Differential activity of interleukin 1α and interleukin 1β in the stimulation of the immune responsein vivo

  Eur. J. Immunol. 1990.20: 317-321 Diana BoraschiO, Luigi Villao, Gianfranco Volpinio, Paola BossUO, Stefan0 CensiniO, Paolo GhiaraO, Giuseppe Scapigliato, Lucian0 NencioniO, Marcella BartaliniO, Giacomo MatteucciO, Franca CioliO, Marzia CarnascialiO, Enza OlmastroniO, Manuela Mengozzia, Pietro Ghezzia and Aldo TagliabueO Research CenterD and In Vivo Quality Control Centero, Sclavo, Siena and Mario Negri Institute*, Milan Comparison of biological activities of interleukin la and interleukin lp 317 Differential activity of interleukin la and interleukin lp in the stimulation of the immune response in zyxw iva The biological activities of human recombinant interleukin (IL) la and IL 1P were compared in different biological systems. The two IL 1 orms were equally active zyxwvut n vitro in inducing proliferation of murine thymocytes and of the murine T helper clone D10.G4.1, and in triggering release of prostaglandin E2 from human skin fibroblasts. zyxwvu n vivo IL la and IL 1P were similarly pyrogenic both in rabbits and mice, and could equally increase the circulating levels of the acute phase protein serum amyloid A in mice. However, only IL lp showed immunostimu- latory activity in vivo as it could enhance the number of specific antibody- producing cells in the spleen of mice immunized with either a T-dependent or a T-independent antigen. Although devoid of immunostimulatory activity, IL la could efficiently compete immunostimulation induced by IL lp, suggesting an effective interaction with the IL 1 receptor. Thus, IL 1P appears to have an important role in the positive regulation of immune responses, while IL la may act as down-regulator of the IL 1P effect. 1 Introduction Interleukin 1 (IL 1) is a family of protein molecules involved in the regulation of a large number of metabolic and immunological defense reactions [l, 21. At least two IL 1 proteins (IL la and IL 1p) have been identified in humans, rodents and other mammalians [MI L la and IL lp share low amino acid homology, but they appear to bind to the same cellular receptor and are very similar in their gene organization and in the array of biological activities [7-lo]. In fact, direct comparison of the biological effects of the two IL 1 orms, either fromnatural or recombinant sources, did not show any clear difference in a variety of in vitro assays [ll-191, as well as in vivo [20]. It is thus believed that the two proteins may derive from genes which duplicated and evolved independently without selective pressure for their distinct maintenance [3, 211 and that the main difference between them may reside in the differential control of gene expression and in different stability and catabolism of the proteins [9]. In the present work, we have addressed the issue of functional distinction between IL la and IL lp proteins by comparing their activities both in vitro and in vivo in assays which could be representative of their immunological and inflammatory roles. Although no difference between the [I 76031 This project was supported by the contract Programma Nazio- nale di Ricerca per la Chimica - Tema 3, Linfochine e Vaccini Sintetici , granted to Sclavo by the Italian Ministry of Scientific Research and Technology. Correspondence: Diana Boraschi, Laboratory of Immunopharma- cology, Sclavo Research Center, Via Fiorentina 1 53100 Siena, Italy Abbreviations: SAA: Serum amyloid A SIII: Pneumococcal polysaccharide type In two IL 1 proteins could be seen in vitro or in fever induction in vivo our data suggest that IL 1P is far more potent than IL la in the stimulation of immune responses in vivo and that IL la may serve as negative regulator of IL 1P-induced immunostimulation. 