Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Two types of mouse T helper cell. IV. Th2 clones secrete a factor that inhibits cytokine production by Th1 clones

Two types of mouse T helper cell. IV. Th2 clones secrete a factor that inhibits cytokine... TWO TYPES T HELPER CELL OF MOUSE IV. Th2 Clones Secrete a Factor that Inhibits Cytokine Production by Thl Clones BY DAVID F. FIORENTINO, MARTHA W BOND, AND TIM R . MOSMANN` From the Department Immunology, DNAX Research Institute, Palo Alto California 94304 of Many long-term mouse Th cell clones can be divided into two types, based on the pattern of cytokines secreted in response to antigen or lectin stimulation . Thl clones secrete IL-2, IFN-y, (LT),' whereas Th2 clones express IL- and lymphotoxin 4, IL-5, function, P600 (1-4) . Both clone se- IL-6, and a gene of unknown types of IL-3, TNF preproenkephalin, crete granulocyte-macrophage CSF (GM-CSF), a, The between Thl and Th2 and several other induction-specific proteins. differences clones cytokine synthesis patterns lead to markedly different functions (5-8) . Th2 helpers (8, can provide are normally better B cell 9), while Thl clones, although they (7, 11), B cell help in some circumstances 10, preferentially induce macrophage acti- vation (12) and delayed-type hypersensitivity (DTH) (6) . Recent evidence (Street, H. T H . Bass, D. F Fiorentino, A . Leverah, N. E., J . Schumacher, T A. Fong, J. (13-15) and T. R . Mosmann, manuscript in preparation) suggests that other differen- Thl repre- tiation states of mouse Th cells exist, and that the and Th2 phenotypes sent activated effector cells . The existence of these two major types ofTh cell may explain the well-documented separation of immune responses into DTH or antibody production, which often ap- pear to be mutually exclusive (16) . If B cell help and DTH are predominantly medi- ated by Th2 and Thl cells, respectively, this suggests that Thl and Th2 cells may be mutually inhibitory (17, 18) . The Thl product IFN-y inhibits proliferation of Th2 clones in vitro (19, 20), while an unknown product of a Th2 clone may inhibit the proliferation of Thl clones (21) . In addition to crossinhibition of proliferation, we considered the possibility that inhibitors of Thl and Th2 effector function also exist . Here, we report the preliminary characterization of a cytokine produced by Th2 clones that inhibits the synthesis of several cytokines by Thl clones . This cytokine, which we have called cytokine synthesis inhibitory factor (CSIF), has been distin- guished from the other known Th2 cytokines . Address correspondence to Tim R. Mosmann, Department ofImmunology, DNAX Research Institute, . . 901 California Avenue, Palo Alto, CA 94304 Dr Mosmann's address afterJanuary 1, 1990 will be the Department ofImmunology, 865 Medical Services Building, University of Alberta, Edmonton, Alberta T6G 2H7, Canada. I paper . synthesis DTH, Abbreviations used in this CSIF, cytokine inhibitory factor; delayed-type hyper- sensitivity; GM-CSF, granulocyte-macrophage CSF; ppCSIF, partially purified CSIF; TGR0, trans- forming growth factor S. J . EXP. MED. ® The Rockefeller University Press - 0022-1007/89/12/2081/15 $2 .00 Volume 170 December 1989 2081-2095 Th2-DERIVED INHIBITOR 2082 OF Thl CYTOKINE SYNTHESIS Materials and Methods Mice. BALB/c, C57BL/6, and CBA/J mice were obtained from TheJackson Laboratory (Bar Harbor, ME) and the Institute of Medical Research (SanJose, CA) . Female mice 8-12 wk old were used in all experiments . Cell Lines. Thl clones included HDK-1, : BALB/c anti-KLH (2) ; MD13-2, BALB/c anti- CRBC (1) ; MD13-10, BALB/c anti-CRBC (1) ; GK15-1, CBA/J anti-CRBC (11) ; LB2-1, C57BL/6 anti-CRBC (1) ; ML3-A-Cl-2-Ba, C57BL/6 anti-BALB/c ; and ML3-A-C2-6-Nb, C57BL/6 anti-BALB/c . Th2 clones included : D10.G4 .1 (D10), AKR/J anti-conalbumin, ob- tained from C . Janeway (Yale University, New Haven, CT) (22) ; CDC25 and CDC35, C3D2 anti-rabbit obtained from D. IgG, Parker (University of Massachusetts, Worcester, MA) (23) ; D9, C57BL/6, G . ; obtained from Nabel (Harvard University, Boston, MA) (24) MB2-1, C57BL/6 (1) ; and M411-2 and M411-6, BALB/c anti-CBA/J . The maintenance and stimula- tion of antigen-specific T cell lines have been described previously (1, 2). The mouse T cell HT2 line (25) was obtained from S . Strober (Stanford University, Stanford, CA) and the WEHI .13 164 line (26) from M . Palladino (Genentech, South San Francisco, CA). Media . Assay medium consisted of RPMI 1640 (J . R . Scientific Inc ., Woodland, CA) R. with 10 17o FCS (J . Scientific Inc .), 0 .05mM 2-ME (Sigma Chemical Co., St. Louis, MO), and 20 mM Hepes (Gibco Laboratories, Grand Island, NY). T cell growth medium con- sisted assay of medium containing 330 U/ml mouse rIL-2 . . IL-2 Antibodies mAbs against (S4B6), IL-5 (TRFK4 and TRFK5), and IFN (XMG1 .2) -'Y have been described previously (1, 2, 27) . The anti-IL-3 mAbs 8F8 .11 and 43D.11 (28), as well as the anti-GM-CSF antibodies 22E9 .11 and 35E10 .11, were obtained fromJ . S. Abrams (DNAX), I1B11 W. the anti-IL-4 antibody (29) was from E . Paul (National Institutes ofHealth, Bethesda, MD), and the 6B4 anti-IL-6 antibody (30) fromJ . van Snick (Ludwig Cancer In- stitute, Brussels, Belgium) . The hamster anti-mouse T3 mAb 145-2CII (31) was obtained . H .) . fromJ Bluestone (N . I . Polyclonal rabbit IgG antibodies specific for transforming growth factor S (TGFfl) were obtained from R & D Systems (Minneapolis, MN), and control rabbit IgG antibodies were obtained from R . Coffman (DNAX) . Cell Supernatants. Antigen-induced T cell supernatants were prepared by incubating x x T cell clones (2 105 cells/ml) with irradiated APC (5 10 6 cells/ml) and antigen in assay medium . Serum-free Con A-induced cell supernatants were prepared by incubating T cell clones (5 x 10 6 cells/ml) with Con A (5 Wg/ml ; Sigma Chemical Co.) in RPMI 1640 lacking phenol red (Gibco Laboratories) and containing 0 .05 mM 2-ME and 20 mM Hepes for 24 h . Cytokines. TGF was obtained from R & D Systems . Purified mouse P40 (32) was gener- ously provided by J . van Snick (Ludwig Cancer Institute) . Purified mouse IL-5 (33) was depleted ofknown Th2 cytokines and providedby R. Coffman, DNAX . Purified recombinant mouse cytokines were provided by: M . Howard, DNAX (IL-4) ; R . Kastelein, DNAX (IL-4); F. Lee, DNAX (IL-6 and IL-7) ; J . Schreurs, DNAX (IL-3 and GM-CSF) ; G . Zurawski, DNAX (human IL-Ia); Schering Research, Bloomfield, NJ (IL-2 and IFN-y) ; and M . Pal- ladino, Genentech (TNFa) . Cytokine Bioassays . The MTT assay (34), as modified (1), was used to measure IL-2, IL-4, and LT/TNFa using mAbs to establish monospecificity for IL-2 and IL-4 assays (2, 26) . WEHI .13 LT/TNFa was assayed on 164 cells essentially as described by Espevik and Nissen- x assay was carried Meyer (26), except that 5 103 target cells/well were used, and the out were calculated in the presence ofsaturating (>2 ng/ml) amounts ofIFN-y . Units of activity (1) . as described previously . Two-site sandwich ELISAs for IL-3, IL-5, and IFN-y were carried out Cytokine ELISAs as described previously (2, 27, 28) . The ELISA for GM-CSF was carried out essentially as mAbs and 22E9 . Intracellular IFN-y levels were described for the IL-5 assay, using the 35E10 1% X-100 Chemical Co.) and removing nuclei measured after lysing cells in Triton (Sigma centrifugation . by Column Chromatography. 1-2 .5-liter batches of serum-free Con A-induced D10 superna- tant were concentrated N10-fold using YM-5 membranes (Amicon Corp., Danvers, MA), passed through a 5-ml mannose-conjugated agarose column (E-Y Laboratories, San Mateo, FIORENTINO ET AL . 2083 CA), then further concentrated another three- to fivefold, for a total concentration of30-50- fold . This material was fractionated by HPLC on a hydroxylapatite-based support (Bio-Gel HPHT, Bio-Rad Laboratories, Richmond, CA) and on a gel filtration column (TSK-G 3000 SW, 60-cm length ; LKB Instruments, Gaithersburg, MD) as previously described (33) . After sequential chromatography on these two media, the fraction with CSIF activity is referred to as partially purified CSIF (ppCSIF) . Antigens. Keyhole limpet hemocyanin was obtained from Pacific Bio-Marine Laborato- ries, Inc . (Venice, CA) and used at a final concentration of 150 /Ag/ml . TNP-KLH (132 TNP residues per 106 daltons protein) was used at 1 p,g/ml (35) . Conalbumin (Sigma Chemical Co .) was used at 100 Kg/ml . Chicken red blood cells were obtained from PMK farms (Vacaville, CA) and used at 0.02% packed volume . Anticytokine Affinity Columns. Six affinity columns were prepared by coupling the mAbs XMG1 .2 anti-IFN-y, 22E9.11 anti-GM-CSF, S4B6 anti-IL-2, 8F8 .11 anti-IL-3, 11BIl anti- IL-4, and TRFK5 anti-IL-5 to Affi-Gel 10 (Bio-Rad Laboratories) . Each 1-2-ml column contained -10-20 mg of coupled antibody. CSIFAssay. Samples were diluted in 96-well flat-bottomed microtiter trays (Falcon Lab- ware, Oxnard, CA) in a volume of 0 .05 ml. Thl cells (5 x 104 cells/well), irradiated (2,500 rad) syngeneic spleen cells (5 x 105 cells/well), and antigen were added in a volume of 0.15 ml . 11BIl anti-IL-4 mAb (10,ug/ml) was added to the assay ifsamples were suspected ofcon- taining IL-4 . Levels of IFN-y and other cytokines in the 24-h supernatants were assayed . Omission of either antigen or APC resulted in negligible cytokine synthesis . When initially assayed, one batch of ppCSIF inhibited IFN-y production by -50% at a dilution of 1 :200 in an assay volume of0 .2 ml, and so the standard unit (SU) was defined by assigning a value of 1,000 SU/ml to this CSIF preparation . CSIF activity in unknown samples was quantitated by comparing levels of inhibition of IFN-y synthesis by the unknown with that of the stan- dard in each assay using a four-parameter curve-fitting program (Assay Zap ; Biosoft, Cam- bridge, UK) . Means t SD of triplicate cultures were determined . mRNA Measurement . Thl cells were stimulated with antigen plus APC that had been depleted of erythrocytes by centrifugation over Ficoll (Histopaque ; Sigma Chemical Co.) . At 8 and 12 h, aliquots of the cells were withdrawn and cytoplasmic RNA was isolated (36) . Dilutions ofRNA were bound to filters and hybridized with synthetic DNA probes that were "P-labeled by fill-in synthesis of overlapping DNA oligonucleotides as described previously (2) . The DNA probes corresponded to bases 900-989 for IFN--Y (2) and 1262-1363 for y-actin (37) . Filters were exposed to x-ray film and the autoradiographs quantitated by laser den- sitometry. Results Supernatantsfrom Th2 but not Thl Clones Inhibit Synthesis ofIFN--y by Thl Clones . Con A-induced supernatants from seven Th2 clones all inhibited IFN-'r production by HDK-1 cells (Fig. 1 A). This inhibitory activity was produced by D10 cells in re- sponse to Con A, antigen plus APC, or antiT3 plus APC (Fig . 1 B) . The inhibitory activity appeared to be produced only by stimulated cells, since Con A-stimulated CDC25 and CDC35 cells produced 866 and 708 U/ml ofinhibitory activity, respec- tively, whereas unstimulated cells produced <40 U/ml . We will call the factor respon- sible for this activity CSIF The decrease in IFN production could be due to a decrease in synthesis and/or - 'Y secretion . If a block in secretion was solely responsible for the reduced IFN-y super- natant levels, then the intracellular levels of IFN-y should be greatly elevated in the presence of CSIF When HDK-1 cells were stimulated with antigen plus APC, the maximum extracellular IFN-y levels h were and 1,100 ng/ml in the pres- at 24 250 ence and absence of CSIF, respectively. The corresponding maximum intracellular levels were 1 .0 and 4 .5 ng/ml . 