Soluble TNF receptor in LPS shock
Eur. J. Immunol. 1991. 21: 2883-2886
2883
Short paper Werner Lesslauer, Hisahiro Tabuchi, Reiner Gentz, Manfred Brockhaus, Emst Juergen Schlaeger, Georges Graua, Pierre Frangois PiguetA, Pascal Pointairen, Pierre Vassallia and Hansruedi Loetscher
F. Hoffmann-La Roche Ltd., Basel and Department of PathologyA, CMU,University of Geneva, Geneva
Recombinant soluble tumor necrosis factor receptor proteins protect mice from lipopolysaccharide-induced lethality The in vivo efficacy of human recombinant soluble tumor necrosis factor (TNF) receptor protein to prevent and to treat lipopolysaccharide (LPS)-induced lethal toxicity in D-galactosamine-treated mice was investigated. Chimeric proteins of the receptor extracellulardomains fused to the hinge region of human IgG3 were expressed in myeloma cells (rST"R-hy3). The fusion proteins had a disulfidebonded dimeric structure. Upon intravenousinjection, their serum concentration decreased relatively slowly after an initial phase of rapid elimination. Dgalactosamine-sensitized mice were fully protected from the toxic effects of LPS, if the animal were pretreated with rsTNFR-hy3 at 20 pglanimal. Partial protection was seen at significantly lower doses and when rdJNFX-hy3 was given up to 3 h after LPS.
1 Introduction
TNF-BP is not fully understood; it has been proposed that they may act as neutralizing buffers of TNF bioactivity, or, Septic shock as a consequence of Gram-negative alternatively as reservoir from which TNF is slowly rebacteremia or endotoxemia in spite of adequate antibiotic leased. We have expressed recombinant soluble proteins therapy remains a critical clinical condition. It is now composed of the complete extracellular regions of the two known that the lethal consequences of septic shock result TNFR fused to human IgG3 and tested their bioactivity in a from an exaggerated host response, mediated by protein murine model of LPS-induced lethality. factors such asTNFand IL 1, rather than from the pathogen directly. For example, LPS has been shown to elicit the release of TNF leading to a strong and transient increase of 2 Materials and methods its serum concentration with a peak maximum about 1-2 h after LPS in man [l-31 and in animal studies [4, 51. 2.1 Construction of vectors, expression and purification Furthermore, TNF causes shock and tissue injury when administered to rats in quantities similar to those induced The cDNA encoding the extracellular domains of TNFRa by endotoxin [6], and passive immunization of primates and TNFRB were amplified by the polymerase chain with antibodies to TNF improves survival during lethal reaction (PCR). Unique restriction sites were introduced at E. coli bacteremia [7].The severity of meningococcemia in both ends of the fragments. The rsTNFRa-hy3 and children correlates with the serum levels of TNF, but also rsTNFRB-hy3fusion constructs were prepared by exchangwith that of IL 1 and IFN-y [8]. In view of these findings, ing the CD4 sequence in the pCD4-hy3-4 vector (kindly TNF is thought to be one of the important early mediators provided by K. Karjalainen and A. Traunecker, Basel in septic shock. However, septic shock has a multifactorial Institute of Immunology) with the extracellular domain pathogenesis; this is reflected by the fact that a receptor sequences of TNFRa and TNFRB [19,20]. The receptor antagonist (ra) to IL 1, IL Ira, or an anti-IL6 mAb also sequences in the chimera thus were inserted in front of the protect against lethal effects of endotoxemia or of E. coli hinge region of the human IgG Cy3 chain. J558L myeloma sepsis in mice [9-111. cells were transfected with the fusion constructs and the expression analyzed as described elsewhere [20].The fusion A few years ago three groups discovered TNF inhibitory proteins were purified from cell-free SN by protein G proteins, now termed TNF-BP I and 11, in human serum affinity chromatography(Pharmacia-LKB, Uppsala, Swedand urine [12-171. TNF-BP I and I1 now are recognized to en) and gel filtration. SDS-PAGE analysis was performed be truncated fragments of the extracellular regions of the according to Laemmli [21]. The affinities of rsTNFRa-hy3 twoTNF receptors,TNFRa (75 kDa) and TNFRP (55 kDa) and rsTNFRp-hy3 for TNF were determined in a solid(for a review see [18]).The physiological function of the two phase assay system using mAb utr4 and htr20, respectively POI. [I 97291 Correspondence: Werner Lesslauer, F. Hoffmann-La Roche Ltd., 2.2 Immunoassays Bldg. 15, Room 6, (33-4002 Basel, Switzerland
Abbreviatons: &NFR u, fl:Recombinant soluble extracellular region of the human TNF receptor type a,P rsTNFR-hy3: Recombinant chimeric protein bf rsTNFR fused to hinge region of human Cy3& chain 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1991
Details of the rsTNFR assay are described elsewhere*. Briefly, 96-well microtiter plates from polyvinyl (Dynatech,
*
Digel, W., Porzsolt, E, Lessiauer, W. and Brockhaus, M.; submitted for publication.
