be expressed and recognized by the p75 TNF-R-specific Ab (Fig. 2,B). Ligand binding analysis showed that all three mutants were expressed on the cell surface (Table III). The induced and the spontaneous shedding of the deletion mutants were assessed (Fig. 3 B). Cleavage of wild-type p75 TNF-R increased 5.8-fold above basal levels after treatment of cells with PMA. Shedding of Δ3b increased 3.4-fold compared with that in untreated transfectants, while shedding of Δ3c and Δ3d each only increased 1.3-fold, a significant reduction in PMA-induced shedding compared with that of wild-type p75 TNF-R (p = 0.0454 for Δ3c; p = 0.0081 for Δ3d), indicating that a motif within these 10 amino acids was required for cleavage.

Several point mutations of the human p75 TNF-R were made within the region deleted by the Δ3 construct (Tables I and II). A motif similar to that found in the p1′ p2′ p3′ positions after the sissile bond in the L-selectin cleavage site (27) was mutated from serine-threonine-serine (STS)4 to serine-arginine-serine (SRS) at amino acids 202 to 204 (in the region deleted by Δ3b). A second mutation was made in a potential Clostridial collagenase site (28) (proline-methionine-glycine-proline, where the sissile bond is between methionine and glycine) mutating glycine 210 to aspartic acid (PMDP mutation). Expression of the mutant receptors was assessed by ligand binding (Table III) and Western blotting (Fig. 2,C), which showed that these mutants were expressed on the surface of COS-7 cells. Shedding of wild-type p75 TNF-R increased 8.5-fold after treatment of cells with PMA compared with similarly transfected untreated control cells (Fig. 4 A), while shedding of SRS increased 3.6-fold and shedding of PMDP increased 4.3-fold above that in untreated transfectants. These findings indicated that the STS motif is not required for cleavage of p75 TNF-R from the cell surface and that the enzyme responsible does not have the same cleavage specificity as Clostridial collagenase.

FIGURE 4.

Fold increase in shedding of wild-type p75 TNF-R and point mutants. Transfected COS-7 cells were incubated with 50 ng/ml recombinant human TNF-α for 18 h or with 10 ng/ml PMA for 3 h, or left in medium alone. Cell extracts and supernatants were collected and analyzed by ELISA for the presence of human p75 TNF-R. The fold increase in shedding above that in untreated transfectants was calculated as described in Materials and Methods. Values were plotted ± 1 SD. A, COS-7 cells transfected with wild-type human p75 TNF-R, SRS, or PMDP. B, COS-7 cells transfected with wild-type human p75 TNF-R, GGS, or GPL point mutants (* indicates p = 0.0172).

FIGURE 4.

Fold increase in shedding of wild-type p75 TNF-R and point mutants. Transfected COS-7 cells were incubated with 50 ng/ml recombinant human TNF-α for 18 h or with 10 ng/ml PMA for 3 h, or left in medium alone. Cell extracts and supernatants were collected and analyzed by ELISA for the presence of human p75 TNF-R. The fold increase in shedding above that in untreated transfectants was calculated as described in Materials and Methods. Values were plotted ± 1 SD. A, COS-7 cells transfected with wild-type human p75 TNF-R, SRS, or PMDP. B, COS-7 cells transfected with wild-type human p75 TNF-R, GGS, or GPL point mutants (* indicates p = 0.0172).

Close modal

The two amino acids at the junction of the deletions of Δ3c and Δ3b were individually mutated. The serine at position 212 was changed to a leucine in one mutant (GPL), and the proline at position 211 was mutated to a serine in the other (GGS; Tables I and II). These mutants were both shown to be expressed on the surface of COS-7 cells by ligand binding analysis (Table III) and Western blotting (Fig. 2,D). Treatment of transfected cells with PMA increased the shedding of wild-type p75 TNF-R 3.7-fold, while shedding of GPL increased 3-fold (Fig. 4 B). However, shedding of GGS was only increased 1.2-fold above that in similarly transfected untreated controls in the presence of PMA, a significant reduction compared with the observed increase in shedding of wild-type p75 TNF-R (p = 0.0172). This indicated that proline 211 played an important role in the PMA-induced shedding of human p75 TNF-R from the cell surface, since its replacement with another amino acid prevented a PMA-induced increase in proteolytic cleavage.

The increase in shedding observed after treatment of transfected COS-7 cells with TNF was much less than that induced by PMA for all the p75 TNF-R constructs analyzed (Figs. 3 and 4). This is in contrast with the results of studies conducted using nontransformed cells such as neutrophils and mononuclear cells, where TNF has been shown to increase shedding of p75 TNF-R severalfold (36). It is possible that COS-7 cells do not express molecules necessary in the signaling pathway of human TNF.

To avoid the possible variation of transient transfection assays, murine Raw 264.7 macrophages were permanently transfected with the deletion mutants of human p75 TNF-R described above. Positive cell lines were identified by flow cytometry after staining with UTR-1 (data not shown), and shedding of p75 TNF-R mutants was assessed after stimulation with either LPS or PMA (Fig. 5 A). Similar to results obtained using COS-7 cells, deletion of amino acids 197 to 171 (Δ3 deletion) significantly reduced shedding (p = 0.0113) after treatment of cells with LPS (a 1.2-fold increase in shedding above that in untreated transfectants) compared with shedding of wild-type human p75 TNF-R from the surface of these cells (a 3-fold increase in shedding above that in untreated transfectants). Shedding of p75 TNF-R was not altered after deletion of amino acids 147 to 171 (Δ1 deletion) or amino acids 172 to 196 (Δ2 deletion). A 4-fold increase in shedding was observed after treatment of cells with LPS in the Δ1 deletion and a 3.4-fold increase in shedding was observed in the Δ2 deletion. PMA induced shedding of the Δ3 mutation was reduced (1.4-fold increase above that in unstimulated transfectants), but this did not reach statistical significance compared with the increase in shedding observed with wild-type p75 TNF-R (a 2-fold increase in shedding).

