Anti-MIF antibodies

Abstract

The present invention relates to methods for screening molecules and methods to detect protein-protein interactions and means used therein. More specifically, the present invention relates to methods for screening candidate drugs for treating or detecting MIF (macrophage migration inhibitor factor) related diseases. In certain aspects, the present invention involves detecting MIF/Jab1 (c-Jun activation domain binding protein) interactions as a basis for modulating cellular regulatory pathways and for identifying candidate drugs for MIF-related diseases. The invention also provides methods for the identification of molecules which dissociate or prevent interaction or binding between MIF and Jab1.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

This application is drugscreeningsued used for coprecipitation. Thus, MIF specifically binds to p38^Jab1 both in vivo and in vitro.

FIG. 2 shows that MIF specifically interacts with Jab1. a, Interaction in vitro. Complexes of rMIF and biotin-Jab1 expressed in TNT lysates were precipitated by MIF-specific antibody, and biotinylated proteins detected by Western blotting (left panel). Right panel, expression control. b, Interaction of immobilised MIF with Jab1. Streptavidin bead-immobilised TNT-translated biotin-MIF-EGFP or unlabelled control was incubated with TNT-translated Jab1, complexes isolated, and Jab1 detected by immunoblotting. c, Interaction of MIF and Jab1 in vivo. Endogenous Jab1/MIF complexes were immunoprecipitated from 293T cell lysates with anti-Jab1 antibody and MIF detected by immunoblotting. Precipitation with anti-TNF antibody or beads alone served as controls.

EXAMPLE 3

Detecting Interaction of MIF with Jab Using Modulation of AP-1-Dependent Reporter Gene Activity.

Transcriptional coactivator function of Jab1 is due to enhancement of AP-1 transcriptional activity. It was tested whether MIF, by binding to Jab1, could modulate this activity. AP-1 dependent reporter gene activity in 293T cells transiently expressing the collagenase TRE luciferase reporter was measured.

The effects of MIF were assessed by measuring AP-1-dependent reporter gene activity in transfected 293T cells using the 5×TRE-luciferase reporter. Recombinant MIF, in a concentration-dependent manner, fully reversed TNF-α-induced AP-1-dependent reporter gene activity and, at a concentration of 1 μM, fully inhibited potentiation of AP-1 reporter gene activity induced by p38^Jab1 which had been transiently cotransfected into cells as full-length cDNA together with the addition of rMIF. Inhibition of AP-1 activity by MIF was not a secondary effect of MIF-mediated altered cell growth. Significant potentiation by p38^Jab1 (2-fold) was observed in the absence of transfected c-Jun, indicating that endogenous c-Jun levels were sufficient for potentiation by p38^Jab1 to occur. Inhibition by MIF was already significant at 1 pM (˜30%) and was complete when 10 nM-1 μM rMIF were applied.

FIG. 3 shows that MIF inhibits stimulated AP-1 activity. a, MIF inhibits Jab1-mediated activation of AP-1 reporter gene activity in 293T cells. Transfections with pCI-neo-Jab1 or control vector are compared with or without (−) rMIF. b, MIF inhibits TNF-induced AP-1 activity. Same as a, but with TNF instead of Jab1 induction. Stimulation by PMA was a control. Data represent means±SD of four determinations and are representative of three experiments. c, MIF does not inhibit stimulated NFκB reporter gene activity. TNF-induced NFKB from antisense MIF macrophages with reduced content of endogenous MIF (+as-MIF) is compared with activities of control (+control) and non-transfected (−) cells. The mean±SD of 3 measurements is given. LUC, relative luciferase activity.