+
0014-2980/91/1111-2883$3.50 .25/0
2884
Eur. J. Immunol. 1991.21: 2883-2886
W. Lesslauer, H. IIgbuchi, R. Gentz et al.
Alexandria, VA) were sensitized with affinity-purified rabbit anti-mouse Ig (10 pg/ml PBS) followed by overnight incubation with the non-inhibitory antibody htr-20 (10 pg/ml in 50 mM Tris-HC1, pH 7.4,140 mM NaCl, 5 mM EDTA, 1% defatted milk powder) against TNFRP. Mouse sera were diluted between 1:10 and 1:lo00 with the same buffer, in order to measure within the linear range of the assay. Diluted samples (45 pl) were incubated in the wells for 3 h at 4 "C. Following removal of the serum and washing of the wells (1 x 0.25 ml PBS) freeTNF-binding sites were detected by incubation with '=I-TNF (450Ci/mmol = 16.65 TBq/mmol, 106 cpm/ml) for 2 h at 4"C.Wells were washed four times with PBS and the residual radioactivity of each well was measured by gamma scintillation counting. Results are the mean of two independent experiments.The assay was calibrated with purified rsTNFRP protein expressed in Sf9 cells.
vivo to allow for a treatment protocol using single bolus injections,various amounts of rsTNFRfLhy3 in the range of 1to 100 pg/animal were injected i.v. into mice. Subsequently, the serum concentrations of the fusion protein were determined after 1, 6, 24, 48, 72 and 96h using a sandwich-type immunoassay (Fig. 2). It can be estimated that about 10% of the administered rsTNFRP-hy3 remained in circulation 1h after injection. This fast initial rate of elimination indicates extravasation and degradation
(B)
200 2.3 Endotoxin lethality model
- 92 - 66
Female BALB/c or (BALB/c x CBA/J)Fl mice, aged 8-10 weeks (from Iffa-Credo, Les Oncins, France), were injected i.p. with a mixture of E. coli LPS (055 :B5, from Difco, Detroit, MI; 10 or 100 pg in 0.1 ml sterile PBS) and D-galactosamine (D-GalN, from Sigma, St. Louis, MO; 18 mg in 0.1 ml sterile PBS), in order to sensitize to the lethal effects of LPS [22, 231. Mortality was monitored after 24, 48 and 72h. No further mortality was seen after 72 h.
3 Results and discussion 3.1 Characterization of -a-hy3 fusion proteins
and flNFRP-hy3
- 45 - 31 Figure 1. SDS-PAGE analysis of rsTNFRa-hy3 and rsTNFRP-hy3 fusion proteins. One microgram of affinity-purified rsTNFRa-hy3 and rtTNFRP-hy3 was denatured in SDS sample buffer in the absence (A) or presence (B) of D'IT and applied to a 12% polyacrylamide gel. The proteins were stained with Serva blue. Numbers on the right indicate the position of standard M, marker
The fusion proteins were expressed in mouse myeloma cells proteins. with a yield of about 0.5-5 pg/ml. SDS-PAGE analysis of the purified proteins revealed for both constructs a disulfide-bonded dimeric structure migrating on nonreducing O0O0 gels with an M, of about 180 kDa (rsTNFRa-hy3) and 150 kDa (rSINFRP-hy3). After reduction of rsTNFRa-hy3 and rsTNFRP-hy3 with DTT, the subunits migrated at about 80 kDa and 70 kDa, respectively (Fig. 1).The slight1000 ly higher molecular mass of rsTNFRa-hy3 reflects the larger size of the extracellular domain of the receptor E type a. Solid-phase binding assays with rsTNFRP-hy3 \ revealed a single class of high-affinity binding sites for cn 100 TNFa and TMF-P with Kd values of 0.10 and 0.12nM, c respectively [20]. Similarly, rs"FRa-hy3 also displayed a single high-affinity binding site for TNF-a and TNF-fJ with I(d values of 0.21 and 0.20 nM, respectively. The TNFR fusion proteins were chosen for the present studies, because they were expected to have a longer in vivo half-life (see Sect. 3.2) and to bind TNF with higher apparent affinity than the soluble extracellular domains of the TNFR. Indeed, monovalent rsTNFRp was found to bind TNF-a with about threefold lower affinity than rsTNFRp-hy3 0 suggestingthat the bivalency of the fusion protein results in an increased binding avidity [20].