4 Thl 208 Th2-DERIVED INHIBITOR OF CYTOKINE SYNTHESIS TH2 SUPERNATANT D10 (%) SUPERNATANT (%) FIGURE 1 . Th2 clones produce an inhibitor ofIFN-y synthesis by Thl cells . HDK-1 cells were stimulated with antigen plus APC in the presence of supernatants from seven Con A-induced /+g/ml) Th2 clones (A) or from D10 cells induced with Con A, Ag plus APC, or antiT3 mAb (4 .5 plus APC (B) . IFN--y secretion is expressed as a percentage of the levels produced in the absence of the Th2-derived inhibitor. Four Con A-induced Thl supernatants were depleted IFN--y and IL-2 (which of would interfere with the CSIF assay) by passage over monoclonal anticytokine affinity columns, and then tested for CSIF activity . TheseThl supernatants, even at a dilu- tion of 1 :10, did not inhibit IFN-7 production (Fig. 2 A) . The slightly increased IFN--y levels in cultures treated with the highest concentration ofM264-37 superna- FIGURE Thl clones secrete CSIF. Super- 2 . do not A-induced Thl clones were natants from four Con depleted of IL-2, IL-3, IFN-y, and GM-CSF, and for activity HDK-1 cells (A) . In- tested CSIF using hibition by ppCSIF is shown for comparison . The were in the presence of 14 same samples also tested ppCSIF (B) . The and lowerdashed lines U/ml upper the IFN-y the absence pres- represent levels of in and ence of CSIF, respectively. All values are expressed as a percentage of the IFN--y levels produced in the absence of both ppCSIF and Thl supernatants . MORENTINO ET AL 2085 tant could be accounted for by the residual IFN-y in this sample. Thus, Thl super- natants either do not contain CSIF, or they contain both CSIF and an inhibitor masks its activity. This second possibility is ruled out by the demonstration that that IL-2- and IFN-y-depleted Thl supernatants did not counteract CSIF activity (Fig . 2 B) . Taken together, these experiments demonstrate that Th2 but not Thl supernatants contain CSIF. CSIF Inhibits Production of Several Thl Cytokines. Since ppCSIF still contained GM- CSF and IL-3, the levels of these cytokines were further reduced by using anticytokine affinity columns before testing for CSIF's ability to inhibit the synthesis of several other Thl Cytokines . Production ofIFN-y was consistently inhibited in all Thl clones tested (e.g., Table 1) . The synthesis of IL-2, LT/TNF, IL-3, and GM-CSF was in- hibited in some cases but not others ; e.g ., IL-2 production was inhibited in HDK-1 cells but not in MD13-10 cells . When inhibition was observed, the extent of max- imum inhibition varied between different clones and cytokines, ranging from -60 to >95To . Even for a single clone, different cytokines were maximally inhibited at different concentrations ofCSIF In particular, the synthesis of LT/TNF was inhibited by relatively low concentrations of CSIF, but the extent of maximal inhibition was much less than that of IFN-y. CSIFInhibits Late but not Early Cytokine Synthesis. The kinetics of inhibition of cytokine synthesis were examined for antigen-stimulated HDK-1 cells (Fig . 3) . In general, the synthesis of all cytokines was not inhibited by CSIF at early times, e.g ., before early times 8.5 h (before 6.5 h for IL-2) . Two cytokines synthesized mainly at (GM- inhibited slightly or not all, whereas the synthesis of CSF and LT/TNF) were at mainly (IFN-y IL-3) was inhibited substan- cytokines produced at later times and exact of action cytokine synthesis tially. Differences in the kinetics CSIF and could for of thus account some of the variability inhibition seen in Table I . Inhibition of IFN- ,y Synthesis Occurs at the mRNA Level . HDK-1 cells were stimu- or absence ppCSIF, and total lated with antigen plus APC in the presence of cyto- RNA was from at and h after stimulation. RNA samples plasmic prepared cells 8 12 were tested for IFN- and y-actin mRNA levels by dot-blot hybridization . IFN-y ,y TABLE Effect of CSIF on the Synthesis of Thl Cytokines Percent of control synthesis level Cell line Cytokine 14 U/ml CSIF 42 U/ml CSIF 125 U/ml CSIF ,y t 1 HDK-1 IFN- 47 .6 t 0.8 29.1 .0 18 .6 t 0.7 71 ± t t IL-2 .7 2.3 59.6 3.9 40 .4 6.1 LT/TNF 41 1 .4 45 .1 t t .9 t 2.3 42 .8 1 .1 11 .4 t t 1 IL-3 63 .9 t 52 .6 6.8 38 .4 .9 8 t GM-CSF 86 .9 t .5 79.1 7.8 66 .8 t 9.3 MD13-10 IFN- ,y 4 .5 t .4 36 .0 t .9 27 3 23 .2 t 2.6 88 .2 5.1 t IL-2 t 109.3 13 .3 96 .0 t 16 .4 IL-3 60 .2 5.1 63 .0 t 6.5 t t 2.7 51 .0 GM-CSF 109.9 13 .0 119.9 t 7.4 .6 t t 97 19 .2 HDK-1 and MD13-10 cells were stimulated with antigen plus APC in the presence of varying dilutions of ppCSIF that had been depleted of GM-CSF and IL-3 . Cytokine levels in the 24-h supernatants are expressed as a percentage of the amount produced in the absence of CSIF . 2086 Th2-DERIVED INHIBITOR Thl OF CYTOKINE SYNTHESIS FIGURE 3 . Kinetics inhibition cytokine synthesis CSIF HDK-1 were of of by cells stimulated with antigen plus APC in the presence or absence of 100 U/ml ppCSIF that had been depleted of IL-3, IL-4, IL-5, and GM-CSF. Supernatants were collected at the indicated times and as- sayed for cytokine levels . mRNA levels were moderately reduced by CSIF at 8 h, and strongly reduced at 12 h over a range of 5-40 U/ml CSIF (Fig . 4) . In contrast, y-actin mRNA levels were not significantly reduced by CSIF at either 8 or 12 h, and possibly showed a slight enhancement at 12 h . The inhibition ofIFN-y protein levels in the 24-h super- natants by CSIF was slightly less than the inhibition of mRNA levels at 12 h . Thl but not Th2 Clones Respond to CSIF. ppCSIF was tested on seven Thl clones that were derived from three mouse strains and were specific for particulate antigens, FIGURE 4 . CSIF reduces mRNA levels for IFN-'Y but not actin . HDK-1 cells were stimulated with antigen plus APC in the presence or absence of ppCSIF. Cellswere harvested at 8 and 12 h for cytoplasmic mRNA isolation, and supernatants were taken at 24 h . y for IFN- protein determination . mRNA were by levels assessed hybrid- ization DNA probes, with synthetic and protein was measured by ELISA . All values are expressed as a percentage of levels produced in the absence of CSIF. FIORENTINO ET AL . MD13.2 A 0 MD13 .10 e HDK-1 L82-1 OK76.1 MUMC1.2 BA 0 N.6. MUMC2-B FIGURE 5. Thl but not Th2 u . clones respond to CSIF. Seven Thl clones were stimulated with antigen APC in plus the pres- ence of dilutions of ppCSIF, and secretion was measured IFN-,y (A). HDK-1 (Thl) and D10 1u (Th2) cellswere stimulated with antigen plus APC in the pres- ppCSIF(%) ence of dilutions ofppCSIF that had been depleted of IL-3, IL- 4, IL-5, and GM-CSF (B) . Levels of IL-2, IL-3, and IFN- (HDK-1) IL-3, IL-4, and ,y or IL-5 (D10) were assayed in the IL3 24-h All values " supernatants . u-a are expressed as apercentage of cytokine levels produced in the n the absence of ppCSIF . IL2 IL3 O IFNy ppCSIF(%) soluble proteins, or alloantigens . IFN- , y synthesis by all seven clones was inhibited by ppCSIF (Fig . 5 A) . Since some of the Th2 cytokines may have been present in the ppCSIF preparation, IL-3, IL-4, and IL-5 were depleted by passage over mAb affinity columns . The resulting CSIF preparation did not suppress the production of IL-3, IL-4, or IL-5 by the Th2 clone D10 (Fig. 5 B) . Two additional experiments with D10 yielded similar results, and in other experiments (not shown), CSIF did not inhibit the production of IL-5 by the Th2 clones MB2-1, CDC25, or CDC35, nor did it inhibit IL-4 synthesis by MB2-1 . Synthesis of other cytokines in response to antigen plus APC in these experiments was too low to reliably determine CSIF effects . In general, we have not seen convincing inhibition of the production of any cytokine by Th2 cells responding to antigen plus APCs . CSIF inhibited synthesis CSIF Action on Thl Cells May Be Indirect. in re- IFN-'y plus APC, but not ,y synthesis stimu- sponse to antigen plus APC or antiT3 IFN- a polystyrene surface, or by Con A (Table II) . Since only lated by anti-T3 bound to were inhibitable, CSIF may act indirectly on the the stimulations involving APC 2088 Th2-DERIVED INHIBITOR Thl CYTOKINE SYNTHESIS OF TABLE II CSIF Is Effective in Stimulations Involving APC Percent of control IFN--y synthesis Exp. Stimulation 5 U/ml CSIF 19 U/ml CSIF 74 U/ml CSIF 294 U/ml CSIF 1 Antigen plus spleen .6 f 72 12 .8 29 .1 t 3 .5 20 .5 t 6 .6 Anti-T3 plus spleen 88 .7 t 8 .0 78 .6 t 2 .8 68 .5 t 2 .6 Anti-T3 (coated) 95 .7 t 6 .7 103 .3 f 6 .7 94 .5 f 9 .3 Con A 102 .7 ± ± 4 .2 95 .4 5 .3 94 .7 t 5 .0 2 Antigen plus spleen 17 .5 t 5 .9 12 .8 1 .6 8 .2 ± .8 Anti-T3 ± 7 .2 plus spleen 54 .5 3 .1 45 .6 t 6 .0 42 .0 t HDK-1 cells were stimulated with antigen plus APC, anti-T3 mAb (9 ug/ml) plus APC, anti-T3 bound to 96-well flat-bottomed trays, or Con A (5 Ag/ml), each in the presence or absence of CSIF-containing 1310 supernatant . Trays were coated with anti-T3 by incubating with 50 ug/ml anti-T3 mAb in 50 141 PBS for 1 h at 37°C, following which, the plates were extensively washed . IFN-1' levels in 24-h supernatants are ex- pressed as a percentage of the amounts produced in the absence of CSIF, using the corresponding stimula- tion condition . T cells, via a cell type in the APC population . CSIF acts mainly at later times (Fig . 3), consistent with the hypothesis that CSIF is first modified into, or induces a sec- ondary factor that then inhibits IFN--y synthesis . Additional evidence for indirect was varying 10-fold T action obtained by the CSIF assay conditions . changes in cell and/or APC numbers did not affect the extent of inhibition of IFN-y synthesis by CSIF or the sensitivity ofdetection of CSIF (data not shown) . In contrast, saturating 16% instead amounts of CSIF inhibited IFN-y synthesis more effectively (to of 35 % ofcontrol IFN-y amounts) if the assay volume were reduced twofold . If the residual IFN-y levels represent secreted early before CSIF action, these the cytokine at times are consistent with inhibition synthesis being dependent on results the of IFN--y a soluble factor produced in the assay. In this case, such a factor would accumulate effective concentrations more rapidly in a smaller assay volume . to CSIF not Directly Inhibit Thl Proliferation . In contrast its effect on Thl cytokine Does to of Thl cells proliferate synthesis, CSIF had no measurable effect on the ability to in response to IL-2 . In the presence of exogenous IL-2 (500 ng/ml), antigen-stimulated proliferated 5.2-fold and by day 6, as assessed by HDK-1 cells by day 3, 42 .6-fold cell counting . The corresponding proliferation values in the presence of IL-2 plus 100 U/ml ppCSIF were 5.0-fold and 40 .0-fold . Similar results (not shown) were ob- tained using the MTT assay. However, in the absence of exogenous IL-2, ppCSIF proliferation of when at the time was sometimes able to decrease HDK-1 cells added not shown) ; presumably, inhibition was indirect of antigen stimulation (results this and was a result of CSIF's ability to reduce IL-2 production by HDK-1 cells (Table I, Fig . 3). Other Cytokines that Affect IFN--y Synthesis. In some but not all experiments, exoge- nous IL-4 and especially IL-2 increased the production ofIFN-y by Thl clones (data not shown) . Since IL-2 was synthesized in variable amounts by stimulated Thl cells, exogenous IL-2 and IL-4 might only be effective in experiments in which nonsaturating IL-2 in Fig . amounts of endogenous were synthesized . In the experiment shown 6 A, anti-IL-2 antibody inhibited the production of indicating that normal IFN-y, IFN--y synthesis was enhanced by IL-2 produced by the Thl cells responding to an- AL . 