FIGURE 5.

Fold increase in shedding of wild-type p75 TNF-R and deletion mutants from the surface of Raw 264.7 macrophages. A, Raw 264.7 cells stably transfected with wild-type human p75 TNF-R, Δ1, Δ2, Δ3, Δ3b, Δ3c, or Δ3d were incubated with 10 ng/ml LPS or with 10 ng/ml PMA for 3 h, or were left in medium with serum. Levels of soluble human p75 TNF-R were quantitated by ELISA. The fold increase in soluble p75 TNF-R above that in similar transfectants treated with medium alone was then calculated for cells treated with LPS or PMA, and the values from at least four different experiments were used to plot graphs. Values were plotted ±1 SD (* indicates p = 0.0113; ** indicates p = 0.0428; § indicates p = 0.0177). No human p75 TNF-R was detected in the supernatants of untransfected Raw 264.7 cells before or after incubation with LPS or PMA (data not shown). B, Raw 264.7 cells stably transfected with wild-type human p75 TNF-R, SRS, or PMDP were treated as described in A. C, Raw 264.7 cells stably transfected with wild-type human p75 TNF-R, GGS, or GPL were treated as described in A (* indicates p = 0.0018; ** indicates p = 0.0043; a indicates p = 0.0008; aa indicates p = 0.0003).

FIGURE 5.

Fold increase in shedding of wild-type p75 TNF-R and deletion mutants from the surface of Raw 264.7 macrophages. A, Raw 264.7 cells stably transfected with wild-type human p75 TNF-R, Δ1, Δ2, Δ3, Δ3b, Δ3c, or Δ3d were incubated with 10 ng/ml LPS or with 10 ng/ml PMA for 3 h, or were left in medium with serum. Levels of soluble human p75 TNF-R were quantitated by ELISA. The fold increase in soluble p75 TNF-R above that in similar transfectants treated with medium alone was then calculated for cells treated with LPS or PMA, and the values from at least four different experiments were used to plot graphs. Values were plotted ±1 SD (* indicates p = 0.0113; ** indicates p = 0.0428; § indicates p = 0.0177). No human p75 TNF-R was detected in the supernatants of untransfected Raw 264.7 cells before or after incubation with LPS or PMA (data not shown). B, Raw 264.7 cells stably transfected with wild-type human p75 TNF-R, SRS, or PMDP were treated as described in A. C, Raw 264.7 cells stably transfected with wild-type human p75 TNF-R, GGS, or GPL were treated as described in A (* indicates p = 0.0018; ** indicates p = 0.0043; a indicates p = 0.0008; aa indicates p = 0.0003).

Close modal

Shedding of the five amino acid deletion mutants described above was also examined in permanently transfected Raw 264.7 cells (Fig. 5 A). Again, the results were similar to those observed in COS-7 cells. The increase in shedding above that in unstimulated transfectants of Δ3c and Δ3d was significantly reduced, compared with that in wild-type human p75 TNF-R, after treatment of cells with LPS. A 1.5-fold increase in shedding above that in untreated transfectants was observed for Δ3c (p = 0.0428 compared with LPS-induced shedding of wild-type human p75 TNF-R), and a 1.4-fold increase above that in untreated transfectants was observed for Δ3d (p = 0.0177 compared with LPS-induced shedding of wild-type human p75 TNF-R). However, shedding of Δ3b was not reduced after treatment of cells with either LPS or PMA compared with wild-type p75 TNF-R (a 5.4-fold increase in shedding above untreated transfectants after treatment with LPS and a 4.2-fold increase after treatment with PMA).

The point mutants of human p75 TNF-R described above were permanently transfected into Raw 264.7 cells, and positive clones were identified by flow cytometry after staining with mAb UTR-1 (data not shown). Shedding of point mutants was assessed in these cells after treatment with LPS or PMA (Fig. 5 B). Point mutations T203 to R (SRS) and G210 to D (PMDP) did not significantly reduce shedding of p75 TNF-R after treatment of Raw 264.7 cells with LPS or PMA (shedding of SRS increased 5.9- and 2.2-fold above that in untreated transfectants, while shedding of PMDP increased 4.7- and 2.7-fold after treatment with LPS or PMA, respectively).

Similar to the effect of the point mutation of P211 to G in COS-7 cells, shedding of GGS in Raw 264.7 cells was significantly reduced after treatment of cells with LPS or PMA (Fig. 5 C). A 0.85-fold (p = 0.0018) and a 0.79-fold (p = 0.0043) increase in shedding of GGS above similarly transfected untreated controls were observed after treatment of cells with LPS and PMA, respectively, compared with 1.8- and 3.6-fold increases in shedding of wild-type p75 TNF-R above that in untreated transfectants. The efficacy of shedding of p75 TNF-R was much reduced by the GPL point mutation in Raw 264.7 cells, in contrast to the lack of effect observed in COS-7 cells. Shedding of GPL was increased 1.2- and 1.3-fold above that in untreated transfectants after treatment of cells with LPS and PMA, respectively, a significantly reduced response compared with the increase in shedding of wild-type p75 TNF-R above that in untreated transfectants after LPS or PMA treatment (p = 0.0008 for LPS-induced shedding and p = 0.0003 for PMA-induced shedding). This suggested that while the presence of proline 211 is an obligatory requirement for PMA/LPS-induced shedding of human p75 TNF-R to occur from the surface of both COS-7 cells and Raw 264.7 macrophages, the presence of a serine at position 212 increased the efficiency of shedding after a physiologically relevant stimulus in Raw 264.7 cells, but was not necessary for PMA-induced shedding from the surface of COS-7 cells.