f
-
3.2 I n vivo half-life of rsTNFRP-hy3
To investigate whether the receptor-IgG3 fusion proteins have a sufficiently long half-life in systemic circulation in
24
48
72
96
hours
Figure 2. In vivo half-life of rsTNFRP-hy3 fusion protein in mice. Mice were injected i.v.with 100,30,10,3and 1 pg of rsTNFRP-hy3 (curves from top to bottom), and the serum concentrations of the fusion protein were determined after 1,6,24,48,72 and 96 h.The mean and range of variation of two independent experiments are
given.
Eur. J. Immunol. 1991. 21: 2883-2886 Table 1. rSTNFRB-hy3 fusion protein prevents acute LPS-induced lethality in D-GalN-sensitized mice
Soluble TNF receptor in LPS shock
2885
Table2. Protective effect of rdJTFRB-hy3 when given after LPS Survival at:
0 100 50
2o 10 5 1 0.1
of the recombinant protein. After the initial phase, the concentration of rsTNFRP-hy3 decreased slowly as compared to the time scale of the animal model to be used; an approximatehalf-life of about 1day may be estimated from Fig. 2. 3.3 TNFR-IgG3fusion proteins prevent LPS-induced lethality
Groups of mice were treated i.v. with various doses of rsTNFRP-hy3 in sterile PBS 4 h before the concomitant injection of LPS and D-GalN and survival was assayed for up to 72 h (Table 1). Complete protection from acute death was seen with rsTNFRP-hy3at doses down to 20 pg/animal, whereas 10or 5 pg/animal provided only partial protection. It is noteworthy, however, that in one study partial protection was seen even at a 0.1 pg/animal dose. 3.4 TNFR-IgG3fusion proteins protect from LPS-induced lethality when given a&er LPS
The potential efficacy of a delayed administration of TNFR-IgG3 fusion proteins to treat an ongoing LPS toxicity syndrome was also investigated. rsTNFRP-hy3was injected i.v. at 50 pg/animal at various times in the range of -4 h to +3 h relative to the injection of a standard dose of LPS and D-GalN (Thble 2). Unlike anti-TNF antibodies in previously reported studies [5,24], rsTNFRP-hy3 (and to a lesser extent riINFRa-hy3; data not shown) was protective, although partially, when given 1 and even 3 h after LPS.
4 Concluding remarks In view of the toxicity of systemicTNF it is to be expected that mechanisms exist to limit TNF activities. It has been proposed that the functional significance of the naturally occurring soluble fragments of the two "FR,TNF-BP I and 11, is to neutralize excess systemic TNE Our finding that exogenous recombinant soluble TNFR protein protects from LPS-induced lethality supports this view; it may also point to novel therapeutic strategiesin the treatment of conditions caused or aggravated by excessive TNE
PBS only -4h -2 h
48h
72 h
115
u5 515 515 445 45 315
YS
+1 h
515 45 45
+3h
Y5
-0
a) Fifty micrograms rsTNFRB-hy3 i.v., relative to the injection of 10 pg of 055 :BSLPS.
Received July 8, 1991.