2 Materials and methods 2.1 Animals C3H/HeN Crl BR mice were obtained from Charles River Italia (Calco, Italy). Endotoxin-resistant C3H/HeJ mice were obtained from The Jackson-Laboratory (Bar Harbor, ME). Mice of both sexes were used between 6 and 10 weeks of age. NZW rabbits of both sexes (Martelli, Molinella, Italy) were used for pyrogenicity tests between 1.8 and 2.3 kg of body weight. 2.2 IL 1 preparations Human rIL la was obtained from Biogen (Geneva, Swit- zerland). In comparative experiments, another human (h) rIL la preparation (kind gift of Dr. I? T. Lomedico, Hoffman-La Roche, Nutley, NJ) was used with similar results. hIL 1P mature fragment 117-269 (Sclavo Research Center, Siena; and DE.BI. Division, Cassina de'Pecchi, Italy) was expressed in E. coli and purified as described by Wingfield et al. [22]. Other hrIL 1p preparations were obtained by Biogen and by Genzyme (Boston, MA). All IL 1 preparations were free of LPS contamination at the concentrations used (LAL chromogenic assay; M.A. Bio- products, Walkersville, MD). 2.3 Murine thymocyte proliferation assay Thymocytes of 6- to 8-wk-old male C3H/HeJ mice were cultured at 6 x lo5 cells/well of Cluster% plates (Costar, Cambridge, MA) for 72 h in culture medium RPh4I 1640; Gibco-Europe, Paisley, Scotland; containing 2 mh4 L- O VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990 OO14-2980/90/02O2-03 17$02.50/0  318 zyxwvusrqpo . Boraschi, L.Villa, G.Volpini et al. Eur. J. Immunol. 1990.20: 317-321 glutamine, 25 mM Hepes buffer, and 50 pg/ml gentamycin sulfate; Sigma Chemical Co., St. Louis, MO) supple- mented with 5% heat-inactivated FBS (HyClone, Sterile Systems, Logan, UT), 1.25 zyxwvuts   M 2-ME, and 1.5 pg/ml purified PHA (Wellcome, Beckenham, GB) [23]. IL 1 was added to thymocytes at the beginning of culture. Thymo- cyte proliferation was assessed as radioactivity incorpora- tion after an overnight pulse with 0.5 pCi/well = 18.5 kBq/well L3H]dThd (sp. act., 185 GBq/mmol; Amersham International, Amersham, GB). 2.7 Determination of serum amyloid A SAA) C3H/HeJ female mice were bled 6 to 8 h after receiving a single i.v. inoculum of IL 1. SAA was evaluated in serum samples by solid-phase competitive RIA [28]. Briefly, 25-pl serum samples were heated at 60 C in tightly capped tubes for 1 h, then diluted with 1 ml casein barbital buffer. Triplicate 200-pl aliquots of the diluted samples were assayed by RIA. SAA were expressed as micrograms of amyloid A equivalendml. 2.4 D10.G4.1 proliferation assay 2.8 Assays for zyxw n zyxw ivo immunostimulation The conalbumin-specific murine T helper clone D10.G4.1 [24] was kindly provided by Dr. J. H. Bertoglio (Institut Gustave Roussy, Villejuif, France). Cells were maintained in culture with IL2 and periodically stimulated with mitomycin C-treated spleen cells of H-2b haplotype. As described in detail elsewhere [25], two weeks after stimu- lation cells were frozen and stored in liquid nitrogen. On the day of assay, cells were thawed and plated at 2 x lo4 cells/well of Cluster96 plates in culture medium with 10% FBS and 6 x lop5 M 2-ME. Cells were incubated for 48 h with 2.5 pg/ml Con A (type IV S; Sigma) and IL 1 dilutions. Proliferation was assessed as radioactivity incorporation after an overnight pulse with 0.5 pCi/well [3H]dThd. 2.5 Prostaglandin Ez PGE2) production by dermal fibroblasts PGE2 production by fibroblasts was assayed as described by Dayer et al. [26]. Briefly, human foreskin fibroblasts F7100 (kindly provided by L. Stanghellini, Interferon Production Unit, Sclavo, Italy) at early passages were seeded at 2 x lo /well of Cluster96 plates in culture medium containing 10% FBS. After 48 h at 37 C, SN was removed, and IL 1 was added for 72 h. PGE2 determinations on culture SN were performed by RIA (Du Pont - NEN, Boston, MA). Determination of numbers of residual fibroblasts at the end of incubation did not reveal any toxic effect by IL 1. 