Membrane-proximal cleavage has recently emerged as a specific mechanism to rapidly modulate the cell surface expression of many different proteins, including the down-regulation of TNF-α and its two receptors, p55 TNF-R and p75 TNF-R; L-selectin; TGF-α; β-amyloid precursor protein; angiotensin-converting enzyme; neu oncogene; and steel growth factor from the cell surface (reviewed in 2 . The proteases involved in membrane-proximal cleavage have not yet been identified for any cellular system, and the mechanisms involved are slowly being elucidated by use of specific protease inhibitors and by site-directed mutagenesis of the membrane-bound targets of proteolysis.

The mechanisms involved in the cleavage of p55 TNF-R and of TNF-α itself have been investigated using both techniques, and there appears to be more than one enzyme involved in the cleavage of both from the cell surface, since both serine proteases (23, 37, 38) and metalloenzymes (39, 40, 41, 42) have been implicated by the use of specific inhibitors. The cleavage sites in both proteins have been determined, between alanine 76 and valine 77 for TNF-α (19) and between asparagine 172 and valine 173 for p55 TNF-R (22). It has previously been shown that a metalloprotease inhibitor of TNF-α cleavage is also able to prevent the shedding of human p75 TNF-R from the cell surface (24), suggesting a possible cleavage site in p75 TNF-R corresponding to the cleavage site of human TNF-α between alanine 177 and valine 178. Our present work indicates that this site is unlikely to be involved in the shedding of human p75 TNF-R from the cell surface, since deletion of amino acids 172 to 196 (Δ2) does not reduce PMA/LPS-induced shedding of p75 TNF-R. Another study implicated the involvement of a serine protease, more specifically elastase, in the shedding of p75 TNF-R by the use of purified fractions from neutrophils and by general serine protease and elastase inhibitors (23). Three potential elastase cleavage sites were suggested in the spacer region of p75 TNF-R at amino acids 151 to 153, 162 to 163, and 177 to 178. Since removal of amino acids 147 to 171 (Δ1) and 172 to 196 (Δ2) does not reduce the PMA/LPS-induced increase in shedding of p75 TNF-R it is unlikely that elastase is directly involved in its cleavage. It may be that elastase provides an alternative mechanism to down-modulate p75 TNF-R from the surface of neutrophils in certain conditions or is involved in the activation of other membrane-bound proteases.

This study has identified proline 211 to be crucial for the cleavage of p75 TNF-R from the cell surface to occur. Removal of large portions of the spacer region of p75 TNF-R (Δ1 and Δ2) resulted in no loss of shedding ability, indicating that there are no structural requirements for protease action in this area. Our data showed an absolute requirement for proline 211 to be present, since substitution with glycine prevented a PMA/LPS-induced increase in shedding from the cell surface in different cell types, while replacement of glycine 210 with aspartic acid had no such effect on receptor shedding. Replacement of serine 212 with leucine had no effect on the PMA-induced shedding of p75 TNF-R from the surface of COS-7 cells, although this shedding was significantly reduced from the surface of murine Raw 264.7 macrophages. This suggests that there may be some differing structural requirements of the protease responsible for the cleavage of p75 TNF-R from the surface of cells from different species. It is interesting to note that the murine p75 TNF-R also contains a proline residue, situated at a similar distance from the cell surface. Indeed, helical alignment of the two receptors from different species (Fig. 6) reveals that these proline residues are situated on the same face of an α-helix, although the residue N terminal to this proline is isoleucine in the mouse. It is therefore interesting that substitution of serine with leucine had such an effect on the PMA/LPS-induced shedding of human p75 TNF-R in murine cells. One possible explanation might be that the side chain of leucine is less hydrophobic than that of isoleucine, and its side chain has a different spatial conformation. In contrast, the side chain of serine is small and therefore may not interfere with binding of the murine enzyme.

FIGURE 6.

Helical alignment of human and murine p75 TNF-R around proline 211. Alignment was performed using the GeneWorks software package for the Macintosh. Closed circles indicate homologous residues. The arrow indicates the target residue, proline 211 (human p75 TNF-R). Boxed residues indicate the start of the transmembrane domain.

FIGURE 6.

Helical alignment of human and murine p75 TNF-R around proline 211. Alignment was performed using the GeneWorks software package for the Macintosh. Closed circles indicate homologous residues. The arrow indicates the target residue, proline 211 (human p75 TNF-R). Boxed residues indicate the start of the transmembrane domain.

Close modal

In addition there are further similarities between the human and mouse p75 TNF-R, both within the transmembrane domain and amino terminal to proline 211, where homologous residues face similar sides of the α-helix. A repeating proline serine motif is observed within this region of both human and murine p75 TNF-R, and this may affect the structure with respect to the cleavage of the receptor from the cell surface. If the hypothetical tertiary structure in this region is correct, it suggests that these are possible areas of contact between the receptor and the peptidase.