5 References 1 Michie, H. R., Manogue, K.R., Sprigs, D. R., Revhaug, A., ODwyer, S., Dinarello, C. A., Cerami, A., Wolff, S. M. and Wilmore, D.W., N . Engl. J. Med. 1988. 318: 1481. 2 Zabel, F!,Wolter, D. T., Schoenharting, M. M. and Schade, U. E, Lancet 1989. ii: 1474. 3 Spinas, G. A., Bloesch, D., Keller, U., Zimmerli, W. and Cammisuli, S., J. Infect. Dis. 1991. 163: 89. 4 Piguet, F! F., Grau, G. and Vassalli,F!,Am. J. Pathol. 1990.136: 103. 5 Beutler, B., Milsark, I.W. and Cerami, A. C., Science 1985.229: 869. 6 Tracey, K.J., Beutler, B., Lowry, S. F., Merryweather,J. ,Wolpe, S., Milsark, I. W., Hariri, R. J., Fahey, T. J., Zentella, A., Albert, J. D., Shires, G.T. and Cerami, A., Science 1986.234: 470. 7 Tracey, K., Fong,Y., Hesse, D. G., Manogue, K.R., Lee, A.T., Kuo, G. C., Lowry, S. F. and Cerami, A., Nature 1987. 330: 662. 8 Girardin, E., Grau, G. E., Dayer, J. M., Roux-Lombard, F,! Group,T. J. S. and Lambert, F! H., N. Engl. J. Med. 1988.319: 397. 9 Alexander, H. R., Doherty, G. M.,Buresh, C. M.V,VenzonD. J. and Norton, J. A., J. Exp. Med. 1991. 173: 1029. 10 Ohlsson, K., Bjork, F?, Bergenfeldt, M., Hageman, R. and Thompson, R. C, Nature 1990. 348: 550. 11 Stames, H. F., Peace, M. K.,Tewari, A.,Yim, J. H., Zou,J. C. and Abrams, J. S., J. Immunol. 1990. 145: 4185. 12 Seckinger, F!, Isaaz, S. and Dayer, J.-M., J. Exp. Med. 1988. 167: 1511. 13 Seckinger, F!, Isaaz, S. and Dayer, J.-M., J. Biol. Chem. 1989. 264: 11966. 14 Peetre, C., Thysell, H., Grubb, A. and Olsson, I., Eur. J. Haematol. 1988. 41: 414. 15 Olsson, I., Lantz, M., Nilsson, E., Peetre, C., Thysell, H., Grubb, A. and Adolf. G., Eur. J. Haematol. 1989. 42: 270. 16 Englemann, H., Aderka, D., Rubinstein, M., Rotman, D. and Wallach, D., J. Biol. Chem. 1989.264: 11974. 17 Engelmann, H., Novick, D. and Wallach, D., J. Biol. Chem. 1990. 265: 1531. 18 Loetscher, H. R., Brockhaus, M., DembiC, Z., Gentz, R., Gubler, U., Hohmann, H. F!, Lahm, H. W.,Van Loon, A. F! G. M., Pan,Y-C. E., Schlaeger, E. J., Steinmetz, M.,Bbuchi, H. and Lesslauer, W., in McLean, N. (Ed.), Oxford Survey on Eucaryotic Genes, vol. 7, Oxford University Press, Oxford 1991, in press.
2886
W. Lesslauer, H. Bbuchi, R. Gentz et al.
19 Traunecker, A., Schneider, J., Kiefer, H. and Karjalainen, K., Nature 1989. 339: 68. 20 Loetscher, H. R., Gem, R., Zulauf, M., Lustig, A., Tabuchi, H., Schlaeger, E. J., Brockhaus, M., Gallati, H., Manneberg, M. and Lesslauer, W., . I Biol. . Chem. 1991, in press. 21 Laemmli, U. K., Nature 1970. 227: 680.
Eur. J. h u n o l . 1991.21: 2883-2886 22 Lehmann,V, Freudenberg, M. A. and Galanos, C., J. Exp. Med. 1987. 165: 657. 23 Rothstein, J. L. and Schreiber, H., Proc. Nutl. Acud. Sci. USA 1988. 85: 607. 24 Grau, G. E.,Vesen, D., De Groote, D., De la Croux, D., Gisler, C., Piguet, F! F. and Lambert, I? H,, Clin.Exp. Zmmunol. 1991. 84: 411.