2.6 Pyrogenicity assays Healthy adult NZW rabbits, restrained and then rested for 2 h to stabilize basal temperature, received a single i.v. administration of pyrogen-free saline (1 mYkg body weight) containing IL 1. Rectal temperature was recorded with a thermocouple thermometer (Ellab Instruments, Copen- hagen, Denmark) every 15 min up to 3 h after IL 1 inoculum. Temperature increases (ATC ) were evaluated as the difference between experimental and preinjection temperature values of each animal. At least three rabbits per experimental group were assayed within each experi- ment. Data reported are the mean k SEM of peak temperature increases (generally achieved between 30 and 60 min after zyxwvuts L 1 inoculum) assessed in 14 experiments. Male rabbits were routinely used, but the use of female rabbits yielded comparable results. Induction of fever by IL 1 in C3H/HeJ female mice was measured as previously described [23,27]. Data are reported as mean ATC at peak temperature increase (usually 20 min after i.v. inoculum of IL 1) zyxwvuts   SEM of 2 to 5 animals per experimental group. Determination of specific antibody-producing cells (PFC) in the spleen of mice was performed as previously described in detail [29]. Briefly, mice received a single antigen inoculum (SRBC, 1 x lo8 to 2 x l@/mouse i.v.; or pneumococcal polysaccharide type 111, SIII, 5 pglmouse i.p.). At the same time of antigen administration, mice received IL 1 .v. The number of specific PFC in the spleen of mice was determined either 4 days (response to SRBC) or 5 days (response to SIII) later by the Cunningham and Szenberg slide technique [30]. Male C3H/HeN mice were routinely used for the in vivo immunostimulation experi- ments.The use of female mice or of LPS-resistant C3H/HeJ mice has yielded comparable results (data not shown). 2.9 Statistical analysis Results were calculated as mean values zk SEM from replicate experiments or as mean EM of triplicate to quintuplicate samples within a representative experiment. SEM < 10% are not reported. Statistical significance was calculated by analysis of variance or by parallel line assay [31] after logarithmic transformation of the variables. 3 Results The biological activities of hrIL la and hrJL 18 have been assayed in vitro in three different experimental systems. The effect on proliferation of T cells has been tested both on normal murine thymocytes (Fig. la) and on theTh clone D10.G4.1 (Fig. lb).The activity of the two IL 1 forms was superimposable in both assays, with half maximal stimula- tion achieved at concentrations around 10 pM on thymo- cytes and 0.3 PM on D10.G4.1 cells. Similarly, IL la and IL lp were equally active in inducing PGE2 release from human dermal fibroblasts F7100, reaching half maximal stimulation at about 3 to 6 pM(Fig. lc). In vivo, the pyrogenic activity of IL la and IL 18 was tested in rabbits and again found to be comparable (Fig. 2a). The fever-inducing capacity of IL 1 was also assayed in the LPS-resistant C3H/HeJ mouse model and again no signif- icant difference between IL la and IL lp could be shown (Fig. 2b). Similarly, IL la and IL lp were comparably able to induce the appearance of the acute-phase protein SAAin the mouse bloodstream (Fig. 2c). A striking difference in activity between IL la and IL 10 was evident when their immunostimulatory capacity in vivo in the mouse was assessed (Fig. 3). In fact, IL lp could  Eur. J. Immunol. 