FIG. 4 shows that MIF inhibits potentiation of AP-1 reporter gene activity (collagenase 5×12-o-tetradecanol phorbol acetate response element (TRE) promoter) by Jab1 and UV stress. 293T cells were transfected with the reporter plasmids and, where indicated, cotransfected with pCl-neo-Jab1 or empty control vector and incubated with the indicated concentrations of rMIF or buffer control for 18 h. a, Western blotting control experiment for FIG. 3a displaying analysis of Jab1 overexpression and control EGFP cotransfection by anti-Jab1 and antiGFP-Western. The co-transfected EGFP construct (pN3-EGFP, Clontech) was added at 0.05 μg per incubation. In addition, EGFP-positive cells were counted at the end of the incubation. Transfection efficiencies judged by this analysis was 45-50% for FIGS. 3a and 35-45% for FIG. 3b. An effect of MIF itself on 293T cell growth under the conditions of the assay was excluded by additional cell counts. Wells after the end of the incubation period contained 2.3-2.5×10 5 cells (FIGS. 3a) and 3.8-4.1×10 4 cells (FIG. 3b). Results are expressed as relative luciferase (LUC) activity. b, MIF inhibits UV stress-induced AP-1 transcriptional activity. Same as FIG. 3b, but with UV light induction (Stratalinker, 3.6 Joule/cm 2 , 20 min+90 min) instead of TNF treatment. Data represent the mean±SD of 4 measurements. c, MIF does not interfere with stimulated NFkappaB activity. Recombinant MIF (rMIF) does not inhibit TNF-induced NFkappaB reporter gene activity in 293T cells. Cells were transfected with the reporter plasmids and incubations with rMIF (40 h) performed at a concentration of 1 μM. TNF was added at a concentration of 10 ng/ml (6 h). Results are relative luciferase (LUC) activities and represent the mean±SD of 3 determinations.

MIF also inhibits TNF- and UV-stress-induced AP-1 transcriptional activity. Neither exogenously added nor endogenously MIF interferes with NFκB activity. Recombinant MIF (1 μM) has no effect on TNF induced NFκB reporter gene activity in 293T cells.

EXAMPLE 4

Detecting Interaction of MIF with Jab1 Using Electromobility-Shift Assay (EMSA).

A transcriptional co-activator function by p38^Jab1 is due to enhancement of the binding of c-Jun to the AP-1 site. Thus, it was assumed that MIF, by binding to p38^Jab1, could modulate this activity. Such an effect on DNA binding was assessed by performing electromobility-shift assay (EMSA) on nuclear extracts from 293 cells that had been transfected with p38^Jab1 and/or c-Jun and that had been incubated in the presence or absence of rMIF, using the 12-o-tetradecanol phorbol acetate response element (TRE). p38^Jab1 promoted binding of c-Jun to the TRE. When MIF was incubated with cells overexpressing both p38^Jab1 and c-Jun, a marked inhibition (2-fold) of the activatory effect of p38^Jab1 on the binding of c-Jun to the AP-1 site was seen. By contrast, when only one or none of the transcriptional factors was overexpressed, MIF exhibited a small activatory effect, but this effect was not seen in EMSA from programmed reticulocyte lysates. EMSA from nuclear extracts of the HeLa Tet-off cell line HtTA, which stably expresses the tetracycline-controlled transactivator system and into which the human MIF cDNA was stably transfected to overexpress the MIF protein 5-6-fold over endogenous MIF following removal of doxycyclin, confirmed that MIF can significantly inhibit AP-1-dependent DNA shifts. The degree of inhibition in this system is probably even more marked, because the band intensity in the doxycyclin-repressed incubation likely represents a DNA shift that is already inhibited by the significant concentrations of endogenous MIF (˜100 fg/cell).

EXAMPLE 5

MIF and p38^Jab1 Regulate Cell Signalling Pathways

To test whether p38^Jab1 and MIF could act to more broadly regulate cell signalling pathways, regulation of AP-1 activity by mechanisms upstream of direct transcriptional control was investigated. It was considered that p38^Jab1 may modulate JNK activity. Transient transfections of 293 cells with p38^Jab1 revealed that immuno-precipitated kinase activity on GST-c-Jun(1-79) was markedly stimulated by p38^Jab1. Enhancement by p38^Jab1 was 2-3-fold and was stronger than N-terminal phosphorylation of GST-c-Jun(1-79) following stimulation with standard concentrations of TNF-α. It was investigated whether MIF would also inhibit this p38^Jab1-mediated effect. Recombinant MIF, in a dose-dependent fashion, completely reversed enhancement of JNK activity by p38^Jab1. Thus, recombinant MIF reversed enhancement of JNK activity by Jab1. 200 nM MIF or more led to complete inhibition. Similarly, TNF-α-induced JNK activity was suppressed to baseline levels following treatment with rMIF. The MIF-p38^Jab1 interaction should occur independently of whether MIF is endogenously overexpressed or exogenously added to cells. This was verified tested by transient transfection experiments, where TNF-α- or p38^Jab1-induced JNK activity was measured in 293 cells that had been co-tranfected with MIF-EGFP or EGFP alone. Similar to the exogenously added rMIF, the endogenously overexpressed MIF fully suppressed both TNF-α- and p38^Jab1-mediated activation of JNK.