Eur. J. Immunol. 1991. 21: 2883-2886
2883
Short paper Werner Lesslauer, Hisahiro Tabuchi, Reiner Gentz, Manfred Brockhaus, Emst Juergen Schlaeger, Georges Graua, Pierre Frangois PiguetA, Pascal Pointairen, Pierre Vassallia and Hansruedi Loetscher
F. Hoffmann-La Roche Ltd., Basel and Department of PathologyA, CMU,University of Geneva, Geneva
Recombinant soluble tumor necrosis factor receptor proteins protect mice from lipopolysaccharide-induced lethality The in vivo efficacy of human recombinant soluble tumor necrosis factor (TNF) receptor protein to prevent and to treat lipopolysaccharide (LPS)-induced lethal toxicity in D-galactosamine-treated mice was investigated. Chimeric proteins of the receptor extracellulardomains fused to the hinge region of human IgG3 were expressed in myeloma cells (rST"R-hy3). The fusion proteins had a disulfidebonded dimeric structure. Upon intravenousinjection, their serum concentration decreased relatively slowly after an initial phase of rapid elimination. Dgalactosamine-sensitized mice were fully protected from the toxic effects of LPS, if the animal were pretreated with rsTNFR-hy3 at 20 pglanimal. Partial protection was seen at significantly lower doses and when rdJNFX-hy3 was given up to 3 h after LPS.
1 Introduction
TNF-BP is not fully understood; it has been proposed that they may act as neutralizing buffers of TNF bioactivity, or, Septic shock as a consequence of Gram-negative alternatively as reservoir from which TNF is slowly rebacteremia or endotoxemia in spite of adequate antibiotic leased. We have expressed recombinant soluble proteins therapy remains a critical clinical condition. It is now composed of the complete extracellular regions of the two known that the lethal consequences of septic shock result TNFR fused to human IgG3 and tested their bioactivity in a from an exaggerated host response, mediated by protein murine model of LPS-induced lethality. factors such asTNFand IL 1, rather than from the pathogen directly. For example, LPS has been shown to elicit the release of TNF leading to a strong and transient increase of 2 Materials and methods its serum concentration with a peak maximum about 1-2 h after LPS in man [l-31 and in animal studies [4, 51. 2.1 Construction of vectors, expression and purification Furthermore, TNF causes shock and tissue injury when administered to rats in quantities similar to those induced The cDNA encoding the extracellular domains of TNFRa by endotoxin [6], and passive immunization of primates and TNFRB were amplified by the polymerase chain with antibodies to TNF improves survival during lethal reaction (PCR). Unique restriction sites were introduced at E. coli bacteremia [7].The severity of meningococcemia in both ends of the fragments. The rsTNFRa-hy3 and children correlates with the serum levels of TNF, but also rsTNFRB-hy3fusion constructs were prepared by exchangwith that of IL 1 and IFN-y [8]. In view of these findings, ing the CD4 sequence in the pCD4-hy3-4 vector (kindly TNF is thought to be one of the important early mediators provided by K. Karjalainen and A. Traunecker, Basel in septic shock. However, septic shock has a multifactorial Institute of Immunology) with the extracellular domain pathogenesis; this is reflected by the fact that a receptor sequences of TNFRa and TNFRB [19,20]. The receptor antagonist (ra) to IL 1, IL Ira, or an anti-IL6 mAb also sequences in the chimera thus were inserted in front of the protect against lethal effects of endotoxemia or of E. coli hinge region of the human IgG Cy3 chain. J558L myeloma sepsis in mice [9-111. cells were transfected with the fusion constructs and the expression analyzed as described elsewhere [20].The fusion A few years ago three groups discovered TNF inhibitory proteins were purified from cell-free SN by protein G proteins, now termed TNF-BP I and 11, in human serum affinity chromatography(Pharmacia-LKB, Uppsala, Swedand urine [12-171. TNF-BP I and I1 now are recognized to en) and gel filtration. SDS-PAGE analysis was performed be truncated fragments of the extracellular regions of the according to Laemmli [21]. The affinities of rsTNFRa-hy3 twoTNF receptors,TNFRa (75 kDa) and TNFRP (55 kDa) and rsTNFRp-hy3 for TNF were determined in a solid(for a review see [18]).The physiological function of the two phase assay system using mAb utr4 and htr20, respectively POI. [I 97291 Correspondence: Werner Lesslauer, F. Hoffmann-La Roche Ltd., 2.2 Immunoassays Bldg. 15, Room 6, (33-4002 Basel, Switzerland
Abbreviatons: &NFR u, fl:Recombinant soluble extracellular region of the human TNF receptor type a,P rsTNFR-hy3: Recombinant chimeric protein bf rsTNFR fused to hinge region of human Cy3& chain 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1991
Details of the rsTNFR assay are described elsewhere*. Briefly, 96-well microtiter plates from polyvinyl (Dynatech,
*
Digel, W., Porzsolt, E, Lessiauer, W. and Brockhaus, M.; submitted for publication.