1990.20: 317-321 Comparison of biological activities of interleukin la and interleukin 10 319 1 13-11 -11 -10 -18 -14 -10 zyxwvutsrqpon 11 11 -10 zyxwvutsrqpo IL zyxwvu   log10 M) zyxwvut igure 1. Comparison of biological activities of IL la and IL 1p n vitro. Proliferation of C3HMeJ thymocytes (a) or of the murineT helper clone D10.G4.1 (b), or PGE2 production by human dermal fibroblasts F7100 (c) was assessed in the presence of increasing concentrations of hrIL la M), hrIL If5 zyxwvu 0) or in the absence of added lymphokine zyxwvutsrqp *). Data are the mean of 10 o 15 ndependent experiments. Proliferation of thymocytes to PHA in the absence of IL 1 was 3366 f 579 cpm in 15 experiments. Proliferation of DlO.G4.ltoConAintheabsenceofIL 1was4944k 630cpmin 11 experiments. Production of PGE2 by F7100 fibroblasts in the absence of IL 1 was 1.61 f 0.40 ng/mg fibroblast protein in 10 experiments. Statistical analysis: IL la vs. IL lp, always not significant (NS). maximally enhance the primary response to the T-depend- ent antigen SRBC when administered i.v. together with the antigen at doses as low as 6 fmoykg (Fig. 3a). In contrast, no significant increase of the response could be observed with IL la even at doses as high as 0.1 nmolkg, although a tendency of increase could be seen at the highest doses. Similarly, in a preliminary experiment IL lp could optimal- ly enhance the immune response to SRBC when adminis- tered i.p. with the antigen at 60 fmol/kg, whereas IL la could show a weak effect only at 0.6 nmoVkg (data not shown). Accordingly, IL lp was also able to maximally increase the immune response to theT-independent antigen SIII at 3-30 fmoykg, whereas IL la did not show any significant effect, either at 30 fmoYkg or at doses up to 1000-fold higher (Fig. 3b). . -13 -12 -11 -10 -12 -11 -10 -13 -12 -11 -10 9 IL 1 log,, mol/kg) Figure 2. Pyrogenicity and SAA induction by IL la and IL lp. Temperature increase in rabbits (a) and in mice (b) and SAA induction in mice (c) was measured after i.v. administration of saline alone (*) or containing increasing doses of hrIL la M) or hrIL lp 0). ata are the mean SEM of 3 to 27 rabbits and 2 to 5 mice for each experimental group. Mean basal temperature before IL 1 noculum was 39.2 f .02 zyxwvutsrq for 77 rabbits and 37.5 0.08 C 3 220 2 8 zyx 9 8 160 W rn .- a E 100 -16 -14 -12 -10 -16 -14 -12 -10 IL 1 log,, mollkg 1 Figure 3. Immunostimulatory activity of IL la and IL 10 in vivo. Stimulation of the number of specific PFC/spleen of mice immu- nized with SRBC (a) or with SIII (b) was assessed upon i.v. administration of increasing doses of hrIL la 1) r rhIL If5 0). Data are the mean of values of 2 to 20 separate experiments. Mean PFC anti-SRBClspleen of control mice receiving the antigen alone was 40 118 f 1096 n 39 experiments. Mean PFC anti-SIWspleen of control mice was 1353 rt 424 in 6 experiments. Statistical analysis: IL lp vs. IL la, p < 0.01. Shaded areas represent the range 2 SD above and below the mean) of control responses. Since IL la and IL 1p have been reported to bind to the same receptor [7, 81 the possibility that inactive IL la could compete with active IL lp in the assay of immuno- stimulation in vivo was investigated. Indeed, although unable to stimulate the immune response by itself, IL la could effectively inhibit the immunostimulatory activity of IL lp, with optimal inhibition achieved at a molar ratio IL la : L 1p of 1: 1 (Fig. 4). Molar ratio IL la : L lp ( o -2 -1 0 1 -17 -16 -15 -14 -13 IL lac (log,, mol kg Figure 4. Competition of IL la with the immunostimulatory activ- ity of IL 1p. Mice were immunized with SRBC alone I) r in the presence of hrIL 18 5.7 fmol/kg, i.v.; O , f increasing doses of hrIL la M), or of IL lp admixed to increasing doses of IL la El). Data are the mean of values obtained in four separate experiments. Mean PFC anti-SRBC/spleen of control mice was 45 502 f 3815. Statistical significance: IL 1p vs control, p < 0.01; IL la vs. control. NS: IL 1B + IL la vs. IL 1p alone, p < 0.01. or 23 mice. Statistical analysis: IL la vs. IL lp, always NS. I  320 D. Boraschi, L.Villa, G.Volpini et al. Eur. J. Immunol. 