9 FIORENTINO ET 208 FIGURE 6 . Effects of IL-2, IL-4, andTGR S on IFN-y synthesis. HDK-1 cells were stimulated with antigen plus APC with var- ious combinations of cytokines and antibodies: IL-2 (500 ng/ml), IL-4 (50 ng/ml), TGF-0 (10 ng/ml), ppCSIF (125 U/ml), 54136 anti-IL-2 (10 1&g/ml), rabbit antiTGR0 (50jig/ml), and normal rabbit IgG (50 Kg/ml) (A and B). Titrations of these reagents showed that these levels were saturating . All cultures contained 111311 anti-IL-4 (10 pg/ml), except those in which IL-2 or IL-4 was added (A) . All values of IFN-y are expressed as a percentage the an- levels produced in the absence of ppCSIF, tibodies, and cytokines. tigen plus APC . When anti-IL-2 antibody was present, IL-4 enhanced IFN-y produc- tion slightly, and excess IL-2 (overcoming the effect ofthe anti-IL-2 antibody) could strongly enhance IFN-y levels . Although both IL-2 and IL-4 antagonized the effect of CSIF by increasing the amount of IFN-y secreted, CSIF was still active in the presence or absence of saturating amounts of IL-2 or IL-4 (Fig . 6 A) . Partially purified antiTGFf3 antibodies slightly enhanced the synthesis of IFN-y, whereas exogenous TGF-0 slightly inhibited IFN-y production (Fig. 6 B) . These results suggest that IFN-y synthesis can be inhibited by TGFf3, and that TGF-0 is produced in the assay under normal conditions . CSIF inhibited IFN-y produc- tion in the presence or absence of TGF-0 or antiTGF0, showing that CSIF is not TGF-i3 . Partial Purification and Biochemical Characterization of CSIF . CSIF-containing D10 supernatants were chromatographed on two media . On both columns, CSIF ac- tivity eluted in a single area. On the hydroxylapatite-based support, 95% of the ap- plied activity was recovered ; also shown are the elution positions of other proteins (Fig . 7 A) . On the gel filtration column, 63 % of the applied activity eluted broadly, with the fractions of high activity corresponding to apparent molecular masses of 27-50 kD (Fig . 7 B) . The width ofthe peak may be the result ofmicroheterogeneity similar to that observed with other cytokines (e .g ., IL-5 ; [33]) . CSIF could be se- quentially chromatographed on both columns in either order with no significant alteration of relative mobility or activity yield (data not shown) . These results sug- gest that CSIF may be a single entity, and we have not yet observed any evidence for directly synergistic interactions of CSIF with any other cytokine. ppCSIF (2,100 U .4) in 0.25 ml 10 mM phosphate buffer, pH 7 was carefully acidified to pH 2 with allowed to stand for 1 h at room temperature, neutralized with 1 M Hepes HCl, (sodium salt), and immediately assayed ; <5% of the activity remained . CSIFIs Distinctfrom Known Thl and Th2 Cytokines. Various blocking anticytokine monoclonal and polyclonal antibodies were tested for the ability to inhibit CSIF ac- 2090 Th2-DERIVED INHIBITOR OF Thl CYTOKINE SYNTHESIS FIGURE 7 . Column chromatography ofCSIFcontaining D10 supernatant . 10 ml of 27-fold con- centrated D10 supernatant applied to a Bio-Gel HPHT column in 10 mM sodium phosphate, pH 7.4, 0.12 mM CaC12, 0.01% Tween-20 at a flow rate of0.5 ml/min was eluted with a 60-min gradient to 0.35 M sodium phosphate, pH 7.4, 3.5 uM CaC12, 0.01% Tween-20. The elution positions of other proteins are indicated by horizontal bars (A) . Gel filtration of 0.2-ml 30-fold concentrated D10 supernatant on TSK-G 3000 SW was performed as previously described (33) (B). The elution positions of proteins of known molecular weight are marked by the arrows. In both cases the absorbance at 280 nm was monitored (curved line) and aliquots from the 1-min fractions were assayed for CSIF activity (shaded areas). tivity. CSIF activity was not reduced by mAbs specific for IL-2, IL-3, IL-4, and IL-6 (10 ttg/ml), IL-5 (2 Wg/ml), and GM-CSF (20 jig/ml), or by polyclonal anti- TGF antibodies (50 jig/ml) . Several cytokines were also tested in the CSIF assay and found to be inactive, either as agonists or antagonists . The synthesis of IFN-'y was enhanced or inhibited by <11% by IL-1 (101 U/ml), IL-5 (2 .4 ng/ml), IL-6 (25 ng/ml), or IL-7 (500 U/ml) . IFN-y synthesis was enhanced by IL-2 and IL-4 (Fig . 6), slightly enhanced by P40 (results not shown), and inhibited by TGF0 (Fig. 6) . Experiments using recombinant cytokines and mAbs, as well as knowledge of its biochemical properties and presence in Th2 but not Thl supernatants, distinguish CSIF from many of the known cytokines (Table III) . In general, we required at least two nonidentical properties before each cytokine was considered distinct from CSIF IL-7, According to these criteria, CSIF is not IL-1, IL-2, IL-3, IL-4, IL-5, IL-6, be GM-CSF, IFN--y, IT, TNF, TGFa, P40, or met-enkephalin, and is unlikely to soluble immune response suppressor (SIRS) (38) . Discussion and DTH responses are often mutu- Several lines of evidence show that antibody If these responses are mainly due to Th2 and Thl cells, ally exclusive (16-18, 39). Thl and Th2 cells reciprocally inhibit the growth and respectively, this suggests that the other cell type. Some of the known mechanisms include the inhibi- function of tion of growth of Th2 clones by IFN-y (19, 20), the inhibition of all IL-4 effects on B cells by IFN--y (40), and the inhibition of Thl proliferation by Th2-derived identified as an autocrine factor (21) . Although the activity in the last study was FIORENTINO ET AL . III TABLE with Known Cytokines Evidence for Nonidentity of CSIF has Biochemically Present in Absent in Ab blocks Cytokine CSIF activity similar to CSIF Cytokine Th2 sup Thl sup CSIF activity No No : separates by size IL-1 No Yes - IL-2 No No No - No No - No : separates on HPHT IL-3 Yes No No: separates on HPHT, size IL-4 Yes Yes No No No No - IL-5 Yes Yes Yes Yes No No No : separates by size IL-6 No No: not acid labile IL-7 - - - No - No : not acid labile GM-CSF Yes No - - - IFN No No LT No No - - - No : not acid labile TNF Yes No - - - - Yes (partial) No : not acid labile TGF-0 No No No : not acid labile P40 Yes - - Yes No - - No : separates by size Met-enk - - - - No : separates by size SIRS growth inhibitor for Th cells, the results need to be reinterpreted in light of recent information (1, 3, 5). The Th2-derived growth inhibitory activity inhibited the growth of a Thl clone "1,000-fold more effectively than D10 growth (21), but it appears to be distinct from CSIF since the two activities have different apparent molecular masses, and ppCSIF does not inhibit proliferation of Thl clones . In addition to IFN-y and the D10-derived proliferation inhibitor, the CSIF identified in this study could be another ofthe crossinhibitory regulators ofThl andTh2 cells. Thl function should be inhibited by CSIF, since IFN-y is an important mediator of several Thl functions (7, 12, 41) . There may also be an indirect enhancement of Th2 growth, since the decrease in IFN-y levels should reduce inhibitory effects on Th2 growth (19, 20). CSIF inhibition of IL-2 but not IL-4 synthesis may also slightly reduce the relative growth rates of Thl cells, since Th2 cells respond better than Thl cells to the proliferative effects of IL-4 (3) . We have suggested previously (3) that these crossinhibitory cytokine effects may be important early in immune responses, whereas the production of specific Ig isotypes may counterbalance this effect later in the response . CSIF could also have a more subtle role in the regulation of immune responses. In vitro, Thl clones are often very poor B cell helpers (8, 9), unless the cytokine levels are adjusted by decreasing IFN-y and increasing IL-2 levels (7) . Since CSIF inhibits IFN-y production, often more effectively than IL-2 production, and freshly isolated Thl clones tend to produce more IL-2 than long-term Thl clones (2), CSIF may modify the functions ofThl cells so that they become more effectiveB cell helpers. If true, this effect would only be seen in responses in which at least some Th2 cells Thl were activated . This may explain the apparently contradictory observations that B DTH cells can be good cell helpers under some circumstances, but that a strong response (presumably due to strong Thl activation) is often unaccompanied by significant antibody synthesis . CSIF appears to selectively inhibit Thl cytokine synthesis, rather than generally Thl CYTOKINE SYNTHESIS 2092 Th2-DERIVED INHIBITOR OF -y-actin growth or viability. CSIF treatment did not reduce the levels of inhibit cell IL-2 . Although mRNA, nor did it inhibit proliferation in response to exogenous IFN- production (42, and Fig. 6), and CSIF often inhibits IL-2 produc- IL-2 enhances -y presence in Fig. 6 also show that CSIF exerts its effect in either the tion, the results IL-2 . This indicates that CSIF does not inhibit IFN-'y synthesis simply or absence of (43, IL-2 production . Although TGF/3 also inhibits IFN-y synthesis by inhibiting saturating Fig . CSIF was able to inhibit IFN--y synthesis more effectively than and 6), TGF0 . TGF0 . This indicates that CSIF does not act only by inducing amounts of CSIF of CSIF on Thl cytokine synthesis may be indirect, since (a) The action .y A ; (b) IFN- IFN- synthesis in response to antigen plus APC but not Con inhibits -y volume of the synthesis is inhibited only at later times after stimulation ; and (c) the Two possible mech- critically important to the degree of inhibition obtained . assay is by a cell in the induction by CSIF of an effector molecule synthesized anisms are form by the APC APC population, or the modification of CSIF to an active the pretreated Preliminary experiments (unpublished data) show that APC population . induce Thl are only marginally reduced in their ability to subsequently with CSIF .y primarily on the APC popu- cells to produce IFN- . This indicates that, if CSF acts of T cells . the effect is mostly reversible, or requires the presence lation, effect of CSIF is mediated at Analysis of IFN- mRNA levels showed that the -y measurements. Possible mRNA level, in agreement with intracellular protein the degradation, or altered include reduction of transcription, accelerated mechanisms reduced at 8 h, and strongly mRNA processing . IFN--y mRNA levels were slightly , y showing that CSIF inhibition reduced by 12 h, consistent with IFN- secretion data ,y of IFN- production may also is detected only at late times . The delayed inhibition inhibits cytokine synthesis ; the re- explain why saturating CSIF never completely of cytokines synthesized at early sidual cytokine levels may represent the amounts times after stimulation . displayed by the majority of long-term Although the Thl and Th2 phenotypes are patterns exist among normal Th cells mouse Th clones, other cytokine secretion N. E., J . H . Schumacher, T. A . T. short-term clones (2, 13, 20, and Street, and Leverah, and T. R. Mosmann, manuscript Fong, H . Bass, D . F. Fiorentino, J. A . phenotypes may be precursors of Thl and submitted for publication) . These other of such precursors are not yet known Th2 cells, and the details of the functions Th types, production and effects in these other It will be important to assess CSIF Thl and Th2 cytokines . particularly in cells secreting both that Th2 cells of CSIF is complicated by the fact The interpretation of the role . This situation partially counteracts the effects of CSIF also produce IL-4, which ,y these two cytokines of IL-2 and IFN- by Thl cells, since is analogous to the production pairs ofcytokines Th2 cells (19, 20) . It is possible that these have opposing effects on that the local clear- different ratios under different conditions, or are produced in to temporal differences in the of each cytokine are very different, leading ance rates awaits improved methods for detecting effects observed . Resolution of these questions and manipulating cytokine levels in vivo. Summary is secreted by Th2 clones in response A cytokine synthesis inhibitory factor (CSIF) from Con A-induced but is absent in supernatants to ConA or antigen stimulation, FIORENTINO ET AL 2093 Thl clones. CSIF can inhibit the production ofIL-2, IL-3, lymphotoxin (LT)/TNF, IFN--y, and granulocyte-macrophage CSF (GM-CSF) Thl responding by cells to antigen significantly Trans- and APC, but