FIG. 5 shows that Jab1 enhances JNK activity and phospho-c-Jun levels and MIF inhibits these effects. a, rMIF inhibits Jab1-mediated enhancement of JNK activity. Left, comparison of phosphorylation of GST-c-Jun(1-79) by JNK immunoprecipitates from 293T cells transfected with pCI-neo-Jab1 versus empty vector and treated with rMIF or buffer. Relative activation indicates band intensities. Control, TNF induction. Right, as left but cotransfection of antisense-Jab1 plasmid where indicated (control, immunoblots of JNK levels; transfection efficiences: 40-50%; variances: ±10%; induction of JNK by Jab1: 1,6-fold±SD of 0.08; n=3). b, Inhibition by MIF of Jab1-induced phosphorylation of endogenous c-Jun. Incubations as in 4a but immunoblots of phospho-c-Jun analysed. Betaactin immunoblots served as control. C, As b, but phosphorylation of c-Jun induced by TNF or UV stress.

FIG. 6 shows modulation of JNK by Jab1 and MIF. a, Inhibition of TNF-stimulated JNK activity by rMIF. 293T cells were transfected with FLAG-tagged JNK-coding vector and incubated with TNF (20 ng/ml) and rMIF or buffer control as indicated for 48 h. JNK immunoprecipitates were used to phosphorylate GST-c-Jun(1-79) and relative band intensities (indicated as relative activation) estimated by phosphoimage scanning. Control Western blots were analysed by anti-FLAG-antibody. Variances were generally in the range of ±10%. b, Enhancement of the binding of JNK to c-Jun beads by Jab1 but absence of interference by MIF. Possible effects of Jab1 and MIF on the binding avidity of c-Jun to JNK were investigated by analysing the effect of Jab1 overexpression and MIF coincubation on the binding of endogenous JNK to GST-c-Jun(1-79) beads in lysates from 293T cells. Transfection of cells with Jab1 led to an enhanced binding of JNK to c-Jun beads compared to control-transfected cells. Addition of rMIF did not influence this Jab1 effect. 293T cells were transfected with Jab1 vector as in FIG. 4a of the manuscript and cell lysates incubated with GSH agarose-bound GST-c-Jun(1-79). Eluted complexes were incubated and precipitated with anti-JNK antibody in the presence or absence of rMIF and bound proteins analysed by anti-JNK (internal control), anti-GST, and anti-c-Jun Western blotting.

EXAMPLE 6

To investigate whether MIF could also negatively regulate Jab1 action with regard to non-AP-1-related activities, the effect of MIF on the cell cycle inhibitor p27^Kip1 which binds to Jab1 and whose degradation is instigated by Jab1 was studied. Contrary to Jab1, which suppresses p27^Kip1 levels, MIF, in a dose-dependent manner, induced p27^Kip1 levels, as indicated by immunoblots prepared from lysates of proliferating NIH 3T3 fibroblasts and Jurkat T cells. This induction was Jab1-dependent as p27^Kip1 levels did not rise in response to rMIF when cells were transfected with antisense Jab1 construct which almost completely suppressed endogenous Jab1 protein. p27^Kip1 induces G1 growth arrest and Jab1 can rescue serum-starved fibroblasts from growth arrest. It was found that overexpressed MIF-EGFP or exogenous rMIF, in a concentration-dependent manner, inhibited both Jab1-induced reduction of serum dependence of fibroblasts and growth of proliferating fibroblasts. As MIF did not directly bind to p27^Kip1, did not stimulate p27^Kip1 mRNA or protein synthesis, and as Jab1 promotes the proteasome-dependent degradation of p27^Kip1, effects of MIF in the context of p27^Kip1 degradation were analysed. p27^Kip1 levels, when measured by immunoprecipitation from synchronised, pulse-chase-labelled fibroblasts were higher in the presence of added rMIF. The stabilising effect of MIF on p27^Kip1 levels was not further enhanced by addition of the proteasome inhibitor LLnL. By contrast, enhancement of p27^Kip1 levels by MIF alone was even more pronounced than by LLnL alone. At the same time, addition of MIF slightly increased p27^Kip1 levels in the presence of a protein synthesis blocker. As coimmunoprecipitation studies indicated that MIF partially interfered with p27^Kip1/Jab1 complex formation, these data show that MIF-mediated effects mirror p27^Kip1-mediated growth arrest and that this occurs via inhibition of Jab1-dependent degradation of p27^Kip1.