The spacer region of p75 TNF-R is proline rich. Since the removal of both amino acids 207 to 211 (Δ3c) and 212 to 216 (Δ3d) abrogated PMA/LPS-induced cleavage, it may be that the proline residues contained in this 10-amino acid sequence are arranged to provide a pocket or groove for interaction with the protease before cleavage of the membrane-bound molecule can occur. It is possible that upon stimulation of cells, the conformation of the receptor, the proteolytic enzyme, or both change before cleavage can occur. Alternatively, the proline located at position 211 may provide a sissile bond for the specific p75 TNF-R-cleaving enzyme. The role of proline is usually thought of as a structural one, providing the breaks in a helical structure (43), although several proteolytic enzymes have been identified that cleave at a post-prolyl bond, such as human thimet oligopeptidase (EC 3.4.24.15) (44), rat kidney endopeptidase (EC 3.4.24.16) (45), and prolyl endopeptidase (EC 3.4.21.26) (46).

It has previously been suggested that a sequence similarity between several membrane-bound molecules that are shed from the cell surface may provide the basis for some similarity in their mechanism of shedding (27). A related motif within the cleavage domains of p75 TNF-R, L-selectin, angiotensin-converting enzyme, and CD16II of serine-phenylalanine-serine (SFS; STS in p75 TNF-R) has been investigated in this study. The SFS motif in L-selectin occurs proximal to the cleavage site at the P1′, P2′, and P3′ positions after the sissile bond, which has been determined by sequencing of the soluble L-selectin molecule and by the part remaining bound to the membrane after cleavage. We have shown that mutation of the related STS motif to SRS had no appreciable effect on shedding of p75 TNF-R. A similar study conducted on the L-selectin molecule, where SFS was mutated to AAA, also had no appreciable effect on shedding (47). This implies that the mechanism of cleavage is different for these molecules despite sequence similarities in their cleavage domain.

The enzyme cascades involved in the shedding of p55 and p75 TNF-R may be related but are not the same, since the two receptors can be differentially induced to shed from the cell surface with different agents (36, 48). This has important implications, since previous work has shown that similar inhibitors (24, 37, 42, 49) can actually inhibit the cleavage of both receptors. Since there is no homology between p55 TNF-R and p75 TNF-R in their cleavage domains, it may be that the inhibitors used were not sufficiently specific and were able to inhibit similar enzymes that activate the specific receptor proteases. It is possible that both TNF-α and its receptors are cleaved by a similar, relatively sequence-specific family of enzymes. Indeed, a membrane-bound metalloprotease belonging to the adamalysin family of metzincin metalloproteases has recently been isolated, cloned, and shown to be responsible for the shedding of TNF (TNF-α-converting enzyme) (41, 50, 51, 52). There are a number of other members of this family of membrane-bound metalloenzymes for whom no specific substrates have been described (53).

With the use of more specific peptide inhibitors it will be possible to further characterize the enzyme responsible for the cleavage of p75 TNF-R from the cell surface.

Table I.

Sequences of oligodeoxynucleotide primers and positions of deleted or mutated amino acidsa

MutationAmino Acid(s)Primer Sequences
Δ1a 147–171 5′-ACTGCCCCTGGGGCCATACTGTGGGGCCTGCAAATATCCG-3′ 
Δ2 172–196 5′-GTGCTTGGAGCAGTGCTGGGCCGGGTGGGGGACGTGGACG-3′ 
Δ3 197–221 5′-AGTCCAACTGGAAGAGCGAATTCTGGAGTTGGCTGCGTGT-3′ 
Δ123 147–221 5′-AGTCCAACTGGAAGAGCGAAGTGGGGCCTGCAAATATCCG-3′ 
Δ3b 202–206 5′-GGGCTGGGGCCCATTGGGAGTGGAGCAGTGCTGGTTCTG-3′ 
Δ3c 207–211 5′-CTCCCTTCAGCTGGGGGGCTCAGGAAGGAGGTGCTTGGAG-3′ 
Δ3d 212–216 5′-GCGAAGTCGCCAGTGCTCCCGGGGCCCATTGGGAGCAGGA-3′ 
STS-SRSb 203 5′-GAGCAGGAAGGACCGGCTTGGAGCAGT-3′ 
PMGPS-PMDPS 210 5′-TGGGGGGCTGGGATCCATTGGGAGCAG-3′ 
PMGPS-PMGGS 211 5′-AGCTGGGGGGCTGCCGCCCATTGGGAG-3′ 
PMGPS-PMGPL 212 5′-TTCAGCTGGGGGGAGGGGGCCCATTGG-3′ 
MutationAmino Acid(s)Primer Sequences
Δ1a 147–171 5′-ACTGCCCCTGGGGCCATACTGTGGGGCCTGCAAATATCCG-3′ 
Δ2 172–196 5′-GTGCTTGGAGCAGTGCTGGGCCGGGTGGGGGACGTGGACG-3′ 
Δ3 197–221 5′-AGTCCAACTGGAAGAGCGAATTCTGGAGTTGGCTGCGTGT-3′ 
Δ123 147–221 5′-AGTCCAACTGGAAGAGCGAAGTGGGGCCTGCAAATATCCG-3′ 
Δ3b 202–206 5′-GGGCTGGGGCCCATTGGGAGTGGAGCAGTGCTGGTTCTG-3′ 
Δ3c 207–211 5′-CTCCCTTCAGCTGGGGGGCTCAGGAAGGAGGTGCTTGGAG-3′ 
Δ3d 212–216 5′-GCGAAGTCGCCAGTGCTCCCGGGGCCCATTGGGAGCAGGA-3′ 
STS-SRSb 203 5′-GAGCAGGAAGGACCGGCTTGGAGCAGT-3′ 
PMGPS-PMDPS 210 5′-TGGGGGGCTGGGATCCATTGGGAGCAG-3′ 
PMGPS-PMGGS 211 5′-AGCTGGGGGGCTGCCGCCCATTGGGAG-3′ 
PMGPS-PMGPL 212 5′-TTCAGCTGGGGGGAGGGGGCCCATTGG-3′ 
a