Th2 cytokine synthesis is not affected . forming S (TGFj3) inhibits IFN-y production, effec- growth factor also although less tively than CSIF, whereas IL-2 and IL-4 partially antagonize activity of the CSIF CSIF inhibition is complete, since early cytokine synthesis of cytokine synthesis not (before 8 h) not significantly affected, whereas later synthesis is strongly inhibited . is In the presence ofCSIF, IFN-y mRNA levels are reduced slightly 8, and strongly at at h after stimulation . Inhibition of cytokine expression by CSIF is not due 12 to a general reduction in Thl cell viability, since actin mRNA levels were not reduced, and proliferation of antigen-stimulated cells in response to IL-2, was unaffected Biochemical characterization, mAbs, recombinant or purified cytokines showed and CSIF from IL-1, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, GM- that is distinct IFN-y, CSF, TNF, and P40 . The role CSIF in crossregulation TGF,3, IT, potential of of Thl and Th2 responses is discussed . We thankJ. Abrams, R. Coffman, M . Howard, R . Kastelein, F Lee, Palladino, J. Schreurs, M. J . van Snick, and Zurawski for providing valuable cytokine and antibody reagents . We G . J . providing technical advice and F Vega for synthesizing oligonucle- thank Schumacher for otides . Note added in proof The identity of CSIF as a novel cytokine has recently been confirmed by the isolation of a cDNA clone that expresses CSIF activity when transfected into COS cells (Moore, K . W. et al ., manuscript in preparation) . mAbs specific for CSIF react with both the recombinant protein and the 1310-derived factor (Mosmann, T. R . et al ., manuscript in of preparation) . The sequence the cDNA clone is different from all previously known cytokines . At the time of submission of this cDNA for clone sequence publication, we will request the Nomenclature Committee of the International Union of Immunological Societies to assign an IL designation for factor. this Received for publication 19June 1989 and in revisedform 28 August 1989. References 1 . Mosmann, T R ., H . Cherwinski, M . W. Bond, M . A . Giedlin, and R . L . Coffman . 1986 . Two types ofmurine helper T cell clone . I . Definition according to profiles of lym- phokine activities and secreted proteins . . Immunol. 136:2348 . 2 . Cherwinski, H . M., J . H . Schumacher, K . D. Brown, and T. R . Mosmann . 1987 . Two types of mouse helper T cell clone. III . Further differences in lymphokine synthesis be- tween Thl and Th2 clones revealed by RNA hybridization, functionally monospecific bioassays, and monoclonal antibodies. . Exp. Med. 166 :1229 . 3 . Mosmann, T R ., and R . L . Coffman . 1989. Heterogeneity of cytokine secretion pat- terns and functions of helper T cells . Adv. Immunol. 46 :11 . 4 . Brown, K . D ., S . M . Zurawski, T R . Mosmann, and G . Zurawski . 1989 . A family of small inducible proteins secreted by leukocytes are members of a new superfamily that includes leukocyte and fibroblast-derived inflammatory agents, growth factors, and indi- cators of various activation processes. . Immunol. 142:679 . 5 . Kim, J ., A . Woods, E . Becker-Dunn, and K. Bottomly. 1985 . Distinct functional pheno- types of cloned la-restricted helper cells . . Exp. Med. 162 :188 . 6 . Cher, D. J ., and T. R . Mosmann . . 1987 Two types of murine helper T cell clone . II . Delayed-type hypersensitivity is mediated by Thl clones . . Immunol. 138:3688 2094 Th2-DERIVED INHIBITOR OF Thl CYTOKINE SYNTHESIS 7 . L ., B . W. D . A . J. Coffman, R . Seymour, Lebman, D. D. Hiraki, A . Christiansen, B . Shrader, H . T. M . Cherwinski, H . F. Savelkoul, F. D. Finkelman, M . W. Bond, and R . Mosmann . 1988 . Th e role of helper T cell products in B differentiation and mouse cell isotype regulation. Immunol. Rev. 102 . :5 8 . Killar, L ., G. MacDonald, A . K . J . West, Woods, and Bottomly. 1987 . Cloned, la-restricted T cells that do not produce interleukin 4(IL-4)/B cell 1(BSF to stimulatory factor 1) fail help antigen-specific B cells . Immunol. 138 :1674 . 9 . Boom, W. H ., D. Liano, and A . K . Abbas 1988 . Heterogeneit helper/inducer T . y of lymphocytes. II . Effects of interleukin 4- and interleukin T cell on 2-producing clones resting B lymphocytes . J . . Med. 167 :1352 Exp . 10 . Stevens, T. L ., A. Bossie, V. T. M . Sanders, R . Fernandez-Botran, R. L . Coffman, R . Mosmann, and E . S . Vitetta. 1988. Subset s of antigen-specific helper T cells regulate isotype secretion by antigen-specific B 334 cells . Nature (Loud.). :255 . 11 . Giedlin, M . A ., B . M . Longenecker, T. . e and R . Mosmann 1986 . Murin T-cell clones specific for chicken erythrocyte . 97 alloantigens . Cell Immunol. :357 . . Stout, . D., and . Bottomly. 1989. 12 R K Antigen-specific activation of effector macrophages by IFNy-producing (Thl) T cell clones Failure of T . IL4-producing (TH2) cell clones to activate effector function in macrophages . Immunol. 142:760 . 13 . Swain, S . L., D. T. McKenzie, A . D. Weinberg, and W. Hancock. 1988 . Characteriza- tion of T helper 1 and cell subsets in normal mice . Helper T cells responsible for IL-4 and IL-5 production are present as precursors that require priming before they develop into lymphokine-secreting cells . Immunol. 141 :3445 . 14 . Budd, R . C .,J . C . Cerottini, and H . R . MacDonald 1987 . Selectively increased produc- tion of interferon-gamma by subsets of Lyt-2' and L3T4' T cells identified by expres- sion of Pgp-1 . Immunol. 138:3583 . 15 . Firestein, G . S ., W. D. Roeder, J . A . Laxer, K . S. Townsend, C . T. Weaver, J . T. Hom, J . Linton, B . E . Torbett, and A. L . Glasebrook . 1989 . A new murine CD4' T cell subset with an unrestricted cytokine profile . . Immunol. 143 :518 . 16 . Parish, C . R . 1972 . The relationship between humoral and cell-mediated immunity. Trans- plant. Rev. 13 :35 . 17 . Katsura, Y. 1977 . Cell-mediated and humoral immune responses in mice . III . Dynamic balance between delayed-type hypersensitivity and antibody response . Immunology. 32 :227 . 18 . Heinzel, F. P, M . D. Sadick, B . J . Holaday, R . L. Coffman, and R. M . Locksley. 1989 . Reciprocal expression of interferon y or interleukin 4 during the resolution or progres- sion of murine leishmaniasis. Evidence for expansion of distinct helper T cell subsets . J Exp. Med. 169 :59 . 19 . Fernandez-Botran, R ., V. M . Sanders, T. R . Mosmann, and E . S . Vitetta . 1988 . Lymphokine-mediated regulation of the proliferative response of clones of T helper 1 and helper 2 cells . . Exp. Med. 168 :543 . 20 . Gajewski, T. F., and F. W. Fitch . 1988. Anti-proliferative effect of IFN-gamma in im- mune regulation . I . IFN-gamma inhibits the proliferation of Th2 but not Thl murine helper T lymphocyte clones . . Immunol. 140:4245 . . J . 21 Horowitz, B., J . Kaye, P. J. Conrad, M . E. Katz, and C . A . Janeway, Jr. 1986 . Auto- crine of line T cells . Proc . :1886 growth inhibition a cloned ofhelper . Nail. Acad. Sci USA . 83 . C . A . . 22 . Kaye, J ., S . Porcelli, J . Tite, G . Jones, and Janeway, Jr 1983 . Both a monoclonal antibody and antisera specific for determinants unique to individual cloned helper lines for T cell can substitute antigen and antigen-presenting cells in the activation of cells . J . 158 Exp. Med :836 . 23 . Tony, H . P, and C . Parker. 1985 . Major histocompatibility complex-restricted, poly- D. clonal B cell responses resulting from helper T cell recognition of antiimmunoglobulin presented by small B lymphocytes . . Exp. Med . 161 :223 . 24 . Nabel, G ., J . S . Greenberger, M . A . Sakakeeny, and H . Cantor. 1981 . Multiple biologic FIORENTINO ET AL . 2095 activities of a cloned inducer T-cell population . Proc. Nail. Acad. Sci. USA . 78 :1157 . 25 . Watson, J . 1979 . Continuous proliferation of murine antigen-specific helper T lympho- cytes in culture . . Exp. Med. 150 :1510 . 26 . Espevik, T., and J . Nissen-Meyer . 1986. A highly line, sensitive cell WEHI 164 clone 13, for measuring cytotoxic factor/tumor necrosis factor from human monocytes . j Im- munol. Methods. 95 :99 . 27 . Schumacher, J . H ., A . O'Garra, B . Shrader, A. van Kimmenade, M . W. Bond, T R . Mosmann, and R . L . Coffman . 1988 . Th e characterization offour monoclonal antibodies specific for mouse IL5 and development of mouse and human IL5 enzyme-linked irri- munosorbent assays . . Immunol. 141 :1576 . 28 . Abrams, J . S ., and M . K . Pearce. 1988 . Development of rat anti-mouse interleukin 3 monoclonal antibodies which neutralize bioactivity in vitro . . Immunol. 140 :131 . 29 . Ohara, J ., and W. E. Paul . 1985 . Production of a monoclonal antibody to and molecular characterization of B-cell stimulatory factor-1 . Nature (Loud.). 315 :333 . 30 . Vink, A., P. G . Coulie, P Wauters, R . P. Nordan, andJ . van Snick . 1988 . B cell growth and differentiation activity of interleukin-HPI and related murine plasmacytoma growth factors . Synergy with interleukin 1 . Eur. Immunol. . 18 :607 31 . Bluestone, J ., D . Pardoll, S. Sharrow, and B. J . Fowlkes . 1987 . Characterization of mu- rine thymocytes with CD3-associated Tcell receptor structures . Nature (Land.). 326 :82 . 32. Uyttenhove, C ., R . J . Simpson, and J . van Snick . 1988. Functional and structural char- acterization of P40, a mouse glycoprotein with Tcell growth factor activity. Proc . Natl. Acad. Sci. USA . 85:6934 . 33. Bond, M . W., B. Shrader, T. R . Mosmann, and R . L. Coffman . 1987 . A mouse T cell product that preferentially enhances IgA production . II . Physicochemical characteriza- tion . Immunol. 139 :3691 . J. 34 Mosmann, T. R . 1983 . Rapid colorimetric assay for cellular growth and survival : appli- cation to proliferation and cytotoxicity assays. Immunol. Methods. 65 :55 . J. 35 . Kiefer, H . 1979 . Chemical modification of proteins, and haptens solid supports . In Im- munological Methods, Vol . 1 . I . Lefkovits, and B . Pernis, editors . Academic Press, New York. 36 . White, B. A ., and F. C . Bancroft . 1982 . Cytoplasmi c dot hybridization . Simple analysis of relative mRNA levels in multiple small cell or tissue samples . Biol. Chem. 257 :8569 . 37 . Leader, D. P, I . Gall, and H . Lehrach. 1985 . The structure of a cloned mouse gamma- actin processed pseudogene . Gene. (Amst.). 36 :369 . 38 . Webb, D. R ., K . Mason, G . Semenuk, T M . Aune, and W. C . Pierce . 1985 . Purificatio n and analysis of isoforms of soluble immune response suppressor (SIRS) . . Immunol. :3238 . 39 . Crowle, A . J ., and C . C . Hu . 1966. Spli t tolerance affecting delayed hypersensitivity and induced in mice by pre-immunization with protein antigens in solution. Clin. Exp. Im- munol. 1:323 . 40 . Mosmann, T. R ., and A . Zlotnik. 1989 . Multiple functions IL4 of and its role in immune regulation. In Growth Factors, Differentiation Factors, and Cytokines. A . Habenicht, editor. Springer-Verlag, Heidelberg . In press . 41 . Fong, T A. T, and T. R . Mosmann . 1989 . The role of IFN-y in Delayed Type Hyper- sensitivity Thl mediated by clones . . Immunol. In press . 42 . Kelso, A ., H . R . MacDonald, K . A . Smith, J . C . Cerottini, and K. T. Brunner. . Interleuki n 2 enhancement oflymphokine secretion by T lymphocytes : analysis of estab- lished clones and primary limiting dilution microcultures . . Immunol. 132 :2932 . 43 . Espevik, T., I . Figari, M . R . Shalaby, G. Lackides, G. Lewis, H . Shepard, and M . A . Palladino . 1987 . Inhibition of cytokine production by cyclosporin A and transforming growth factor S. . Exp. Med. 166 :571 . http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Journal of Experimental Medicine Pubmed Central

Two types of mouse T helper cell. IV. Th2 clones secrete a factor that inhibits cytokine production by Th1 clones

The Journal of Experimental Medicine , Volume 170 (6) – Dec 1, 1989

Loading next page...