FIG. 7 shows that MIF stabilises p27^Kip1 protein and inhibits fibroblast growth in a Jab1-dependent manner. a, MIF induces p27^Kip1 expression in fibroblasts (p27^Kip1 immunoblots and c-Jun control blots). b, p27^Kip1 induction by MIF is Jab1-dependent. Left, As a, but additional transfection of antisense Jab1 or control plasmid (efficiency control, pEGFP cotransfection. Right, The antisense construct reduces Jab1 in fibroblasts. c, Inhibition by MIF of Jab1-mediated reduction of serum dependence of fibroblasts. Proliferation of GFP-expressing cells is analysed in Jab1- and MIF-EFGP-overexpressing versus control vector-treated cells via BrdU incorporation. Data are means±SD of four determinations. d, MIF inhibits proteasome-mediated degradation of p27^Kip1. Left, MIF reduces degradation of p27^Kip1 in pulse-chase labelled fibroblasts. [35S]p27^Kip1 levels shown from immunoprecipitations from rMIF-(+) versus buffer-treated cells. Right, MIF effect is proteasome-linked. As a, but incubations followed by treatment with DMSO, cycloheximide (CHX), or LLnL.

FIG. 8 shows the mechanism of p27 Kip1 induction by MIF. a, Effect of rMIF on p27^Kip1 mRNA levels as analysed by Northern blotting. 1×10 7 NIH 3T3 fibroblasts were incubated in the presence or absence of rMIF (as indicated) for 40 h, cells washed in ice-cold PBS, and lysed in Trizol reagent (Gibco BRL, Life Technologies). Total cellular RNA was isolated by the Trizol protocol, RNA quantitated, and applied to a standard Northern blotting procedure (Burger-Kentischer et al., Kidney Int. 55, 1417-1425, 1999). Probes were generated by the PCR-based DIG-labelling method (Roche Diagnostics). The Kip probe corresponded to bases 8-181 of mouse Kip1 (Genbank accession number U09968) and the human GAPDH probe was as described in Burger-Kentischer et al. For transfections, cells were incubated with the plasmids pjabl and pClneo for 5 h and rested for 1 h before rMIF was added. Kip mRNA levels were found to be at comparable levels independent of whether fibroblasts were incubated with increasing concentrations of rMIF. Overexpression of Jab1 following transient transfection of the pJab1 vector also did not lead to a stimulatory effect of MIF on Kip RNA formation. A similar result was obtained when mRNA levels of Kip were analysed in Jurkat T cells (Kip/GAPDH ratios were: buffer, 1.0; 100 nM rMIF 0.8; 1 μM rMIF, 0.78). b, Effect of rMIF on p27 Kip1 protein synthesis. 4×10⁶ NIH 3T3 fibroblasts were synchronised by an overnight incubation in media containing 1% FCS, rMIF (1 μM) or control buffer were added, and cells incubated for another 36 h. Cells were washed in cysteine/methione-free media, incubated for 15 min in this media. and radioactivity (PROMIX, Amersham-Pharmacia Biotech) added. Cells were incubated for 60 min in the presence of the label, washed and Kip immunoprecipitated with anti-Kip antibody. Samples were electrophoresed in a 13% SDS-PAGE gel and radioactivity detected by a phoshoimager. c, Effect of recombinant MIF on formation of Kip/Jab1 complexes in fibroblasts. NIH 3T3 fibroblasts were either incubated with LLnL for 4 h. Biotin-Jab1 was overexpressed in TNT reticulocyte lysates. For control, TNT lysates were programmed with the pClneo control vector and aliquots from both lysates added to fibroblast lysates as indicated. Recombinant MIF (1 μM) or control buffer was added and the mixtures incubated for 2 h at coimmunoprecipitation conditions. Complexes were precipitated with anti-Kip antibody (Santa Cruz). Immunoblots stained for either biotin-Jab1 (upper panel) or Kip1 (control, lower panel) were then performed to evaluate the potential effect of MIF on Kip/Jab complex formation.

EXAMPLE 7

Mutant Analysis and Competition Experiments.