+
0014-2980/91/1111-2883$3.50 .25/0
2884
Eur. J. Immunol. 1991.21: 2883-2886
W. Lesslauer, H. IIgbuchi, R. Gentz et al.
Alexandria, VA) were sensitized with affinity-purified rabbit anti-mouse Ig (10 pg/ml PBS) followed by overnight incubation with the non-inhibitory antibody htr-20 (10 pg/ml in 50 mM Tris-HC1, pH 7.4,140 mM NaCl, 5 mM EDTA, 1% defatted milk powder) against TNFRP. Mouse sera were diluted between 1:10 and 1:lo00 with the same buffer, in order to measure within the linear range of the assay. Diluted samples (45 pl) were incubated in the wells for 3 h at 4 "C. Following removal of the serum and washing of the wells (1 x 0.25 ml PBS) freeTNF-binding sites were detected by incubation with '=I-TNF (450Ci/mmol = 16.65 TBq/mmol, 106 cpm/ml) for 2 h at 4"C.Wells were washed four times with PBS and the residual radioactivity of each well was measured by gamma scintillation counting. Results are the mean of two independent experiments.The assay was calibrated with purified rsTNFRP protein expressed in Sf9 cells.
vivo to allow for a treatment protocol using single bolus injections,various amounts of rsTNFRfLhy3 in the range of 1to 100 pg/animal were injected i.v. into mice. Subsequently, the serum concentrations of the fusion protein were determined after 1, 6, 24, 48, 72 and 96h using a sandwich-type immunoassay (Fig. 2). It can be estimated that about 10% of the administered rsTNFRP-hy3 remained in circulation 1h after injection. This fast initial rate of elimination indicates extravasation and degradation
(B)
200 2.3 Endotoxin lethality model
- 92 - 66
Female BALB/c or (BALB/c x CBA/J)Fl mice, aged 8-10 weeks (from Iffa-Credo, Les Oncins, France), were injected i.p. with a mixture of E. coli LPS (055 :B5, from Difco, Detroit, MI; 10 or 100 pg in 0.1 ml sterile PBS) and D-galactosamine (D-GalN, from Sigma, St. Louis, MO; 18 mg in 0.1 ml sterile PBS), in order to sensitize to the lethal effects of LPS [22, 231. Mortality was monitored after 24, 48 and 72h. No further mortality was seen after 72 h.
3 Results and discussion 3.1 Characterization of -a-hy3 fusion proteins
and flNFRP-hy3
- 45 - 31 Figure 1. SDS-PAGE analysis of rsTNFRa-hy3 and rsTNFRP-hy3 fusion proteins. One microgram of affinity-purified rsTNFRa-hy3 and rtTNFRP-hy3 was denatured in SDS sample buffer in the absence (A) or presence (B) of D'IT and applied to a 12% polyacrylamide gel. The proteins were stained with Serva blue. Numbers on the right indicate the position of standard M, marker
The fusion proteins were expressed in mouse myeloma cells proteins. with a yield of about 0.5-5 pg/ml. SDS-PAGE analysis of the purified proteins revealed for both constructs a disulfide-bonded dimeric structure migrating on nonreducing O0O0 gels with an M, of about 180 kDa (rsTNFRa-hy3) and 150 kDa (rSINFRP-hy3). After reduction of rsTNFRa-hy3 and rsTNFRP-hy3 with DTT, the subunits migrated at about 80 kDa and 70 kDa, respectively (Fig. 1).The slight1000 ly higher molecular mass of rsTNFRa-hy3 reflects the larger size of the extracellular domain of the receptor E type a. Solid-phase binding assays with rsTNFRP-hy3 \ revealed a single class of high-affinity binding sites for cn 100 TNFa and TMF-P with Kd values of 0.10 and 0.12nM, c respectively [20]. Similarly, rs"FRa-hy3 also displayed a single high-affinity binding site for TNF-a and TNF-fJ with I(d values of 0.21 and 0.20 nM, respectively. The TNFR fusion proteins were chosen for the present studies, because they were expected to have a longer in vivo half-life (see Sect. 3.2) and to bind TNF with higher apparent affinity than the soluble extracellular domains of the TNFR. Indeed, monovalent rsTNFRp was found to bind TNF-a with about threefold lower affinity than rsTNFRp-hy3 0 suggestingthat the bivalency of the fusion protein results in an increased binding avidity [20].