Primers for deletions were designed to anneal to 20 bases on either side of the area to be deleted from p75 TNF-R cDNA.

b

Primers for site-directed mutagenesis were designed to anneal to at least 12 bases on either side of the bases to be altered. Underlined nucleotides represent sites of introduced point mutations.

Table II.

Amino acids deleted or altered by mutagenesis of wild-type p75 TNF-R

 140a 147 172 197 202 207 212 217 222 
WTp75 TSSTDICRPHQICNVVAIPGNASMDAVCTSTSPTRSMAPGAVHLPQPVSTRSQHTQPTPEPSTAPSTSFLLPMGPSPPAEGSTGDFALPVGLIVG 
Δ1 TSSTDICRPH-------------------------SMAPGAVHLPQPVSTRSQHTQPTPEPSTAPSTSFLLPMGPSPPAEGSTGDFALPVGLIVG 
Δ2 TSSTDICRPHQICNVVAIPGNASMDAVCTSTSPTR-------------------------PSTAPSTSFLLPMGPSPPAEGSTGDFALPVGLIVG 
Δ3 TSSTDICRPHQICNVVAIPGNASMDAVCTSTSPTRSMAPGAVHLPQPVSTRSQHTQPTPE-------------------------FALPVGLIVG 
Δ123 TSSTDICRPH---------------------------------------------------------------------------FALPVGLIVG 
Δ3b TSSTDICRPHQICNVVAIPGNASMDAVCTSTSPTRSMAPGAVHLPQPVSTRSQHTQPTPEPSTAP-----LPMGPSPPAEGSTGDFALPVGLIVG 
Δ3c TSSTDICRPHQICNVVAIPGNASMDAVCTSTSPTRSMAPGAVHLPQPVSTRSQHTQPTPEPSTAPSTSFL-----SPPAEGSTGDFALPVGLIVG 
Δ3d TSSTDICRPHQICNVVAIPGNASMDAVCTSTSPTRSMAPGAVHLPQPVSTRSQHTQPTPEPSTAPSTSFLLPMGP-----GSTGDFALPVGLIVG 
SRS TSSTDICRPHQICNVVAIPGNASMDAVCTSTSPTRSMAPGAVHLPQPVSTRSQHTQPTPEPSTAPSRSFLLPMGPSPPAEGSTGDFALPVGLIVG 
PMDP TSSTDICRPHQICNVVAIPGNASMDAVCTSTSPTRSMAPGAVHLPQPVSTRSQHTQPTPEPSTAPSTSFLLPMDPSPPAEGSTGDFALPVGLIVG 
GGS TSSTDICRPHQICNVVAIPGNASMDAVCTSTSPTRSMAPGAVHLPQPVSTRSQHTQPTPEPSTAPSTSFLLPMGGSPPAEGSTGDFALPVGLIVG 
GPL TSSTDICRPHQICNVVAIPGNASMDAVCTSTSPTRSMAPGAVHLPQPVSTRSQHTQPTPEPSTAPSTSFLLPMGPLPPAEGSTGDFALPVGLIVG 
 140a 147 172 197 202 207 212 217 222 
WTp75 TSSTDICRPHQICNVVAIPGNASMDAVCTSTSPTRSMAPGAVHLPQPVSTRSQHTQPTPEPSTAPSTSFLLPMGPSPPAEGSTGDFALPVGLIVG 
Δ1 TSSTDICRPH-------------------------SMAPGAVHLPQPVSTRSQHTQPTPEPSTAPSTSFLLPMGPSPPAEGSTGDFALPVGLIVG 
Δ2 TSSTDICRPHQICNVVAIPGNASMDAVCTSTSPTR-------------------------PSTAPSTSFLLPMGPSPPAEGSTGDFALPVGLIVG 
Δ3 TSSTDICRPHQICNVVAIPGNASMDAVCTSTSPTRSMAPGAVHLPQPVSTRSQHTQPTPE-------------------------FALPVGLIVG 
Δ123 TSSTDICRPH---------------------------------------------------------------------------FALPVGLIVG 
Δ3b TSSTDICRPHQICNVVAIPGNASMDAVCTSTSPTRSMAPGAVHLPQPVSTRSQHTQPTPEPSTAP-----LPMGPSPPAEGSTGDFALPVGLIVG 
Δ3c TSSTDICRPHQICNVVAIPGNASMDAVCTSTSPTRSMAPGAVHLPQPVSTRSQHTQPTPEPSTAPSTSFL-----SPPAEGSTGDFALPVGLIVG 
Δ3d TSSTDICRPHQICNVVAIPGNASMDAVCTSTSPTRSMAPGAVHLPQPVSTRSQHTQPTPEPSTAPSTSFLLPMGP-----GSTGDFALPVGLIVG 
SRS TSSTDICRPHQICNVVAIPGNASMDAVCTSTSPTRSMAPGAVHLPQPVSTRSQHTQPTPEPSTAPSRSFLLPMGPSPPAEGSTGDFALPVGLIVG 
PMDP TSSTDICRPHQICNVVAIPGNASMDAVCTSTSPTRSMAPGAVHLPQPVSTRSQHTQPTPEPSTAPSTSFLLPMDPSPPAEGSTGDFALPVGLIVG 
GGS TSSTDICRPHQICNVVAIPGNASMDAVCTSTSPTRSMAPGAVHLPQPVSTRSQHTQPTPEPSTAPSTSFLLPMGGSPPAEGSTGDFALPVGLIVG 
GPL TSSTDICRPHQICNVVAIPGNASMDAVCTSTSPTRSMAPGAVHLPQPVSTRSQHTQPTPEPSTAPSTSFLLPMGPLPPAEGSTGDFALPVGLIVG 
a