To investigate more closely those structural parts of MIF which are essential for Jab1 binding and modulation, mutants were created. One mutant is mutant C60SMIF wherein in the wildtype MIF sequence the cysteine at position 60 has been replaced by a serine residue. Other mutants are MIF (50-65) which is a peptide fragment of wildtype MIF consisting of the 16 amino acid residues being present at position 50-65 of wildtype MIF.

Mutant Ser⁵⁷ Ser⁶⁰ MIF(50-65) represents a mutant consisting of the wildtype amino acid residues at positions 50-65 of wildtype MIF wherein at position 57 a serine and in position 60 another serine has been inserted instead of the wildtype amino acids Cys at that position. The above given positions of amino acids are given in correlation with the published human MIF sequence described in Kleeman et al., FIG. 2 (1998 b) whose disclosure content is with respect to the amino acid sequence of MIF and its preparation wholly included in the disclosure of the present teaching.

Mutant C60SMIF exhibits a clear structure activity profile. This mutant can be readily folded for use in activity assays but is devoid of the enzymatic oxidoreductase and immunological activity of MIF. It was found that C60SMIF, while capable of binding to biotin-Jab1, did not inhibit TNF-induced AP-1 activity, did not reduce Jab1-induced enhancement of c-Jun DNA binding, and exhibited reduced p27^Kip1- inducing properties. Cys60-spanning 16 residue MIF peptide, MIF (50-65) and Ser⁵⁷ Ser⁶⁰ MIF (50-65) strongly competed with wildtype MIF for Jab1 binding, indicating together that this region is involved in the binding and modulation of Jab1.

FIG. 9 shows the characterisation of the binding site between MIF and Jab1. Sequence (50-65) of MIF but not the Cys60 residue alone is critical for interaction between MIF and Jab1. a, To investigate whether mutant C60SMIF also bound to Jab1, binding of rC60SMIF (r: recombinant) to biotin-labelled Jab1 was compared with that of rwtMIF in vitro in the TNT reticulocyte lysate. An anti-biotin S—HRP Western blot is shown (left panel). Direct Western blotting analysis of rwtMIF and rC60SMIF verified that the anti-MIF antibody used for coimmunoprecipitation exhibited comparable binding properties to both proteins (right panel). b, MIF peptide (50-65), in a concentration-dependent manner, competes with wildtype MIF for binding of biotin-Jab1 in vitro in the TNT reticulocyte lysate system. An anti-biotin S—HRP Western blot is shown. An analogue of the MIF (50-65) peptide with the two Cys residues substituted for Ser was also tested and also competed for Jab1 binding, confirming that the CALC Cys residues themselves (see a) are not critical for binding of Jab1. c, Mutant C60SMIF only showed reduced p27 Kip1-inducing properties as compared to wildtype MIF. Anti-Kip1 Western blots from lysates of NIH 3T3 fibroblasts and Jurkat T cells incubated with rMIF or mutant C60SMIF (20 nM) are shown.

FIG. 10 shows the effect of MIF on growth arrest of fibroblasts. a, Effect of exogenously added rMIF on Jab1-mediated rescue of fibroblasts from starvation. NIH 3T3 fibroblasts were transfected with Jab1 or control vector and cotransfected with EGFP. Cells were serum-starved for 2 h, rMIF added at the indicated concentrations, cells starved for another 36 h, incubated with BrdU for 18 h, and stained with anti-BrdU antibody. Only GFP-expressing cells were analysed. Data represent the mean±SEM of 4 determinations (>150 cells each) and are representative of 3 independent experiments. Exogenously added rMIF inhibited the Jab1 effect in a dose-dependent manner. With 1.8 μM rMIF added a similar degree of inhibition was obtained as seen when EGFP-MIF was transfected (inhibition from −80% to 40%; see manuscript). b, Effect of rMIF on starvation-induced growth arrest of fibroblasts. Semi-confluent NIH 3T3 fibroblasts were synchronized by incubation in serum-free medium for 24 h. Culture medium was changed to medium with 0.5% FCS, containing either no further addition or rMIF at the indicated concentration. Incubation was continued in the presence of labelled thymidine for 16 h. FIG. 10 is representative of two independent experiments and the data represent the mean±SD of 6 measurements. c, Effect of anti-MIF antibody on cell growth of proliferating or serum-starved fibroblasts. Semi-confluent 3T3 cells were synchronized by incubation in serum-free medium for 24 h. Medium was changed to either serum-free (latter not shown) or 10% FCS-containing medium, containing either no further addition or 200 μg/ml of polyclonal anti-MIF antibody or 200 μg/ml unrelated rabbit IgG as control. Incubation was continued in the presence of labelled thymidine for 16 h. FIG. 10 shows one of four experiments, using anti-MIF IgG in combination with 10% FCS. Similar results were obtained when following synchronization, incubations were continued in serum-free media. d, Effect of endogenously overexpressed MIF on the growth of proliferating fibroblasts. NIH 3T3 fibroblasts were synchronized by incubation in media containing 0.5% serum for 18 h. Cells were switched to full-serum media conditions, transfected with Jab1 or control vector and EGFP-MIF or EGFP, and incubated for another 40 h. The percentage of BrdU-positive cells was assessed as before. Data represent the mean±SD of 4 determinations and are representative of 3 independent experiments.