 
/lp/pubmed-central/two-types-of-mouse-t-helper-cell-iv-th2-clones-secrete-a-factor-that-pomSq2E0O4

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Pubmed Central
ISSN
0022-1007
eISSN
1540-9538
Publisher site
See Article on Publisher Site

Abstract

TWO TYPES T HELPER CELL OF MOUSE IV. Th2 Clones Secrete a Factor that Inhibits Cytokine Production by Thl Clones BY DAVID F. FIORENTINO, MARTHA W BOND, AND TIM R . MOSMANN` From the Department Immunology, DNAX Research Institute, Palo Alto California 94304 of Many long-term mouse Th cell clones can be divided into two types, based on the pattern of cytokines secreted in response to antigen or lectin stimulation . Thl clones secrete IL-2, IFN-y, (LT),' whereas Th2 clones express IL- and lymphotoxin 4, IL-5, function, P600 (1-4) . Both clone se- IL-6, and a gene of unknown types of IL-3, TNF preproenkephalin, crete granulocyte-macrophage CSF (GM-CSF), a, The between Thl and Th2 and several other induction-specific proteins. differences clones cytokine synthesis patterns lead to markedly different functions (5-8) . Th2 helpers (8, can provide are normally better B cell 9), while Thl clones, although they (7, 11), B cell help in some circumstances 10, preferentially induce macrophage acti- vation (12) and delayed-type hypersensitivity (DTH) (6) . Recent evidence (Street, H. T H . Bass, D. F Fiorentino, A . Leverah, N. E., J . Schumacher, T A. Fong, J. (13-15) and T. R . Mosmann, manuscript in preparation) suggests that other differen- Thl repre- tiation states of mouse Th cells exist, and that the and Th2 phenotypes sent activated effector cells . The existence of these two major types ofTh cell may explain the well-documented separation of immune responses into DTH or antibody production, which often ap- pear to be mutually exclusive (16) . If B cell help and DTH are predominantly medi- ated by Th2 and Thl cells, respectively, this suggests that Thl and Th2 cells may be mutually inhibitory (17, 18) . The Thl product IFN-y inhibits proliferation of Th2 clones in vitro (19, 20), while an unknown product of a Th2 clone may inhibit the proliferation of Thl clones (21) . In addition to crossinhibition of proliferation, we considered the possibility that inhibitors of Thl and Th2 effector function also exist . Here, we report the preliminary characterization of a cytokine produced by Th2 clones that inhibits the synthesis of several cytokines by Thl clones . This cytokine, which we have called cytokine synthesis inhibitory factor (CSIF), has been distin- guished from the other known Th2 cytokines . Address correspondence to Tim R. Mosmann, Department ofImmunology, DNAX Research Institute, . . 901 California Avenue, Palo Alto, CA 94304 Dr Mosmann's address afterJanuary 1, 1990 will be the Department ofImmunology, 865 Medical Services Building, University of Alberta, Edmonton, Alberta T6G 2H7, Canada. I paper . synthesis DTH, Abbreviations used in this CSIF, cytokine inhibitory factor; delayed-type hyper- sensitivity; GM-CSF, granulocyte-macrophage CSF; ppCSIF, partially purified CSIF; TGR0, trans- forming growth factor S. J . EXP. MED. ® The Rockefeller University Press - 0022-1007/89/12/2081/15 $2 .00 Volume 170 December 1989 2081-2095 Th2-DERIVED INHIBITOR 2082 OF Thl CYTOKINE SYNTHESIS Materials and Methods Mice. BALB/c, C57BL/6, and CBA/J mice were obtained from TheJackson Laboratory (Bar Harbor, ME) and the Institute of Medical Research (SanJose, CA) . Female mice 8-12 wk old were used in all experiments . Cell Lines. Thl clones included HDK-1, : BALB/c anti-KLH (2) ; MD13-2, BALB/c anti- CRBC (1) ; MD13-10, BALB/c anti-CRBC (1) ; GK15-1, CBA/J anti-CRBC (11) ; LB2-1, C57BL/6 anti-CRBC (1) ; ML3-A-Cl-2-Ba, C57BL/6 anti-BALB/c ; and ML3-A-C2-6-Nb, C57BL/6 anti-BALB/c . Th2 clones included : D10.G4 .1 (D10), AKR/J anti-conalbumin, ob- tained from C . Janeway (Yale University, New Haven, CT) (22) ; CDC25 and CDC35, C3D2 anti-rabbit obtained from D. IgG, Parker (University of Massachusetts, Worcester, MA) (23) ; D9, C57BL/6, G . ; obtained from Nabel (Harvard University, Boston, MA) (24) MB2-1, C57BL/6 (1) ; and M411-2 and M411-6, BALB/c anti-CBA/J . The maintenance and stimula- tion of antigen-specific T cell lines have been described previously (1, 2). The mouse T cell HT2 line (25) was obtained from S . Strober (Stanford University, Stanford, CA) and the WEHI .13 164 line (26) from M . Palladino (Genentech, South San Francisco, CA). Media . Assay medium consisted of RPMI 1640 (J . R . Scientific Inc ., Woodland, CA) R. with 10 17o FCS (J . Scientific Inc .), 0 .05mM 2-ME (Sigma Chemical Co., St. Louis, MO), and 20 mM Hepes (Gibco Laboratories, Grand Island, NY). T cell growth medium con- sisted assay of medium containing 330 U/ml mouse rIL-2 . . IL-2 Antibodies mAbs against (S4B6), IL-5 (TRFK4 and TRFK5), and IFN (XMG1 .2) -'Y have been described previously (1, 2, 27) . The anti-IL-3 mAbs 8F8 .11 and 43D.11 (28), as well as the anti-GM-CSF antibodies 22E9 .11 and 35E10 .11, were obtained fromJ . S. Abrams (DNAX), I1B11 W. the anti-IL-4 antibody (29) was from E . Paul (National Institutes ofHealth, Bethesda, MD), and the 6B4 anti-IL-6 antibody (30) fromJ . van Snick (Ludwig Cancer In- stitute, Brussels, Belgium) . The hamster anti-mouse T3 mAb 145-2CII (31) was obtained . H .) . fromJ Bluestone (N . I . Polyclonal rabbit IgG antibodies specific for transforming growth factor S (TGFfl) were obtained from R & D Systems (Minneapolis, MN), and control rabbit IgG antibodies were obtained from R . Coffman (DNAX) . Cell Supernatants. Antigen-induced T cell supernatants were prepared by incubating x x T cell clones (2 105 cells/ml) with irradiated APC (5 10 6 cells/ml) and antigen in assay medium . Serum-free Con A-induced cell supernatants were prepared by incubating T cell clones (5 x 10 6 cells/ml) with Con A (5 Wg/ml ; Sigma Chemical Co.) in RPMI 1640 lacking phenol red (Gibco Laboratories) and containing 0 .05 mM 2-ME and 20 mM Hepes for 24 h . Cytokines. TGF was obtained from R & D Systems . Purified mouse P40 (32) was gener- ously provided by J . van Snick (Ludwig Cancer Institute) . Purified mouse IL-5 (33) was depleted ofknown Th2 cytokines and providedby R. Coffman, DNAX . Purified recombinant mouse cytokines were provided by: M . Howard, DNAX (IL-4) ; R . Kastelein, DNAX (IL-4); F. Lee, DNAX (IL-6 and IL-7) ; J . Schreurs, DNAX (IL-3 and GM-CSF) ; G . Zurawski, DNAX (human IL-Ia); Schering Research, Bloomfield, NJ (IL-2 and IFN-y) ; and M . Pal- ladino, Genentech (TNFa) . Cytokine Bioassays . The MTT assay (34), as modified (1), was used to measure IL-2, IL-4, and LT/TNFa using mAbs to establish monospecificity for IL-2 and IL-4 assays (2, 26) . WEHI .13 LT/TNFa was assayed on 164 cells essentially as described by Espevik and Nissen- x assay was carried Meyer (26), except that 5 103 target cells/well were used, and the out were calculated in the presence ofsaturating (>2 ng/ml) amounts ofIFN-y . Units of activity (1) . as described previously . Two-site sandwich ELISAs for IL-3, IL-5, and IFN-y were carried out Cytokine ELISAs as described previously (2, 27, 28) . The ELISA for GM-CSF was carried out essentially as mAbs and 22E9 . Intracellular IFN-y levels were described for the IL-5 assay, using the 35E10 1% X-100 Chemical Co.) and removing nuclei measured after lysing cells in Triton (Sigma centrifugation . by Column Chromatography. 1-2 .5-liter batches of serum-free Con A-induced D10 superna- tant were concentrated N10-fold using YM-5 membranes (Amicon Corp., Danvers, MA), passed through a 5-ml mannose-conjugated agarose column (E-Y Laboratories, San Mateo, FIORENTINO ET AL . 2083 CA), then further concentrated another three- to fivefold, for a total concentration of30-50- fold . This material was fractionated by HPLC on a hydroxylapatite-based support (Bio-Gel HPHT, Bio-Rad Laboratories, Richmond, CA) and on a gel filtration column (TSK-G 3000 SW, 60-cm length ; LKB Instruments, Gaithersburg, MD) as previously described (33) . After sequential chromatography on these two media, the fraction with CSIF activity is referred to as partially purified CSIF (ppCSIF) . Antigens. Keyhole limpet hemocyanin was obtained from Pacific Bio-Marine Laborato- ries, Inc . (Venice, CA) and used at a final concentration of 150 /Ag/ml . TNP-KLH (132 TNP residues per 106 daltons protein) was used at 1 p,g/ml (35) . Conalbumin (Sigma Chemical Co .) was used at 100 Kg/ml . Chicken red blood cells were obtained from PMK farms (Vacaville, CA) and used at 0.02% packed volume . Anticytokine Affinity Columns. Six affinity columns were prepared by coupling the mAbs XMG1 .2 anti-IFN-y, 22E9.11 anti-GM-CSF, S4B6 anti-IL-2, 8F8 .11 anti-IL-3, 11BIl anti- IL-4, and TRFK5 anti-IL-5 to Affi-Gel 10 (Bio-Rad Laboratories) . Each 1-2-ml column contained -10-20 mg of coupled antibody. CSIFAssay. Samples were diluted in 96-well flat-bottomed microtiter trays (Falcon Lab- ware, Oxnard, CA) in a volume of 0 .05 ml. Thl cells (5 x 104 cells/well), irradiated (2,500 rad) syngeneic spleen cells (5 x 105 cells/well), and antigen were added in a volume of 0.15 ml . 11BIl anti-IL-4 mAb (10,ug/ml) was added to the assay ifsamples were suspected ofcon- taining IL-4 . Levels of IFN-y and other cytokines in the 24-h supernatants were assayed . Omission of either antigen or APC resulted in negligible cytokine synthesis . When initially assayed, one batch of ppCSIF inhibited IFN-y production by -50% at a dilution of 1 :200 in an assay volume of0 .2 ml, and so the standard unit (SU) was defined by assigning a value of 1,000 SU/ml to this CSIF preparation . CSIF activity in unknown samples was quantitated by comparing levels of inhibition of IFN-y synthesis by the unknown with that of the stan- dard in each assay using a four-parameter curve-fitting program (Assay Zap ; Biosoft, Cam- bridge, UK) . Means t SD of triplicate cultures were determined . mRNA Measurement . Thl cells were stimulated with antigen plus APC that had been depleted of erythrocytes by centrifugation over Ficoll (Histopaque ; Sigma Chemical Co.) . At 8 and 12 h, aliquots of the cells were withdrawn and cytoplasmic RNA was isolated (36) . Dilutions ofRNA were bound to filters and hybridized with synthetic DNA probes that were "P-labeled by fill-in synthesis of overlapping DNA oligonucleotides as described previously (2) . The DNA probes corresponded to bases 900-989 for IFN--Y (2) and 1262-1363 for y-actin (37) . Filters were exposed to x-ray film and the autoradiographs quantitated by laser den- sitometry. Results Supernatantsfrom Th2 but not Thl Clones Inhibit Synthesis ofIFN--y by Thl Clones . Con A-induced supernatants from seven Th2 clones all inhibited IFN-'r production by HDK-1 cells (Fig. 1 A). This inhibitory activity was produced by D10 cells in re- sponse to Con A, antigen plus APC, or antiT3 plus APC (Fig . 1 B) . The inhibitory activity appeared to be produced only by stimulated cells, since Con A-stimulated CDC25 and CDC35 cells produced 866 and 708 U/ml ofinhibitory activity, respec- tively, whereas unstimulated cells produced <40 U/ml . We will call the factor respon- sible for this activity CSIF The decrease in IFN production could be due to a decrease in synthesis and/or - 'Y secretion . If a block in secretion was solely responsible for the reduced IFN-y super- natant levels, then the intracellular levels of IFN-y should be greatly elevated in the presence of CSIF When HDK-1 cells were stimulated with antigen plus APC, the maximum extracellular IFN-y levels h were and 1,100 ng/ml in the pres- at 24 250 ence and absence of CSIF, respectively. The corresponding maximum intracellular levels were 1 .0 and 4 .5 ng/ml . 