f
-
3.2 I n vivo half-life of rsTNFRP-hy3
To investigate whether the receptor-IgG3 fusion proteins have a sufficiently long half-life in systemic circulation in
24
48
72
96
hours
Figure 2. In vivo half-life of rsTNFRP-hy3 fusion protein in mice. Mice were injected i.v.with 100,30,10,3and 1 pg of rsTNFRP-hy3 (curves from top to bottom), and the serum concentrations of the fusion protein were determined after 1,6,24,48,72 and 96 h.The mean and range of variation of two independent experiments are
given.
Eur. J. Immunol. 1991. 21: 2883-2886 Table 1. rSTNFRB-hy3 fusion protein prevents acute LPS-induced lethality in D-GalN-sensitized mice
Soluble TNF receptor in LPS shock
2885
Table2. Protective effect of rdJTFRB-hy3 when given after LPS Survival at:
0 100 50
2o 10 5 1 0.1
of the recombinant protein. After the initial phase, the concentration of rsTNFRP-hy3 decreased slowly as compared to the time scale of the animal model to be used; an approximatehalf-life of about 1day may be estimated from Fig. 2. 3.3 TNFR-IgG3fusion proteins prevent LPS-induced lethality
Groups of mice were treated i.v. with various doses of rsTNFRP-hy3 in sterile PBS 4 h before the concomitant injection of LPS and D-GalN and survival was assayed for up to 72 h (Table 1). Complete protection from acute death was seen with rsTNFRP-hy3at doses down to 20 pg/animal, whereas 10or 5 pg/animal provided only partial protection. It is noteworthy, however, that in one study partial protection was seen even at a 0.1 pg/animal dose. 3.4 TNFR-IgG3fusion proteins protect from LPS-induced lethality when given a&er LPS
The potential efficacy of a delayed administration of TNFR-IgG3 fusion proteins to treat an ongoing LPS toxicity syndrome was also investigated. rsTNFRP-hy3was injected i.v. at 50 pg/animal at various times in the range of -4 h to +3 h relative to the injection of a standard dose of LPS and D-GalN (Thble 2). Unlike anti-TNF antibodies in previously reported studies [5,24], rsTNFRP-hy3 (and to a lesser extent riINFRa-hy3; data not shown) was protective, although partially, when given 1 and even 3 h after LPS.
4 Concluding remarks In view of the toxicity of systemicTNF it is to be expected that mechanisms exist to limit TNF activities. It has been proposed that the functional significance of the naturally occurring soluble fragments of the two "FR,TNF-BP I and 11, is to neutralize excess systemic TNE Our finding that exogenous recombinant soluble TNFR protein protects from LPS-induced lethality supports this view; it may also point to novel therapeutic strategiesin the treatment of conditions caused or aggravated by excessive TNE
PBS only -4h -2 h
48h
72 h
115
u5 515 515 445 45 315
YS
+1 h
515 45 45
+3h
Y5
-0
a) Fifty micrograms rsTNFRB-hy3 i.v., relative to the injection of 10 pg of 055 :BSLPS.
Received July 8, 1991.