REFERENCES

• 1. Bernhagen, J. Calandra, T., & Bucala, R. Regulation of the immune response by macrophage migration inhibitory factor: biological and structural features. J. Mol. Med. 76, 151-161 (1998).
• 2. Bernhagen, J. et al. MIF is a pituitary-derived cytokine that potentiates lethal endotoxaemia. Nature 365, 756-759 (1193 1993).
• 3. Calandra, T. et al. MIF as a glucocorticoid-induced modulator of cytokine production. Nature 377, 68-71 (1995).
• 4. Donnelly, S. C. et al. Regulatory role for macrophage migration inhibitory factor in acute respiratory distress syndrome. Nature Med. 3, 320-323 (1997).
• 5. Kleeman, R. et al. Disulfide analysis reveals a role for macrophage migration inhibitory factor (MIF) as a thiol-protein oxidoreductase. J. Mol. Biol. 280, 85-102 (1998 a).
• 6. Kleeman, R., Kapurniotu, A, Mischke, R. Held, J. & Bernhagen, J. Characterization of catalytic center mutants of macrophage migration inhibitory factor (MIF) and comparison with C81S MIF. Eur. J. Biochem. 261, 753-766 (1999).
• 7. Rosengren, E. et al. The immunoregulatory mediator migration inhibitory factor (MIF) catalyzes a tautomerization reaction. Mol. Med. 2, 143-149 (1996).
• 8. Claret, F.-X Hibi, M. Dhut, S. Toda, T. & Karin, M. A new group of conserved coactivators the increase the specificity of AP-1 transcription factors. Nature 383, 453-457 (1996).
• 9. Tomoda, K. Kubota, Y & Kato, J-Y. Degradation of the cyclin-dependent-kinase inhibitor p27^Kpi1 is instigated by Jab1. Nature 398, 160-164 (1999).
• 10. Bacher, M. et al. Migration inhibitory factor expression in experimentally induced endotoxemia. Am. J. Pathol. 15, 235-246 (1997).
• 11. Fields, S. & Song, O. A novel genetic system to detect protein-protein interactions. Nature 340, 245-246 (1989).
• 12. Bacher, M. et al. MIF expression in the rat brain—implications for neuronal function. Mol. Med. 4, 217-230 (1998).
• 13. Ogata, A. Nishihira, J. Suzuki, T. Nagashima, K. & Tashiro, K. Identification of macrophage migration inhibitory factor mRNA expression in neural cells of the rat brain by in situ hybridization. Neurosci. Lett. 246, 173-177 (1998).
• 14. Bernhagen, J. et al. Purification, bioactivity, and secondary structure analysis of mouse and human macrophage migration inhibitory factor (MIF). Biochemistry 33, 14144-14155 (1994).
• 15. Sun, H., Bernhagen, J., Bucala, R. & Lolis, E. Crystal structure at 2.6 Å resolution of human macrophage migration inhibitory factor. Proc. Natl. Acad. Sci. USA 93, 5191-5196 (1996).
• 16. Seeger, M. et al. A novel protein complex involved in signal transduction possessing similarities to 26S proteasome subunits. FASEB J. 12, 469-478 (1998).
• 17. Hofmann, K. & Bucher P. The PCl domain: a common theme in three multiprotein complexes. Trends Biochem. Sci. 23, 204-205 (1998).
• 18. Asano, K. et al. Structure of cDNAs encoding human eukaryotic initiation factor 3 subunits. J. Biol. Chem. 272, 27042-27052 (1997).
• 19. Karin, M. The regulation of AP-1 activity by mitogen-activated protein kinases. J. Biol. Chem. 270, 16483-16486 (1995).
• 20. Hirota, K. et al. AP-1 transcriptional activity is regulated by a direct association between thioredoxin and Ref-1. Proc. Natl. Acad. Sci. USA 94, 3633-3638 (1997).
• 21. Wilhelm, D., Bender, K., Knebel, A. & Angel, P. The level of intracellular glutathione is a kex factor for the induction of stress-activated signal transduction pathways including Jun N-terminal protein kinases and p38 kinase by alkylating agents. Mol. Cell. Biol. 17, 4792-4800 (1997).
• 22. Lanathan, A., Williams, J. B., Sanders, L. K. & Nathans, D. Growth factor-induced delayed early response genes. Mol. Cell. Biol. 12, 3919-3929 (1992).
• 23. Wistow, G. J., Shaughnessy, M. P., Lee, D. C., Hodin, J. & Zelenka, P. S. A macrophage migration inhibitory factor is expressed in the differentiating cells of the eye lens. Proc. Natl. Acad. Sci. USA 90, 1271-1280 (1993).
• 24. Scheinman, R. I., Cogswell, P. C., Lofquist, A. K. & Baldwin, A. S.;. Role of transcriptional activation of IκBα in mediation of immunosuppression by glucocorticoids. Science 270, 283-289 (1995).
• 25. Mitchell, R. A. Metz, C. N., Peng, T. & Bucala, R. Sustained mitogen-activated protein kinase (MAPK) and cytoplasmic phospholipase A2 activaion by macrophage migration inhibitory factor (MIF). Regulatory role in cell proliferation and glucocorticoid action. J. Biol. Chem. 274, 18100-18106 (1999).
• 26. Dedio, J., Jahnen-Dechent, W., Bachmann, M. & Muller-Esterl, W. The multiligand-binding protein gC1qR, putative C1q receptor, is a mitochondrial protein. J. Immunol. 160, 3534-3542 (1998).
• 27. Johannes, F. J. et al. Protein kinase Cu down-regulation of tumor-necrosis-factor-induced apoptosis correlates with enhanced expression of nuclear-factor-kappaB-dependent protective genes. Eur.J. Biochem. 257, 47-54 (1998).
• 28. Angel, P. et al. Phorbol ester-inducible genes contain a common cis element recognized by a TPA-modulated transactivating factor. Cell 49, 729-736 (1987).
• 29. Berberich, I. et al. Cross-linking CD40 on B cells preferentially induces stress-activated protein kinases rather than mitogen-activated protein kinases. EMBO J. 15, 92-101 (1996).
• 30. Roger et al. Increased AP-1 and NFκB activity and increased stability of IL-8 mRNA are implicated in super-induction of IL-8 mRNA by H292 cells. Biochem. J. 330, 429-435 (1998)
• 31. Waeber et al. Insulin secretion is regulated by the glucose dependent production of islet beta cell MIF. Proc. Natl. Acad. Sci. USA 94, 4782-4-787 (1997)
• 32. Kleemann et al. Specific reduction of insulin disulfides by macrophage migration inhibitory factor (MIF) with glutathione and dihydrolipoamide: potential role in cellular redox processes. FEBS Lett. 430, 191-196 (1998 b).