4 Thl 208 Th2-DERIVED INHIBITOR OF CYTOKINE SYNTHESIS TH2 SUPERNATANT D10 (%) SUPERNATANT (%) FIGURE 1 . Th2 clones produce an inhibitor ofIFN-y synthesis by Thl cells . HDK-1 cells were stimulated with antigen plus APC in the presence of supernatants from seven Con A-induced /+g/ml) Th2 clones (A) or from D10 cells induced with Con A, Ag plus APC, or antiT3 mAb (4 .5 plus APC (B) . IFN--y secretion is expressed as a percentage of the levels produced in the absence of the Th2-derived inhibitor. Four Con A-induced Thl supernatants were depleted IFN--y and IL-2 (which of would interfere with the CSIF assay) by passage over monoclonal anticytokine affinity columns, and then tested for CSIF activity . TheseThl supernatants, even at a dilu- tion of 1 :10, did not inhibit IFN-7 production (Fig. 2 A) . The slightly increased IFN--y levels in cultures treated with the highest concentration ofM264-37 superna- FIGURE Thl clones secrete CSIF. Super- 2 . do not A-induced Thl clones were natants from four Con depleted of IL-2, IL-3, IFN-y, and GM-CSF, and for activity HDK-1 cells (A) . In- tested CSIF using hibition by ppCSIF is shown for comparison . The were in the presence of 14 same samples also tested ppCSIF (B) . The and lowerdashed lines U/ml upper the IFN-y the absence pres- represent levels of in and ence of CSIF, respectively. All values are expressed as a percentage of the IFN--y levels produced in the absence of both ppCSIF and Thl supernatants . MORENTINO ET AL 2085 tant could be accounted for by the residual IFN-y in this sample. Thus, Thl super- natants either do not contain CSIF, or they contain both CSIF and an inhibitor masks its activity. This second possibility is ruled out by the demonstration that that IL-2- and IFN-y-depleted Thl supernatants did not counteract CSIF activity (Fig . 2 B) . Taken together, these experiments demonstrate that Th2 but not Thl supernatants contain CSIF. CSIF Inhibits Production of Several Thl Cytokines. Since ppCSIF still contained GM- CSF and IL-3, the levels of these cytokines were further reduced by using anticytokine affinity columns before testing for CSIF's ability to inhibit the synthesis of several other Thl Cytokines . Production ofIFN-y was consistently inhibited in all Thl clones tested (e.g., Table 1) . The synthesis of IL-2, LT/TNF, IL-3, and GM-CSF was in- hibited in some cases but not others ; e.g ., IL-2 production was inhibited in HDK-1 cells but not in MD13-10 cells . When inhibition was observed, the extent of max- imum inhibition varied between different clones and cytokines, ranging from -60 to >95To . Even for a single clone, different cytokines were maximally inhibited at different concentrations ofCSIF In particular, the synthesis of LT/TNF was inhibited by relatively low concentrations of CSIF, but the extent of maximal inhibition was much less than that of IFN-y. CSIFInhibits Late but not Early Cytokine Synthesis. The kinetics of inhibition of cytokine synthesis were examined for antigen-stimulated HDK-1 cells (Fig . 3) . In general, the synthesis of all cytokines was not inhibited by CSIF at early times, e.g ., before early times 8.5 h (before 6.5 h for IL-2) . Two cytokines synthesized mainly at (GM- inhibited slightly or not all, whereas the synthesis of CSF and LT/TNF) were at mainly (IFN-y IL-3) was inhibited substan- cytokines produced at later times and exact of action cytokine synthesis tially. Differences in the kinetics CSIF and could for of thus account some of the variability inhibition seen in Table I . Inhibition of IFN- ,y Synthesis Occurs at the mRNA Level . HDK-1 cells were stimu- or absence ppCSIF, and total lated with antigen plus APC in the presence of cyto- RNA was from at and h after stimulation. RNA samples plasmic prepared cells 8 12 were tested for IFN- and y-actin mRNA levels by dot-blot hybridization . IFN-y ,y TABLE Effect of CSIF on the Synthesis of Thl Cytokines Percent of control synthesis level Cell line Cytokine 14 U/ml CSIF 42 U/ml CSIF 125 U/ml CSIF ,y t 1 HDK-1 IFN- 47 .6 t 0.8 29.1 .0 18 .6 t 0.7 71 ± t t IL-2 .7 2.3 59.6 3.9 40 .4 6.1 LT/TNF 41 1 .4 45 .1 t t .9 t 2.3 42 .8 1 .1 11 .4 t t 1 IL-3 63 .9 t 52 .6 6.8 38 .4 .9 8 t GM-CSF 86 .9 t .5 79.1 7.8 66 .8 t 9.3 MD13-10 IFN- ,y 4 .5 t .4 36 .0 t .9 27 3 23 .2 t 2.6 88 .2 5.1 t IL-2 t 109.3 13 .3 96 .0 t 16 .4 IL-3 60 .2 5.1 63 .0 t 6.5 t t 2.7 51 .0 GM-CSF 109.9 13 .0 119.9 t 7.4 .6 t t 97 19 .2 HDK-1 and MD13-10 cells were stimulated with antigen plus APC in the presence of varying dilutions of ppCSIF that had been depleted of GM-CSF and IL-3 . Cytokine levels in the 24-h supernatants are expressed as a percentage of the amount produced in the absence of CSIF . 2086 Th2-DERIVED INHIBITOR Thl OF CYTOKINE SYNTHESIS FIGURE 3 . Kinetics inhibition cytokine synthesis CSIF HDK-1 were of of by cells stimulated with antigen plus APC in the presence or absence of 100 U/ml ppCSIF that had been depleted of IL-3, IL-4, IL-5, and GM-CSF. Supernatants were collected at the indicated times and as- sayed for cytokine levels . mRNA levels were moderately reduced by CSIF at 8 h, and strongly reduced at 12 h over a range of 5-40 U/ml CSIF (Fig . 4) . In contrast, y-actin mRNA levels were not significantly reduced by CSIF at either 8 or 12 h, and possibly showed a slight enhancement at 12 h . The inhibition ofIFN-y protein levels in the 24-h super- natants by CSIF was slightly less than the inhibition of mRNA levels at 12 h . Thl but not Th2 Clones Respond to CSIF. ppCSIF was tested on seven Thl clones that were derived from three mouse strains and were specific for particulate antigens, FIGURE 4 . CSIF reduces mRNA levels for IFN-'Y but not actin . HDK-1 cells were stimulated with antigen plus APC in the presence or absence of ppCSIF. Cellswere harvested at 8 and 12 h for cytoplasmic mRNA isolation, and supernatants were taken at 24 h . y for IFN- protein determination . mRNA were by levels assessed hybrid- ization DNA probes, with synthetic and protein was measured by ELISA . All values are expressed as a percentage of levels produced in the absence of CSIF. FIORENTINO ET AL . MD13.2 A 0 MD13 .10 e HDK-1 L82-1 OK76.1 MUMC1.2 BA 0 N.6. MUMC2-B FIGURE 5. Thl but not Th2 u . clones respond to CSIF. Seven Thl clones were stimulated with antigen APC in plus the pres- ence of dilutions of ppCSIF, and secretion was measured IFN-,y (A). HDK-1 (Thl) and D10 1u (Th2) cellswere stimulated with antigen plus APC in the pres- ppCSIF(%) ence of dilutions ofppCSIF that had been depleted of IL-3, IL- 4, IL-5, and GM-CSF (B) . Levels of IL-2, IL-3, and IFN- (HDK-1) IL-3, IL-4, and ,y or IL-5 (D10) were assayed in the IL3 24-h All values " supernatants . u-a are expressed as apercentage of cytokine levels produced in the n the absence of ppCSIF . IL2 IL3 O IFNy ppCSIF(%) soluble proteins, or alloantigens . IFN- , y synthesis by all seven clones was inhibited by ppCSIF (Fig . 5 A) . Since some of the Th2 cytokines may have been present in the ppCSIF preparation, IL-3, IL-4, and IL-5 were depleted by passage over mAb affinity columns . The resulting CSIF preparation did not suppress the production of IL-3, IL-4, or IL-5 by the Th2 clone D10 (Fig. 5 B) . Two additional experiments with D10 yielded similar results, and in other experiments (not shown), CSIF did not inhibit the production of IL-5 by the Th2 clones MB2-1, CDC25, or CDC35, nor did it inhibit IL-4 synthesis by MB2-1 . Synthesis of other cytokines in response to antigen plus APC in these experiments was too low to reliably determine CSIF effects . In general, we have not seen convincing inhibition of the production of any cytokine by Th2 cells responding to antigen plus APCs . CSIF inhibited synthesis CSIF Action on Thl Cells May Be Indirect. in re- IFN-'y plus APC, but not ,y synthesis stimu- sponse to antigen plus APC or antiT3 IFN- a polystyrene surface, or by Con A (Table II) . Since only lated by anti-T3 bound to were inhibitable, CSIF may act indirectly on the the stimulations involving APC 2088 Th2-DERIVED INHIBITOR Thl CYTOKINE SYNTHESIS OF TABLE II CSIF Is Effective in Stimulations Involving APC Percent of control IFN--y synthesis Exp. Stimulation 5 U/ml CSIF 19 U/ml CSIF 74 U/ml CSIF 294 U/ml CSIF 1 Antigen plus spleen .6 f 72 12 .8 29 .1 t 3 .5 20 .5 t 6 .6 Anti-T3 plus spleen 88 .7 t 8 .0 78 .6 t 2 .8 68 .5 t 2 .6 Anti-T3 (coated) 95 .7 t 6 .7 103 .3 f 6 .7 94 .5 f 9 .3 Con A 102 .7 ± ± 4 .2 95 .4 5 .3 94 .7 t 5 .0 2 Antigen plus spleen 17 .5 t 5 .9 12 .8 1 .6 8 .2 ± .8 Anti-T3 ± 7 .2 plus spleen 54 .5 3 .1 45 .6 t 6 .0 42 .0 t HDK-1 cells were stimulated with antigen plus APC, anti-T3 mAb (9 ug/ml) plus APC, anti-T3 bound to 96-well flat-bottomed trays, or Con A (5 Ag/ml), each in the presence or absence of CSIF-containing 1310 supernatant . Trays were coated with anti-T3 by incubating with 50 ug/ml anti-T3 mAb in 50 141 PBS for 1 h at 37°C, following which, the plates were extensively washed . IFN-1' levels in 24-h supernatants are ex- pressed as a percentage of the amounts produced in the absence of CSIF, using the corresponding stimula- tion condition . T cells, via a cell type in the APC population . CSIF acts mainly at later times (Fig . 3), consistent with the hypothesis that CSIF is first modified into, or induces a sec- ondary factor that then inhibits IFN--y synthesis . Additional evidence for indirect was varying 10-fold T action obtained by the CSIF assay conditions . changes in cell and/or APC numbers did not affect the extent of inhibition of IFN-y synthesis by CSIF or the sensitivity ofdetection of CSIF (data not shown) . In contrast, saturating 16% instead amounts of CSIF inhibited IFN-y synthesis more effectively (to of 35 % ofcontrol IFN-y amounts) if the assay volume were reduced twofold . If the residual IFN-y levels represent secreted early before CSIF action, these the cytokine at times are consistent with inhibition synthesis being dependent on results the of IFN--y a soluble factor produced in the assay. In this case, such a factor would accumulate effective concentrations more rapidly in a smaller assay volume . to CSIF not Directly Inhibit Thl Proliferation . In contrast its effect on Thl cytokine Does to of Thl cells proliferate synthesis, CSIF had no measurable effect on the ability to in response to IL-2 . In the presence of exogenous IL-2 (500 ng/ml), antigen-stimulated proliferated 5.2-fold and by day 6, as assessed by HDK-1 cells by day 3, 42 .6-fold cell counting . The corresponding proliferation values in the presence of IL-2 plus 100 U/ml ppCSIF were 5.0-fold and 40 .0-fold . Similar results (not shown) were ob- tained using the MTT assay. However, in the absence of exogenous IL-2, ppCSIF proliferation of when at the time was sometimes able to decrease HDK-1 cells added not shown) ; presumably, inhibition was indirect of antigen stimulation (results this and was a result of CSIF's ability to reduce IL-2 production by HDK-1 cells (Table I, Fig . 3). Other Cytokines that Affect IFN--y Synthesis. In some but not all experiments, exoge- nous IL-4 and especially IL-2 increased the production ofIFN-y by Thl clones (data not shown) . Since IL-2 was synthesized in variable amounts by stimulated Thl cells, exogenous IL-2 and IL-4 might only be effective in experiments in which nonsaturating IL-2 in Fig . amounts of endogenous were synthesized . In the experiment shown 6 A, anti-IL-2 antibody inhibited the production of indicating that normal IFN-y, IFN--y synthesis was enhanced by IL-2 produced by the Thl cells responding to an- AL . 