Amino acid number.

We thank Profs. T. Hardingham and M. Feldmann for useful suggestions during this work and Drs. Dudhia and Gould for their comments and critical review of this manuscript.

1

This work was supported by the Arthritis and Rheumatism Council, United Kingdom.

4

Abbreviations used in this paper: STS, serine-threonine-serine; SRS, serine-arginine-serine; SFS, serine-phenylalanine-serine.

1
Bazil, V..
1995
. Physiological enzymatic cleavage of leukocyte membrane molecules.
Immunol. Today
16
:
135
2
Ehlers, M. R., J. F. Riordan.
1991
. Membrane proteins with soluble counterparts: role of proteolysis in the release of transmembrane proteins.
Biochemistry
30
:
10065
3
Rose-John, S., P. C. Heinrich.
1994
. Soluble receptors for cytokines and growth factors: generation and biological function.
Biochem. J.
300
:
281
4
Beutler, B., A. Cerami.
1989
. The biology of cachetin/TNF: a primary mediator of the host response.
Annu. Rev. Immunol.
7
:
625
5
Loetscher, H., Y. Pan, H. Lahm, R. Gentz, M. Brockhaus, H. Tabuchi, W. Lesslauer.
1990
. Molecular cloning and expression of the human 55kD tumour necrosis factor receptor.
Cell
61
:
351
-359.
6
Schall, T. J., M. Lewis, K. J. Koller, A. Lee, G. C. Rice, G. H. W. Wong, T. Gatanaga, G. A. Granger, R. Lentz, H. Raab, W. J. Kohr, D. V. Goeddel.
1990
. Molecular cloning and expression of a receptor for human tumour necrosis factor.
Cell
61
:
361
7
Gray, P., K. Barrett, D. Chantry, M. Turner, M. Feldmann.
1990
. Cloning of human tumour necrosis factor (TNF) receptor cDNA and expression of recombinant soluble TNF binding protein.
Proc. Natl. Acad. Sci. USA
87
:
7380
8
Smith, C., T. Davis, D. Anderson, L. Solam, M. Beckmann, R. Jerzy, S. Dower, D. Cosman, R. Goodwin.
1990
. A receptor for tumour necrosis factor defines an unusual family of cellular and viral proteins.
Science
248
:
1019
9
Dembic, Z., H. Loetscher, U. Gubler, Y. Pan, H. Lahm, R. Gentz, M. Brockhaus, W. Lesslauer.
1990
. Two human TNF receptors have similar extracellular, but distinct intracellular, domain sequences.
Cytokine
2
:
231
10
Lantz, M., U. Gullberg, E. Nilsson, I. Olsson.
1990
. Characterization in vitro of a human tumour necrosis factor binding protein, a soluble form of a tumour necrosis factor receptor.
J. Clin. Invest.
86
:
1396
11
Engelmann, H., D. Novick, D. Wallach.
1990
. Two tumour necrosis factor-binding proteins purified from human urine.
J. Biol. Chem.
265
:
1531
12
Khono, T., M. Brewer, S. Baker, P. Schwartz, M. King, K. Hale, C. Squires, R. Thompson, J. Vannice.
1990
. A second tumour necrosis factor receptor gene product can shed a naturally occurring tumour necrosis factor inhibitor.
Proc. Natl. Acad. Sci. USA
87
:
8331
13
Gatanaga, T., C. Hwang, M. Gatanaga, F. Cappuccini, R. Yamamoto, G. Granger.
1991
. The regulation of TNF receptor mRNA synthesis, membrane expression, and release by PMA- and LPS-stimulated human monocytic THP-1 cells in vitro.
Cell. Immunol.
138
:
1
14
Porteu, F., C. Nathan.
1990
. Shedding of tumour necrosis factor receptors by activated human neutrophils.
J. Exp. Med.
172
:
599
15
Aderka, D., H. Engelmann, Y. Maor, C. Brakebusch, D. Wallach.
1992
. Stabilization of the bioactivity of tumour necrosis factor by its soluble receptors.
J. Exp. Med.
175
:
323
16
Brakebusch, C., E. E. Varfolomeev, M. Batkin, D. Wallach.
1994
. Structural requirements for inducible shedding of the p55 tumour necrosis factor receptor.
J. Biol. Chem.
269
:
32488
17
Fernandez-Botran, R., E. Vitetta.
1990
. A soluble, high affinity, interleukin-4-binding-protein is present in the biological fluids of mice.
Proc. Natl. Acad. Sci. USA
87
:
4202
18
Goodwin, R., D. Friend, S. Ziegler, R. Jerzy, B. Falk, S. Gimpel, D. Cosman, S. Dower, C. March, A. Namen, L. Park.
1990
. Cloning of the human and murine interleukin-7 receptors: demonstration of a soluble form and homology to a new receptor superfamily.
Cell
60
:
941
19
Kriegler, M., C. Perez, K. De Fay, I. Albert, S. D. Lu.
1988
. A novel form of TNF/cachetin is a cell surface cytotoxic transmembrane protein: ramifications for the complex physiology of TNF.