Claims

1. A method for detecting an interaction between MIF and Jab1, the method comprising

a) providing a cell containing a reporter gene comprising a DNA-binding protein recognition site, wherein the reporter gene expresses a reporter protein when the reporter gene is activated by a transcriptional activation domain when the transcriptional activation domain is in sufficient proximity to the reporter gene;

b) providing a first chimeric gene that is operably linked to expression control sequences in host cells, the first chimeric gene comprising a DNA sequence that encodes a first hybrid protein, the first hybrid protein comprising:

i) a DNA-binding domain that recognizes the DNA-binding protein recognition site of the reporter gene in the cell; and

ii) the MIF protein or a part thereof;

c) providing a second chimeric gene that is operably linked to expression control sequence in the host cell, the second chimeric gene comprising a DNA sequence that encodes a second hybrid protein, the second hybrid protein comprising

i) the transcriptional activation domain, and

ii) the Jab1 protein or a part thereof;

iii) wherein interaction between the MIF protein and the Jab1 in the cell causes the transcriptional activation domain to activate transcription of the reporter gene;

d) introducing the first chimeric gene and the second chimeric gene into the cell;

e) expressing the first hybrid protein and the second hybrid protein in sufficient quantity for the reporter gene to be activated; and

f) determining expression of the reporter gene wherein the expression measures MIF-Jab1 binding.