1990.20: 317-321 other cytokines such asTNFa [44] and IL 6 (J.Van Damme and D. Boraschi, unpublished observations).These factors, which can be directly immunostimulatory if present at sufficient concentrations [45,46], at suboptimal doses may synergize with residual IL 10 and produce a significant biological effect, as already demonstrated in zy n vitro systems [47-501. Since the difference between ILla and ILlp in the immunostimulatory activity in vivo was evident also for responses independent of the presence of Th cells (such as that to the pneumococcal polysaccharide SIII) [52], it is possible that the activity of IL lp could be mainly directed to B cells. Indeed, B cells appear to possess an IL 1R different from that described for T lymphocytes, to which IL lp binds more abundantly but with lower affinity than IL la [40-421. It is thus proposed that IL 1p may be the major IL 1 orm involved in immunostimulation in vivo and that IL la may act as negative regulator of IL lp activity, possibly by direct competion for receptor occupancy. On the other hand, IL la and IL 1fJ showed comparable inflammation-related activities in vivo in the mouse (pyro- genicity, SAA induction) as well as in the rabbit (pyroge- nicity) and in vitro on human cells (PGE2 production by skin fibroblasts). It should be noted that the doses of z L lp necessary for induction of inflammatory responses in vivo are much higher (103 to 104-fold) than those necessary for immunostimulation. It is therefore possible that IL lp and IL la could act as physiological regulators of the immune response when locally present at very low concentrations. In pathological situations where IL 1 production is increased, both IL 1 forms would become important for triggering the inflammatory response while their role in the immune response regulation is possibly less relevant. z 4 Discussion IL 1 appears to play a central role in the stimulation and regulation of immune responses and in the onset and development of inflammatory reactions [l, 21. The multi- plicity of IL 1 activities and the notion of the existence of two different, though distantly related, IL 1 proteins has led to the hypothesis that IL la and IL lp may have distinct biological roles. In fact, some differences in activity between IL la and IL 1p have been described. IL la appears to be more active than IL 10 in inducing PAF release [13] and in inhibiting proliferation of human endothelial cells [32]. Conversely, IL lp is more effective than IL la in inducing ACTH release in vitro and in vivo in the rat [33-351, in inhibiting insulin release from rat pancreatic islets [36] and glycosaminoglycan synthesis in rat chondrocytes [37], in inducing al-acid glycoprotein in rat hepatoma cells [38], and as hyperalgesic agent in vivo in rats [39]. Although in these studies the IL 1 preparations used were always carefully compared, only in a few cases was the difference in efficacy between IL la and IL lp striking [13, 32-35, 38, 391, and the possibility still exists that some of these differences might be due to the animal species used e.g. IL 1p appears to be generally more active than IL la in the rat or on rat cells) [33-391. In fact, a large body of experimental evidence indicates that the two IL 1 forms generally share most of their biological activities [ll-201 and apparently bind to a common receptor [7,8].Thus, it is still unclear whether the two IL 1 forms may play distinct roles in the regulation of biological responses. The study reported here suggests that IL lp might be a selective regulator of the immune response in vivo in the mouse. In fact, IL 16 could significantly enhance the number of specific PFC in the spleen of mice immunized with either a T-dependent SRBC) or a T-independent (SIII) antigen. In contrast, IL la did not reproducibly show any significant effect when administered at doses up to 104-fold higher than those necessary for IL lp to achieve maximal immunostimulation. The fact that in vitro IL la and IL lp were equally active in stimulating