Cell
53
:
45
20
Nophar, Y., O. Kemper, C. Brakebusch, H. Engelmann, R. Zwang, D. Aderka, H. Holtmann, D. Wallach.
1990
. Soluble forms of tumour necrosis factor receptors (TNF-Rs): the cDNA for type 1 TNF-R, cloned using amino acid sequence data of its soluble form, encodes both the cell surface and a soluble form of the receptor.
EMBO J.
9
:
3269
21
Perez, C., I. Albert, K. De Fay, N. Zachariades, L. Gooding, M. Kriegler.
1990
. A nonsecretable cell surface mutant of tumor necrosis factor (TNF) kills by cell-to-cell contact.
Cell
63
:
251
22
Gullberg, U., M. Lantz, L. Lindvall, I. Olsson, A. Himmler.
1992
. Involvement of an Asn/Val cleavage site in the production of a soluble form of human tumour necrosis factor (TNF) receptor: site directed mutagenesis of a putative cleavage site in the p55 TNF receptor chain.
Eur. J. Cell Biol.
58
:
307
23
Porteu, F., M. Brockhaus, D. Wallach, H. Engelmann, C. Nathan.
1991
. Human neutrophil elastase releases a ligand binding fragment from the 75-kDa tumour necrosis factor (TNF) receptor.
J. Biol. Chem.
266
:
18846
24
Crowe, P. D., B. N. Walter, K. M. Mohler, C. Otten-Evans, R. A. Black, C. F. Ware.
1995
. A metalloprotease inhibitor blocks shedding of the 80-kD TNF receptor and TNF processing in T-lymphocytes.
J. Exp. Med.
181
:
1205
25
Crowe, P., T. Van Arsdale, R. Goodwin, C. Ware.
1993
. Specific induction of 80-kDa tumour necrosis factor receptor shedding in T lymphocytes involves the cytoplasmic domain and phosphorylation.
J. Immunol.
151
:
6882
26
Pocsik, E., E. Duda, D. Wallach.
1995
. Phosphorylation of the 26 kDa TNF precursor in monocytic cells and in transfected HeLa cells.
J. Inflamm.
45
:
152
27
Kahn, J., R. I. Ingraham, F. Shirley, G. I. Migaki, T. K. Kishimoto.
1994
. Membrane proximal cleavage of L-selectin: identification of the cleavage site and a 6-kD transmembrane peptide fragment of L-selectin.
J. Cell Biol.
125
:
461
28
Van Wart, H. E., D. R. Steinbrink.
1981
. A continuous spectrophotometric assay for Clostridiumhistolyticum collagenase.
Anal. Biochem.
113
:
356
29
Leeuwenberg, J. F. M., T. M. A. A. Jeunhomme, W. A. Buurman.
1994
. Slow release of soluble TNF receptors by monocytes in vitro.
J. Immunol.
152
:
4036
30
Zeman, K., E. M. Paleolog, H. Tchorzewski, B. Szymanska.
1993
. Comparison of the functional activities of two different preparations of human recombinant tumour necrosis factor α.
Arch. Immunol. Ther. Exp.
41
:
159
31
Kunkel, T. A..
1987
. Oligonucleotide-directed mutagenesis without phenotypic selection. F. M. Ausubel, and R. Brent, and R. E. Kingston, and D. D. Moore, and J. G. Seidman, and J. A. Smith, and K. Struhl, eds.
Current Protocols in Molecular Biology
8.1.1
-8.1.6. Wiley Interscience, New York.
32
Busslinger, M., N. Moschonas, R. Flavell.
1981
. Beta+ thalassemia: aberrant splicing results from a single point mutant in an intron.
Cell
27
:
289
33
De Wet, J. R., K. V. Wood, M. DeLuca, D. R. Helinski, S. Subramani.
1987
. Firefly luciferase gene: structure and expression in mammalian cells.
Mol. Cell. Biol.
7
:
725
34
Greenwood, F. C., W. M. Hunter, J. S. Glover.
1963
. The preparation of 131I-labelled human growth hormone of high specific radioactivity.
Biochem. J.
89
:
114
35
Chen, P. C. H., G. C. Du Bois, M. J. Chen.
1995
. Mapping the domains critical for the binding of human tumour necrosis factor α to its two receptors.
J. Biol. Chem.
270
:
2874
36
Porteu, F., C. Hieblot.
1994
. Tumour necrosis factor induces a selective shedding of its p75 receptor from human neutrophils.
J. Biol. Chem.
269
:
2834
37
Hwang, C., M. Gatanaga, G. Granger, T. Gatanaga.
1993
. mechanism of release of soluble forms of tumour necrosis factor/lymphotoxin receptors by phorbol myristate acetate stimulated human THP-1 cells in vitro.
J. Immunol.
151
:
5631
38
Robache-Gallea, S., V. Morand, J. M. Bruneau, B. Schoot, E. Tagat, E. Realo, S. Chuaib, S. Roman-Roman.
1995
. In vitro processing of human tumour necrosis factor α.
J. Biol. Chem.
270
:
23688
39