9 FIORENTINO ET 208 FIGURE 6 . Effects of IL-2, IL-4, andTGR S on IFN-y synthesis. HDK-1 cells were stimulated with antigen plus APC with var- ious combinations of cytokines and antibodies: IL-2 (500 ng/ml), IL-4 (50 ng/ml), TGF-0 (10 ng/ml), ppCSIF (125 U/ml), 54136 anti-IL-2 (10 1&g/ml), rabbit antiTGR0 (50jig/ml), and normal rabbit IgG (50 Kg/ml) (A and B). Titrations of these reagents showed that these levels were saturating . All cultures contained 111311 anti-IL-4 (10 pg/ml), except those in which IL-2 or IL-4 was added (A) . All values of IFN-y are expressed as a percentage the an- levels produced in the absence of ppCSIF, tibodies, and cytokines. tigen plus APC . When anti-IL-2 antibody was present, IL-4 enhanced IFN-y produc- tion slightly, and excess IL-2 (overcoming the effect ofthe anti-IL-2 antibody) could strongly enhance IFN-y levels . Although both IL-2 and IL-4 antagonized the effect of CSIF by increasing the amount of IFN-y secreted, CSIF was still active in the presence or absence of saturating amounts of IL-2 or IL-4 (Fig . 6 A) . Partially purified antiTGFf3 antibodies slightly enhanced the synthesis of IFN-y, whereas exogenous TGF-0 slightly inhibited IFN-y production (Fig. 6 B) . These results suggest that IFN-y synthesis can be inhibited by TGFf3, and that TGF-0 is produced in the assay under normal conditions . CSIF inhibited IFN-y produc- tion in the presence or absence of TGF-0 or antiTGF0, showing that CSIF is not TGF-i3 . Partial Purification and Biochemical Characterization of CSIF . CSIF-containing D10 supernatants were chromatographed on two media . On both columns, CSIF ac- tivity eluted in a single area. On the hydroxylapatite-based support, 95% of the ap- plied activity was recovered ; also shown are the elution positions of other proteins (Fig . 7 A) . On the gel filtration column, 63 % of the applied activity eluted broadly, with the fractions of high activity corresponding to apparent molecular masses of 27-50 kD (Fig . 7 B) . The width ofthe peak may be the result ofmicroheterogeneity similar to that observed with other cytokines (e .g ., IL-5 ; [33]) . CSIF could be se- quentially chromatographed on both columns in either order with no significant alteration of relative mobility or activity yield (data not shown) . These results sug- gest that CSIF may be a single entity, and we have not yet observed any evidence for directly synergistic interactions of CSIF with any other cytokine. ppCSIF (2,100 U .4) in 0.25 ml 10 mM phosphate buffer, pH 7 was carefully acidified to pH 2 with allowed to stand for 1 h at room temperature, neutralized with 1 M Hepes HCl, (sodium salt), and immediately assayed ; <5% of the activity remained . CSIFIs Distinctfrom Known Thl and Th2 Cytokines. Various blocking anticytokine monoclonal and polyclonal antibodies were tested for the ability to inhibit CSIF ac- 2090 Th2-DERIVED INHIBITOR OF Thl CYTOKINE SYNTHESIS FIGURE 7 . Column chromatography ofCSIFcontaining D10 supernatant . 10 ml of 27-fold con- centrated D10 supernatant applied to a Bio-Gel HPHT column in 10 mM sodium phosphate, pH 7.4, 0.12 mM CaC12, 0.01% Tween-20 at a flow rate of0.5 ml/min was eluted with a 60-min gradient to 0.35 M sodium phosphate, pH 7.4, 3.5 uM CaC12, 0.01% Tween-20. The elution positions of other proteins are indicated by horizontal bars (A) . Gel filtration of 0.2-ml 30-fold concentrated D10 supernatant on TSK-G 3000 SW was performed as previously described (33) (B). The elution positions of proteins of known molecular weight are marked by the arrows. In both cases the absorbance at 280 nm was monitored (curved line) and aliquots from the 1-min fractions were assayed for CSIF activity (shaded areas). tivity. CSIF activity was not reduced by mAbs specific for IL-2, IL-3, IL-4, and IL-6 (10 ttg/ml), IL-5 (2 Wg/ml), and GM-CSF (20 jig/ml), or by polyclonal anti- TGF antibodies (50 jig/ml) . Several cytokines were also tested in the CSIF assay and found to be inactive, either as agonists or antagonists . The synthesis of IFN-'y was enhanced or inhibited by <11% by IL-1 (101 U/ml), IL-5 (2 .4 ng/ml), IL-6 (25 ng/ml), or IL-7 (500 U/ml) . IFN-y synthesis was enhanced by IL-2 and IL-4 (Fig . 6), slightly enhanced by P40 (results not shown), and inhibited by TGF0 (Fig. 6) . Experiments using recombinant cytokines and mAbs, as well as knowledge of its biochemical properties and presence in Th2 but not Thl supernatants, distinguish CSIF from many of the known cytokines (Table III) . In general, we required at least two nonidentical properties before each cytokine was considered distinct from CSIF IL-7, According to these criteria, CSIF is not IL-1, IL-2, IL-3, IL-4, IL-5, IL-6, be GM-CSF, IFN--y, IT, TNF, TGFa, P40, or met-enkephalin, and is unlikely to soluble immune response suppressor (SIRS) (38) . Discussion and DTH responses are often mutu- Several lines of evidence show that antibody If these responses are mainly due to Th2 and Thl cells, ally exclusive (16-18, 39). Thl and Th2 cells reciprocally inhibit the growth and respectively, this suggests that the other cell type. Some of the known mechanisms include the inhibi- function of tion of growth of Th2 clones by IFN-y (19, 20), the inhibition of all IL-4 effects on B cells by IFN--y (40), and the inhibition of Thl proliferation by Th2-derived identified as an autocrine factor (21) . Although the activity in the last study was FIORENTINO ET AL . III TABLE with Known Cytokines Evidence for Nonidentity of CSIF has Biochemically Present in Absent in Ab blocks Cytokine CSIF activity similar to CSIF Cytokine Th2 sup Thl sup CSIF activity No No : separates by size IL-1 No Yes - IL-2 No No No - No No - No : separates on HPHT IL-3 Yes No No: separates on HPHT, size IL-4 Yes Yes No No No No - IL-5 Yes Yes Yes Yes No No No : separates by size IL-6 No No: not acid labile IL-7 - - - No - No : not acid labile GM-CSF Yes No - - - IFN No No LT No No - - - No : not acid labile TNF Yes No - - - - Yes (partial) No : not acid labile TGF-0 No No No : not acid labile P40 Yes - - Yes No - - No : separates by size Met-enk - - - - No : separates by size SIRS growth inhibitor for Th cells, the results need to be reinterpreted in light of recent information (1, 3, 5). The Th2-derived growth inhibitory activity inhibited the growth of a Thl clone "1,000-fold more effectively than D10 growth (21), but it appears to be distinct from CSIF since the two activities have different apparent molecular masses, and ppCSIF does not inhibit proliferation of Thl clones . In addition to IFN-y and the D10-derived proliferation inhibitor, the CSIF identified in this study could be another ofthe crossinhibitory regulators ofThl andTh2 cells. Thl function should be inhibited by CSIF, since IFN-y is an important mediator of several Thl functions (7, 12, 41) . There may also be an indirect enhancement of Th2 growth, since the decrease in IFN-y levels should reduce inhibitory effects on Th2 growth (19, 20). CSIF inhibition of IL-2 but not IL-4 synthesis may also slightly reduce the relative growth rates of Thl cells, since Th2 cells respond better than Thl cells to the proliferative effects of IL-4 (3) . We have suggested previously (3) that these crossinhibitory cytokine effects may be important early in immune responses, whereas the production of specific Ig isotypes may counterbalance this effect later in the response . CSIF could also have a more subtle role in the regulation of immune responses. In vitro, Thl clones are often very poor B cell helpers (8, 9), unless the cytokine levels are adjusted by decreasing IFN-y and increasing IL-2 levels (7) . Since CSIF inhibits IFN-y production, often more effectively than IL-2 production, and freshly isolated Thl clones tend to produce more IL-2 than long-term Thl clones (2), CSIF may modify the functions ofThl cells so that they become more effectiveB cell helpers. If true, this effect would only be seen in responses in which at least some Th2 cells Thl were activated . This may explain the apparently contradictory observations that B DTH cells can be good cell helpers under some circumstances, but that a strong response (presumably due to strong Thl activation) is often unaccompanied by significant antibody synthesis . CSIF appears to selectively inhibit Thl cytokine synthesis, rather than generally Thl CYTOKINE SYNTHESIS 2092 Th2-DERIVED INHIBITOR OF -y-actin growth or viability. CSIF treatment did not reduce the levels of inhibit cell IL-2 . Although mRNA, nor did it inhibit proliferation in response to exogenous IFN- production (42, and Fig. 6), and CSIF often inhibits IL-2 produc- IL-2 enhances -y presence in Fig. 6 also show that CSIF exerts its effect in either the tion, the results IL-2 . This indicates that CSIF does not inhibit IFN-'y synthesis simply or absence of (43, IL-2 production . Although TGF/3 also inhibits IFN-y synthesis by inhibiting saturating Fig . CSIF was able to inhibit IFN--y synthesis more effectively than and 6), TGF0 . TGF0 . This indicates that CSIF does not act only by inducing amounts of CSIF of CSIF on Thl cytokine synthesis may be indirect, since (a) The action .y A ; (b) IFN- IFN- synthesis in response to antigen plus APC but not Con inhibits -y volume of the synthesis is inhibited only at later times after stimulation ; and (c) the Two possible mech- critically important to the degree of inhibition obtained . assay is by a cell in the induction by CSIF of an effector molecule synthesized anisms are form by the APC APC population, or the modification of CSIF to an active the pretreated Preliminary experiments (unpublished data) show that APC population . induce Thl are only marginally reduced in their ability to subsequently with CSIF .y primarily on the APC popu- cells to produce IFN- . This indicates that, if CSF acts of T cells . the effect is mostly reversible, or requires the presence lation, effect of CSIF is mediated at Analysis of IFN- mRNA levels showed that the -y measurements. Possible mRNA level, in agreement with intracellular protein the degradation, or altered include reduction of transcription, accelerated mechanisms reduced at 8 h, and strongly mRNA processing . IFN--y mRNA levels were slightly , y showing that CSIF inhibition reduced by 12 h, consistent with IFN- secretion data ,y of IFN- production may also is detected only at late times . The delayed inhibition inhibits cytokine synthesis ; the re- explain why saturating CSIF never completely of cytokines synthesized at early sidual cytokine levels may represent the amounts times after stimulation . displayed by the majority of long-term Although the Thl and Th2 phenotypes are patterns exist among normal Th cells mouse Th clones, other cytokine secretion N. E., J . H . Schumacher, T. A . T. short-term clones (2, 13, 20, and Street, and Leverah, and T. R. Mosmann, manuscript Fong, H . Bass, D . F. Fiorentino, J. A . phenotypes may be precursors of Thl and submitted for publication) . These other of such precursors are not yet known Th2 cells, and the details of the functions Th types, production and effects in these other It will be important to assess CSIF Thl and Th2 cytokines . particularly in cells secreting both that Th2 cells of CSIF is complicated by the fact The interpretation of the role . This situation partially counteracts the effects of CSIF also produce IL-4, which ,y these two cytokines of IL-2 and IFN- by Thl cells, since is analogous to the production pairs ofcytokines Th2 cells (19, 20) . It is possible that these have opposing effects on that the local clear- different ratios under different conditions, or are produced in to temporal differences in the of each cytokine are very different, leading ance rates awaits improved methods for detecting effects observed . Resolution of these questions and manipulating cytokine levels in vivo. Summary is secreted by Th2 clones in response A cytokine synthesis inhibitory factor (CSIF) from Con A-induced but is absent in supernatants to ConA or antigen stimulation, FIORENTINO ET AL 2093 Thl clones. CSIF can inhibit the production ofIL-2, IL-3, lymphotoxin (LT)/TNF, IFN--y, and granulocyte-macrophage CSF (GM-CSF) Thl responding by cells to antigen significantly Trans- and APC, but Th2 cytokine synthesis is not affected . forming S (TGFj3) inhibits IFN-y production, effec- growth factor also although less tively than CSIF, whereas IL-2 and IL-4 partially antagonize activity of the CSIF CSIF inhibition is complete, since early cytokine synthesis of cytokine synthesis not (before 8 h) not significantly affected, whereas later synthesis is strongly inhibited . is In the presence ofCSIF, IFN-y mRNA levels are reduced slightly 8, and strongly at at h after stimulation . Inhibition of cytokine expression by CSIF is not due 12 to a general reduction in Thl cell viability, since actin mRNA levels were not reduced, and proliferation of antigen-stimulated cells in response to IL-2, was unaffected Biochemical characterization, mAbs, recombinant or purified cytokines showed and CSIF from IL-1, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, GM- that is distinct IFN-y, CSF, TNF, and P40 . The role CSIF in crossregulation TGF,3, IT, potential of of Thl and Th2 responses is discussed . We thankJ. Abrams, R. Coffman, M . Howard, R . Kastelein, F Lee, Palladino, J. Schreurs, M. J . van Snick, and Zurawski for providing valuable cytokine and antibody reagents . We G . J . providing technical advice and F Vega for synthesizing oligonucle- thank Schumacher for otides . Note added in proof The identity of CSIF as a novel cytokine has recently been confirmed by the isolation of a cDNA clone that expresses CSIF activity when transfected into COS cells (Moore, K . W. et al ., manuscript in preparation) . mAbs specific for CSIF react with both the recombinant protein and the 1310-derived factor (Mosmann, T. R . et al ., manuscript in of preparation) . The sequence the cDNA clone is different from all previously known cytokines . At the time of submission of this cDNA for clone sequence publication, we will request the Nomenclature Committee of the International Union of Immunological Societies to assign an IL designation for factor. this Received for publication 19June 1989 and in revisedform 28 August 1989. References 1 . Mosmann, T R ., H . Cherwinski, M . W. Bond, M . A . Giedlin, and R . L . Coffman . 1986 . Two types ofmurine helper T cell clone . I . Definition according to profiles of lym- phokine activities and secreted proteins . . Immunol. 136:2348 . 2 . Cherwinski, H . M., J . H . Schumacher, K . D. Brown, and T. R . Mosmann . 1987 . Two types of mouse helper T cell clone. III . Further differences in lymphokine synthesis be- tween Thl and Th2 clones revealed by RNA hybridization, functionally monospecific bioassays, and monoclonal antibodies. . Exp. Med. 166 :1229 . 3 . Mosmann, T R ., and R . L . Coffman . 1989. Heterogeneity of cytokine secretion pat- terns and functions of helper T cells . Adv. Immunol. 46 :11 . 4 . Brown, K . D ., S . M . Zurawski, T R . Mosmann, and G . Zurawski . 1989 . A family of small inducible proteins secreted by leukocytes are members of a new superfamily that includes leukocyte and fibroblast-derived inflammatory agents, growth factors, and indi- cators of various activation processes. . Immunol. 142:679 . 5 . Kim, J ., A . Woods, E . Becker-Dunn, and K. Bottomly. 1985 . Distinct functional pheno- types of cloned la-restricted helper cells . . Exp. Med. 162 :188 . 6 . Cher, D. J ., and T. R . Mosmann . . 1987 Two types of murine helper T cell clone . II . Delayed-type hypersensitivity is mediated by Thl clones . . Immunol. 138:3688 2094 Th2-DERIVED INHIBITOR OF Thl CYTOKINE SYNTHESIS 7 . L ., B . W. D . A . J. Coffman, R . Seymour, Lebman, D. D. Hiraki, A . Christiansen, B . Shrader, H . T. M . Cherwinski, H . F. Savelkoul, F. D. Finkelman, M . W. Bond, and R . Mosmann . 1988 . Th e role of helper T cell products in B differentiation and mouse cell isotype regulation. Immunol. Rev. 102 . :5 8 . Killar, L ., G. MacDonald, A . K . J . West, Woods, and Bottomly. 1987 . Cloned, la-restricted T cells that do not produce interleukin 4(IL-4)/B cell 1(BSF to stimulatory factor 1) fail help antigen-specific B cells . Immunol. 138 :1674 . 9 . Boom, W. H ., D. Liano, and A . K . Abbas 1988 . Heterogeneit helper/inducer T . y of lymphocytes. II . Effects of interleukin 4- and interleukin T cell on 2-producing clones resting B lymphocytes . J . . Med. 167 :1352 Exp . 10 . Stevens, T. L ., A. Bossie, V. T. M . Sanders, R . Fernandez-Botran, R. L . Coffman, R . Mosmann, and E . S . Vitetta. 1988. Subset s of antigen-specific helper T cells regulate isotype secretion by antigen-specific B 334 cells . Nature (Loud.). :255 . 11 . Giedlin, M . A ., B . M . Longenecker, T. . e and R . Mosmann 1986 . Murin T-cell clones specific for chicken erythrocyte . 97 alloantigens . Cell Immunol. :357 . . Stout, . D., and . Bottomly. 1989. 12 R K Antigen-specific activation of effector macrophages by IFNy-producing (Thl) T cell clones Failure of T . IL4-producing (TH2) cell clones to activate effector function in macrophages . Immunol. 142:760 . 13 . Swain, S . L., D. T. McKenzie, A . D. Weinberg, and W. Hancock. 1988 . Characteriza- tion of T helper 1 and cell subsets in normal mice . Helper T cells responsible for IL-4 and IL-5 production are present as precursors that require priming before they develop into lymphokine-secreting cells . Immunol. 141 :3445 . 14 . Budd, R . C .,J . C . Cerottini, and H . R . MacDonald 1987 . Selectively increased produc- tion of interferon-gamma by subsets of Lyt-2' and L3T4' T cells identified by expres- sion of Pgp-1 . Immunol. 138:3583 . 15 . Firestein, G . S ., W. D. Roeder, J . A . Laxer, K . S. Townsend, C . T. Weaver, J . T. Hom, J . Linton, B . E . Torbett, and A. L . Glasebrook . 1989 . A new murine CD4' T cell subset with an unrestricted cytokine profile . . Immunol. 143 :518 . 16 . Parish, C . R . 1972 . The relationship between humoral and cell-mediated immunity. Trans- plant. Rev. 13 :35 . 17 . Katsura, Y. 1977 . Cell-mediated and humoral immune responses in mice . III . Dynamic balance between delayed-type hypersensitivity and antibody response . Immunology. 32 :227 . 18 . Heinzel, F. P, M . D. Sadick, B . J . Holaday, R . L. Coffman, and R. M . Locksley. 1989 . Reciprocal expression of interferon y or interleukin 4 during the resolution or progres- sion of murine leishmaniasis. Evidence for expansion of distinct helper T cell subsets . J Exp. Med. 169 :59 . 19 . Fernandez-Botran, R ., V. M . Sanders, T. R . Mosmann, and E . S . Vitetta . 1988 . Lymphokine-mediated regulation of the proliferative response of clones of T helper 1 and helper 2 cells . . Exp. Med. 168 :543 . 20 . Gajewski, T. F., and F. W. Fitch . 1988. Anti-proliferative effect of IFN-gamma in im- mune regulation . I . IFN-gamma inhibits the proliferation of Th2 but not Thl murine helper T lymphocyte clones . . Immunol. 140:4245 . . J . 21 Horowitz, B., J . Kaye, P. J. Conrad, M . E. Katz, and C . A . Janeway, Jr. 1986 . Auto- crine of line T cells . Proc . :1886 growth inhibition a cloned ofhelper . Nail. Acad. Sci USA . 83 . C . A . . 22 . Kaye, J ., S . Porcelli, J . Tite, G . Jones, and Janeway, Jr 1983 . Both a monoclonal antibody and antisera specific for determinants unique to individual cloned helper lines for T cell can substitute antigen and antigen-presenting cells in the activation of cells . J . 158 Exp. Med :836 . 23 . Tony, H . P, and C . Parker. 1985 . Major histocompatibility complex-restricted, poly- D. clonal B cell responses resulting from helper T cell recognition of antiimmunoglobulin presented by small B lymphocytes . . Exp. Med . 161 :223 . 24 . Nabel, G ., J . S . Greenberger, M . A . Sakakeeny, and H . Cantor. 1981 . Multiple biologic FIORENTINO ET AL . 2095 activities of a cloned inducer T-cell population . Proc. Nail. Acad. Sci. USA . 78 :1157 . 25 . Watson, J . 1979 . Continuous proliferation of murine antigen-specific helper T lympho- cytes in culture . . Exp. Med. 150 :1510 . 26 . Espevik, T., and J . Nissen-Meyer . 1986. A highly line, sensitive cell WEHI 164 clone 13, for measuring cytotoxic factor/tumor necrosis factor from human monocytes . j Im- munol. Methods. 95 :99 . 27 . Schumacher, J . H ., A . O'Garra, B . Shrader, A. van Kimmenade, M . W. Bond, T R . Mosmann, and R . L . Coffman . 1988 . Th e characterization offour monoclonal antibodies specific for mouse IL5 and development of mouse and human IL5 enzyme-linked irri- munosorbent assays . . Immunol. 141 :1576 . 28 . Abrams, J . S ., and M . K . Pearce. 1988 . Development of rat anti-mouse interleukin 3 monoclonal antibodies which neutralize bioactivity in vitro . . Immunol. 140 :131 . 29 . Ohara, J ., and W. E. Paul . 1985 . Production of a monoclonal antibody to and molecular characterization of B-cell stimulatory factor-1 . Nature (Loud.). 315 :333 . 30 . Vink, A., P. G . Coulie, P Wauters, R . P. Nordan, andJ . van Snick . 1988 . B cell growth and differentiation activity of interleukin-HPI and related murine plasmacytoma growth factors . Synergy with interleukin 1 . Eur. Immunol. . 18 :607 31 . Bluestone, J ., D . Pardoll, S. Sharrow, and B. J . Fowlkes . 1987 . Characterization of mu- rine thymocytes with CD3-associated Tcell receptor structures . Nature (Land.). 326 :82 . 32. Uyttenhove, C ., R . J . Simpson, and J . van Snick . 1988. Functional and structural char- acterization of P40, a mouse glycoprotein with Tcell growth factor activity. Proc . Natl. Acad. Sci. USA . 85:6934 . 33. Bond, M . W., B. Shrader, T. R . Mosmann, and R . L. Coffman . 1987 . A mouse T cell product that preferentially enhances IgA production . II . Physicochemical characteriza- tion . Immunol. 139 :3691 . J. 34 Mosmann, T. R . 1983 . Rapid colorimetric assay for cellular growth and survival : appli- cation to proliferation and cytotoxicity assays. Immunol. Methods. 65 :55 . J. 35 . Kiefer, H . 1979 . Chemical modification of proteins, and haptens solid supports . In Im- munological Methods, Vol . 1 . I . Lefkovits, and B . Pernis, editors . Academic Press, New York. 36 . White, B. A ., and F. C . Bancroft . 1982 . Cytoplasmi c dot hybridization . Simple analysis of relative mRNA levels in multiple small cell or tissue samples . Biol. Chem. 257 :8569 . 37 . Leader, D. P, I . Gall, and H . Lehrach. 1985 . The structure of a cloned mouse gamma- actin processed pseudogene . Gene. (Amst.). 36 :369 . 38 . Webb, D. R ., K . Mason, G . Semenuk, T M . Aune, and W. C . Pierce . 1985 . Purificatio n and analysis of isoforms of soluble immune response suppressor (SIRS) . . Immunol. :3238 . 39 . Crowle, A . J ., and C . C . Hu . 1966. Spli t tolerance affecting delayed hypersensitivity and induced in mice by pre-immunization with protein antigens in solution. Clin. Exp. Im- munol. 1:323 . 40 . Mosmann, T. R ., and A . Zlotnik. 1989 . Multiple functions IL4 of and its role in immune regulation. In Growth Factors, Differentiation Factors, and Cytokines. A . Habenicht, editor. Springer-Verlag, Heidelberg . In press . 41 . Fong, T A. T, and T. R . Mosmann . 1989 . The role of IFN-y in Delayed Type Hyper- sensitivity Thl mediated by clones . . Immunol. In press . 42 . Kelso, A ., H . R . MacDonald, K . A . Smith, J . C . Cerottini, and K. T. Brunner. . Interleuki n 2 enhancement oflymphokine secretion by T lymphocytes : analysis of estab- lished clones and primary limiting dilution microcultures . . Immunol. 132 :2932 . 43 . Espevik, T., I . Figari, M . R . Shalaby, G. Lackides, G. Lewis, H . Shepard, and M . A . Palladino . 1987 . Inhibition of cytokine production by cyclosporin A and transforming growth factor S. . Exp. Med. 166 :571 .

Journal

The Journal of Experimental MedicinePubmed Central

Published: Dec 1, 1989

There are no references for this article.