proliferation of murineTcells (both thymocytes and aTh clone) may suggest that a differential clearance of the two IL 1 forms after i.v. administration could be responsible for their different immunostimulatory capacity. Although careful pharmaco- kinetics studies are necessary to verify this hypothesis, data presented here indicate that IL la, although unable to trigger an antigen-specific immune response, is nonet heless able to inhibit the immunostimulatory activity of IL 1p at ratios (0.1 zyxwvutsrq   1 1 1) which suggests direct competition for receptor binding. It could be thus hypothesized that IL la can effectively bind to immunocompetent cells in vivo but that receptor occupancy is not a sufficient event for triggering of the immune response. It should be noted that the inhibitory effect of IL la on the IL lp activity was significantly and reproducibly reduced by increasing the IL la to IL 16 ratio (Fig. 4, and data not shown). Different hypotheses might be formulated to explain this finding, including altered receptor affinity or kinetics of receptor binding by increasing ligand concentra- tion [7] and heterogeneity in the receptor binding sites and affinities for IL la and IL lp [40-431. Furthermore, higher doses of IL la, although unable to stimulate the immune response by itself, may trigger the production in vivo of The authors are grateful to Dr. Albert0 Mantovani Mario Negri Institute, Milan, Italy) for helpful suggestions during the prepara- tion of this work, to Dr. Jean D. Sipe Arthritis Center, Boston University School of Medicine, Boston, MA) for the kind gift of anti-SAA antibody, to Catherine Mallia for secretarial assistance, and to Giorgio Corsi for artwork. Received April 28, 1989; in revised form September 12, 1989. 5 References 1 Dinarello, C. A., Rev. Infect. Dis. 1984. 6: 51. 2 Oppenheim, J. J., Kovacs, E. J., Matsushima, K. and Durum S. K., Immunol. Today 1986. 7: 45. 3 Lomedico, P T., Gubler, U. and Mizel, S. B., Lymphokines 1987. 13: 139. 4 March, C. J., Mosley, B., Larsen, A., Cerretti, D. I? Braedt, G., Price, V. Gillis, S., Henney, C. S., Kronheim, S. R., Grabstein, K., Conlon, F J., Hopp, T. F and Cosman, D., Nature 1985. 3I5: 641. 5 Saklatvala, J., Sarsfield, S. J. and Townsend,Y., z   Exp. Med. 1985. 162: 1208. 6 Maliszewski, C. R., Baker, F E., Schoenborn, M. A., Davis, B. S., Cosman, D., Gillis, S. and Cerretti, D. F , Mol. Immunol 1988. 25: 429. 7 Dower, S. K. and Urdal, D. L., Immunol. Today 1987. 8: 4 6 8 Sims, J. E., March, C. J., Cosman, D., Widmer, M. B., MacDonald, H. R., McMahan, C. J., Grubin, C. E.,Wignall, J. M., Jackson, J. L., Call, S. M., Friend, D., Alpert, A. R., Gillis, S., Urdal, D. L. and Dower, S. K., Science 1988. 241: 585.  Eur. J. Immunol. 1990. 20: 317-321 Comparison of biological activities of interleukin la and interleukin lp 321 9 Auron, P. E. and Webb, A. C., Lymphokines 1987. 14: 33. 10 Rupp, E. A., Cameron, P. M., Ranawat, C. S., Schmidt, J. A. and Kahn Bayne, E., J. Clin. Invest. 1986. 78: 836. 11 Wood, D. D., Kahn Bayne, E., Goldring, M. B., Gowen, M., Hamerman, D., Humes, J. L., Ihrie, E. J., Lipsky, P. E. and Staruch, M.-J., J. Immunol. 1985. 134: 895. 12 Lowenthal, J. W., Cerottini, J.-C. and MacDonald, H. R., J. Immunol. 1986. 137: 1226. 13 Dejana, E., Breviario, F., Erroi, A., Bussolino, E, Mussoni, L., Gramse, M., Pintucci, G., Casali, B., Dinarello, C. A.,Van Damme, J and Mantovani, A., Blood 1987. 69: 695. 14 Tsai, S.-C. J. and Gaffney, E.V., J. Natl. Cancer Int. 1987. 79: 77. 15 Sakai, K., Hattori,T., Matsuoka, M., Asou, N.,Yamamoto, S., Sagawa, K. and Takatsuki, K., zyxwvut   Exp. Med. 1987. 166: 1597. 16 Pober, J. S., Lapierre, L. A., Stolpen, A. H., Brock,T. A., Springer, T. A., Fiers,W., Bevilacqua, M. A., Mendrick, D. L. and Gimbrone, Jr., M. A., J. Immunol. 1987. 138: 3319. 17 Elias, J. A., Gustilo, K., Baeder, Wand Freundlich, B., J. Immunol. 