2. The method of claim 1, wherein the expression of the reporter gene is determined in cells subjected to a candidate drug to be screened and compared to the expression of the reporter gene in cells not subjected to the drug.

3. The method of claim 1, wherein the DNA-binding domain and transcriptional activation domain are derived from transcriptional activators having separately DNA-binding and transcriptional activations domain.

4. The method of claim 1, wherein the DNA-binding domain is selected from the group consisting of transcriptional activators GAL4, GCN4 and ADR1.

5. The method of claim 1, wherein the first hybrid protein or the second hybrid protein, or the first and the second hybrid proteins, is encoded on a library of plasmids containing DNA inserts, wherein the DNA inserts are obtained from the group consisting of genomic DNA, cDNA and synthetically generated DNA.

6. The method of claim 1, wherein the chimeric genes are introduced into the cell in the form of plasmids.

7. The method of claim 1, wherein the first chimeric gene is integrated into a chromosome of the cell.

8. The method of claim 1, wherein the first chimeric gene is integrated into a chromosome of the cell and the second chimeric gene is introduced into the host cell as a part of a plasmid.

9. The method of claim 1, wherein the DNA-binding domain and the transcriptional activation domain are from different transcriptional activators.

10. The method according to claim 1, wherein the interaction between MIF and Jab1 is tested in the presence of a MIF competitive peptide.

11. The method according to claim 10, wherein the MIF competitive peptide is

MIF (50-65) (SEQ ID NO: 1) or

Ser⁵⁷ Ser⁶⁰ MIF (50-65) (SEQ ID NO: 2).

12. A method of preparing MIF comprising

a) providing a first source containing MIF;

b) contacting the first source containing MIF with a second source containing isolated Jab1 under conditions allowing for the binding of MIF and Jab1; and

c) separating MIF from Jab1.

13. The method of claim 12, wherein the first source is a cell, tissue, cell culture, cell culture supernatant, cell extract, protein preparation.

14. The method of claim 12, wherein the sources are disrupted by a method selected from the group consisting of sonication, chemical lysis, enzymatic lysis, and pulse or constant electrical field exposure prior to contacting the sources.

15. A method of preparing Jab1 comprising

a) providing a first source containing Jab1;

b) contacting the first source containing Jab1 protein with a second source containing isolated MIF under conditions allowing for the binding of MIF and Jab1; and

c) separating Jab1 from MIF.

16. The method of claim 15, wherein the first source is a cell, tissue, cell culture, cell culture supernatant, cell extract, or protein preparation.

17. The method according to claim 15, wherein the sources are disrupted by a method selected from the group consisting of sonication, chemical lysis, enzymatic lysis, and pulse or constant electrical field exposure prior to contacting the sources.

18. An isolated protein complex comprising MIF or a fragment thereof in natural association with Jab1, or a fragment thereof, optionally in natural association with a protein selected from the group consisting of, p27^Kip1, c-Jun, c-Jun-amino-terminal kinase, steroid receptor coactivator 1, integrin LFA-1, progesterone receptor and a glucocorticoid receptor.

19. A kit for screening candidate drugs, which kit comprises

an isolated protein complex according to claim 18.

20. A composition comprising an isolated protein complex according to claim 18 in a pharmaceutically acceptable carrier.

21. An antibody that is (a) specifically reactive with MIF50-65 (SEQ ID NO: 1) and (b) is not specifically reactive to the MIF/Jab1 complex.

22. The antibody of claim 21, wherein the antibody is a humanized antibody.

23. The antibody of claim 21, wherein the antibody is a monoclonal antibody.

24. The antibody of claim 21, wherein the antibody is a polyclonal antibody.

25. The antibody of claim 21, wherein the antibody consists essentially of an antibody fragment.

26. The antibody of claim 21, wherein the antibody consists essentially of a Fab fragment.

27. The antibody of claim 21, wherein the antibody consists essentially of an Fv fragment.

28. The antibody of claim 21, wherein the antibody is a single chain antibody.