Gearing, A. J. H., P. Beckett, M. Christodoulou, M. Churchill, J. Clements, A. H. Davidson, A. H. Drummond, W. A. Galloway, R. Gilbert, J. L. Gordon, T. M. Leber, M. Mangan, K. Miller, P. Nayee, K. Owen, S. Patel, W. Thomas, G. Wells, L. M. Wood, K. Woolley.
1994
. Processing of tumour necrosis factor-α precursor by metalloproteinases.
Nature
370
:
555
40
McGeehan, G., J. D. Becherer, R. C. Bast, C. M. Boyer, B. Champion, K. M. Connolly, J. G. Conway, P. Furdon, S. Karp, S. Kidao, A. B. McElroy, J. Nichols, K. M. Pryzwansky, F. Schoenen, L. Sekut, A. Truesdale, M. Verghese, J. Warner, J. P. Ways.
1994
. Regulation of tumour necrosis factor a processing by a metalloproteinase inhibitor.
Nature
370
:
558
41
Mohler, K. M., P. R. Sleath, J. N. Fitzner, D. P. Cerretti, M. Alderson, S. S. Kerwar, D. S. Torrance, C. Otten-Evans, T. Greenstreet, K. Weerawarna, S. R. Kronheim, M. Peterson, M. Gerhart, C. J. Kozlosky, C. J. March, R. A. Black.
1994
. Protection against a lethal dose of endotoxin by an inhibitor of tumour necrosis factor processing.
Nature
370
:
218
42
Mullberg, J., F. H. Durie, C. Otten-Evans, M. R. Alderson, S. Rose-John, D. Cosman, R. A. Black, K. M. Mohler.
1995
. A metalloprotease inhibitor blocks shedding of the IL-6 receptor and the p60 TNF receptor.
J. Immunol.
155
:
5198
43
Williamson, M. P..
1994
. The structure and function of proline-rich regions.
Biochem. J.
297
:
249
44
Dando, P. M., M. A. Brown, A. J. Barrett.
1993
. Human thimet oligopeptidase.
Biochem. J.
294
:
451
45
Barelli, H., J. Vincent, F. Checler.
1993
. Rat kidney endopeptidase 24.16: purification, physico-chemical characteristics and differential specificity towards opiates, tachykinins and neurotensin-related peptides.
Eur. J. Biochem.
211
:
79
46
Yoshimoto, T., T. Nishimura, T. Kita, D. Tsuru.
1983
. Post-proline cleaving enzyme (prolyl endopeptidase) from bovine brain.
J. Biochem.
94
:
1179
47
Migaki, G. I., J. Kahn, T. K. Kishimoto.
1995
. Mutational analysis of the membrane-proximal cleavage site of L-selectin: relaxed sequence specificity surrounding the cleavage site.
J. Exp. Med.
182
:
549
48
Joyce, D. A., D. P. Gibbons, P. Green, J. H. Steer, M. Feldmann, F. M. Brennan.
1994
. Two inhibitors of pro-inflammatory cytokine release, interleukin-10 and interleukin-4, have contrasting effects on release of soluble p75 tumour necrosis factor receptor by cultured monocytes.
Eur. J. Immunol.
24
:
2699
49
Williams, L., D. Gibbons, A. Gearing, R. Maini, M. Feldmann, F. Brennan.
1996
. Paradoxical effects of a synthetic metalloproteinase inhibitor that blocks both p55 and p75 TNF receptor shedding and TNF α processing in RA synovial membrane cell cultures.
J. Clin. Invest.
97
:
2833
50
Black, R., F. Durie, C. Otten-Evans, R. Miller, J. Slack, D. Lynch, B. Castner, K. Mohler, M. Gerhart, R. Johnson, Y. Itoh, Y. Okada, H. Nagase.
1996
. Relaxed specificity of matrix metalloproteinases (MMPS) and TIMP insensitivity of tumor necrosis factor-α (TNF-α) production suggest the major TNF-α converting enzyme is not an MMP.
Biochem. Biophys. Res. Commun.
225
:
400
51
Black, R. A., C. T. Rauch, C. J. Kozlosky, J. J. Peschon, J. L. Slack, M. F. Wolfson, B. J. Castner, K. L. Stocking, P. Reddy, S. Srinivasan, N. Nelson, N. Boiani, K. A. Schooley, M. Gerhart, R. Davis, J. N. Fitzner, R. S. Johnson, R. J. Paxton, C. J. March, D. P. Cerretti.
1997
. A metalloprotease disintegrin that releases tumour-necrosis factor α from cells.
Nature
385
:
729
52
Moss, M. L., S. L. C. Jin, M. E. Milla, W. Burkhart, H. L. Carter, W. J. Chen, W. C. Clay, J. R. Didsbury, D. Hassler, C. R. Hoffman, T. A. Kost, M. H. Lambert, M. A. Leesnitzer, P. McCauley, G. McGeehan, J. Mitchell, M. Moyer, G. Pahel, W. Rocque, L. K. Overton, F. Schoenen, T. Seaton, J. L. Su, J. Warner, D. Willard, J. D. Becherer.
1997
. Cloning of a disintegrin metalloprotease that processes precursor tumour-necrosis factor α.
Nature
385
:
733
53
Wolfsberg, T. G., P. Primakoff, D. G. Myles, J. M. White.
1995
. ADAM, a novel family of membrane proteins containing a disintegrin and metalloprotease domain: multipotential functions in cell-cell and cell-matrix interactions.
J. Cell Biol.
131
:
275