1987. 138: 3812. 18 VanDamme, J., Opdenakker, G., Simpson, R. J., Rubira, M. R., Cayphas, S. ,Vink, A., Billiau, A. and Van Snick, J., J Exp. Med. 1987. 165: 914. 19 Burch, R. M., Connor, J. R. and Axelrod, J., Proc. Natl. Acad. Sci. USA 1988. 85: 6306. 20 Sabatini, M., Boyce, B., Aufdemorte, T., Bonewald, L. and Mundy, G. R., Proc. Natl. Acad. Sci. USA 1988. 85: 5235. 21 Clark, B. D., Collins, K. L., Gandy, M. S., Webb, A. C. and Auron, P E., Nucleic Acids Res. 1986. 14: 7897. 22 Wingfield, P., Payton, M. ,Tavernier, J., Barnes, M., Shaw, A., Rose, K., Simona, M. G., Demczuk, S.,Williamson, K. and Dayer, J.-M., Eur. J. Biochem. 1986. 160: 491. 23 Antoni, G., Presentini, R., Perin, F.,Tagliabue, A., Ghiara, P., Censini, S. ,Volpini, G. ,Villa, L. and Boraschi, D., J. Immunol. 1986. 137: 3201. 24 Kaye, J., Gillis, S., Mizel, S. B., Shevach, E. M., Malek,T. R., Dinarello, C. A., Lachman, L. B. and Janeway, Jr., C. A., J. Immunol. 1984. 133: 1339. 25 Bertoglio, J. H. and Leroux, E., J. Immunol. 1988. 141: 1191. 26 Dayer, J.-M., Zavadil-Grob, C., Ucla, C. and Mach, B., Eur. J. Immunol. 1984. 14: 898. 27 Boraschi, D., Nencioni, L., Villa, L., Censini, S., Bossii, P., Ghiara, P., Presentini, R., Perin, F., Frasca, D., Doria, G., Forni, G., Musso, T., Giovarelli, M., Ghezzi, P., Bertini, R., Besedovsky, H. O., Del Rey, A., Sipe, J. D., Antoni, G., Silvestri, S. and Tagliabue, A., J. Exp. Med. 1988. zyxwvut 68: 675. 28 Sipe, J. D., Ignaczak,T. F., Pollock, zyxwvut . S. and Glenner, G. G., J. Immunol. 1976. 6: 1151. 29 Nencioni, L.,Villa, L., Tagliabue, A., Antoni, G., Presentini, R., Perin, E, Silvestri, S. and Boraschi, D., J. Immunol. 1987. 139: 800. 30 Cunningham, A. J. and Szenberg, A., Immunology 1968. 14: 599. 31 Finney, D. J., Statistical Method in Biological Assay, 2nd Edit., Griffin, London 1971, p. 99. 32 Norioka, K., Hara, M., Kitani, A., Hirose, T., Hirose, W., Harigai, M., Suzuki, K., Kawakami, M.,Tabata,T., Kawagoe, M. and Nakamura, H., Biochem. Biophys. Res. Commun. 1987. 145: 969. 33 Uehara, A., Gillis, S. and Arimura, A., Neuroendocrinology 1987. 45: 343. 34 Uehara, A., Gottschall, P E., Dahl, R. R. and Arimura, A., Biochem. Biophys. Res. Commun. 1987. 146: 1286. 35 Nakamura, H., Motoyoshi, S. and Kadokawa, T., Eur. J. Pharmacol. 1988. 151: 67. 36 Mandrup-Poulsen, T., Bendtzen, K., Dinarello, C. A. and Nerup, J., J. Immunol. 1987. 139: 4077. 37 Ikebe, T., Hirata, M. and Koga, T., J. Immunol. 1988. 140: 827. 38 Geiger,T., Andus,T., Klapproth, J., Northoff, H. andHeinrich, P. C., J. Biol. Chem. 1988. 263: 7141. 39 Ferreira, S. H., Lorenzetti, B. B., Bristow, A. F. and Poole, S., Nature 1988. 334: 698. 4 Matsushima, K., Akahoshi, T.,Yamada, M., Furutani,Y. and Oppenheim, J. J., J. Immunol. 1986. 136: 4496. 41 Horuk, R., Huang, J. J., Covington, M. and Newton, R. C., J. Biol. Chem. 1987. 262: 16275. 42 Scapigliati, G., Ghiara, P., Bartalini, M., Tagliabue, A. and Boraschi, D., FEBS Lett. 1989. 243: 394. 43 Dinarello, C. A., Clark, B. D., Puren, A. J., Savage, N. and Rosoff, P M., Immunol. Today 1989. 10: 49. 44 Philip, R. and Epstein, L. B., Nature 1986. 323: 86. 45 Ghiara, P., Boraschi, D., Nencioni, L., Ghezzi, P. and Taglia- bue, A., J. Immunol. 1987. 139: 3676. 46 Takatsuki, F., Okano, A., Suzuki, C., Chieda, R.,Takahara,Y., Hirano, T., Kishimoto, T., Hamuro, J. and Akiyama, Y, J Immunol. 1988. 141: 3072. 47 Hurme, M., Eur, J. Immunol. 1988. 8: 1303. 48 Ranges, G. E., Zlotnik, A., Espevik, T., Dinarello, C. A., Cerami, A. and Palladino, M. A., J. Exp. Med. 1988. 167: 1472. 49 Houssiau, F. A., Coulie, P. G., Olive, D. and Van Snick, J., Eur. J. Immunol. 1988. 8: 653. 50 Helle, M., Brakenhoff, J. P J., De Groot, E. R. and Aarden, L. A., Eur. J. Immunol. 1988. 8: 957. 51 Howard, J. G., Christie, G. H., Courtenay, B. M., Leuchars, E. and Davies, A. J. S., Cell Immunol. 1971. 2: 614.