5 References 1 Michie, H. R., Manogue, K.R., Sprigs, D. R., Revhaug, A., ODwyer, S., Dinarello, C. A., Cerami, A., Wolff, S. M. and Wilmore, D.W., N . Engl. J. Med. 1988. 318: 1481. 2 Zabel, F!,Wolter, D. T., Schoenharting, M. M. and Schade, U. E, Lancet 1989. ii: 1474. 3 Spinas, G. A., Bloesch, D., Keller, U., Zimmerli, W. and Cammisuli, S., J. Infect. Dis. 1991. 163: 89. 4 Piguet, F! F., Grau, G. and Vassalli,F!,Am. J. Pathol. 1990.136: 103. 5 Beutler, B., Milsark, I.W. and Cerami, A. C., Science 1985.229: 869. 6 Tracey, K.J., Beutler, B., Lowry, S. F., Merryweather,J. ,Wolpe, S., Milsark, I. W., Hariri, R. J., Fahey, T. J., Zentella, A., Albert, J. D., Shires, G.T. and Cerami, A., Science 1986.234: 470. 7 Tracey, K., Fong,Y., Hesse, D. G., Manogue, K.R., Lee, A.T., Kuo, G. C., Lowry, S. F. and Cerami, A., Nature 1987. 330: 662. 8 Girardin, E., Grau, G. E., Dayer, J. M., Roux-Lombard, F,! Group,T. J. S. and Lambert, F! H., N. Engl. J. Med. 1988.319: 397. 9 Alexander, H. R., Doherty, G. M.,Buresh, C. M.V,VenzonD. J. and Norton, J. A., J. Exp. Med. 1991. 173: 1029. 10 Ohlsson, K., Bjork, F?, Bergenfeldt, M., Hageman, R. and Thompson, R. C, Nature 1990. 348: 550. 11 Stames, H. F., Peace, M. K.,Tewari, A.,Yim, J. H., Zou,J. C. and Abrams, J. S., J. Immunol. 1990. 145: 4185. 12 Seckinger, F!, Isaaz, S. and Dayer, J.-M., J. Exp. Med. 1988. 167: 1511. 13 Seckinger, F!, Isaaz, S. and Dayer, J.-M., J. Biol. Chem. 1989. 264: 11966. 14 Peetre, C., Thysell, H., Grubb, A. and Olsson, I., Eur. J. Haematol. 1988. 41: 414. 15 Olsson, I., Lantz, M., Nilsson, E., Peetre, C., Thysell, H., Grubb, A. and Adolf. G., Eur. J. Haematol. 1989. 42: 270. 16 Englemann, H., Aderka, D., Rubinstein, M., Rotman, D. and Wallach, D., J. Biol. Chem. 1989.264: 11974. 17 Engelmann, H., Novick, D. and Wallach, D., J. Biol. Chem. 1990. 265: 1531. 18 Loetscher, H. R., Brockhaus, M., DembiC, Z., Gentz, R., Gubler, U., Hohmann, H. F!, Lahm, H. W.,Van Loon, A. F! G. M., Pan,Y-C. E., Schlaeger, E. J., Steinmetz, M.,Bbuchi, H. and Lesslauer, W., in McLean, N. (Ed.), Oxford Survey on Eucaryotic Genes, vol. 7, Oxford University Press, Oxford 1991, in press.
2886
W. Lesslauer, H. Bbuchi, R. Gentz et al.
19 Traunecker, A., Schneider, J., Kiefer, H. and Karjalainen, K., Nature 1989. 339: 68. 20 Loetscher, H. R., Gem, R., Zulauf, M., Lustig, A., Tabuchi, H., Schlaeger, E. J., Brockhaus, M., Gallati, H., Manneberg, M. and Lesslauer, W., . I Biol. . Chem. 1991, in press. 21 Laemmli, U. K., Nature 1970. 227: 680.
Eur. J. h u n o l . 1991.21: 2883-2886 22 Lehmann,V, Freudenberg, M. A. and Galanos, C., J. Exp. Med. 1987. 165: 657. 23 Rothstein, J. L. and Schreiber, H., Proc. Nutl. Acud. Sci. USA 1988. 85: 607. 24 Grau, G. E.,Vesen, D., De Groote, D., De la Croux, D., Gisler, C., Piguet, F! F. and Lambert, I? H,, Clin.Exp. Zmmunol. 1991. 84: 411.
