CA2190275A1 - Screening for cytokine modulators - Google Patents

Abstract

This invention provides a method for screening for agents useful for treatment of diseases and pathological conditions affected by cytokines. These agents interact directly or indirectly with an intracellular receptor, which in turn modulates the binding of a rel-like protein, a rel-like protein complex, or other transcriptional proteins to a rel site on the promoter of a cytokine gene. The intracellular receptor can be the estrogen receptor, retinoid acid receptors, retinoid X receptors, glucocorticoid receptor, progesterone receptors, androgen receptor, thyroid hormone receptors, or vitamin D receptor. The select agents can be used to treat osteoporosis, rheumatoid arthritis, inflammation, psoriasis, Kaposi's sarcoma, septic shock and multiple myeloma.

Description

2 i 9027~
Wl> 9~/31722 r~ 4 S~ ;N1N~; FOR ~:Y~L~KLN~: MODULATORS
FIELD OF THE INVF~TION
This invention relates to a method for screening for agents useful for treatment o diseases and pathological condltions affected by cytokines and novel agents identif ied using such screening method .
BACKGROUND OF THE I NVEN- ~m I ON
Cytokines are a group of molecules capable of signalling cellular development. Aberrant expression of cytokines is known to be associated with pathological conditions in~ ;n~ ;n~ i ^ diseases, septic shock, rheumatoid arthritis, psoriasis, inflammation, postmenopau6al osteoporosis, and some cancers. Common treatment for these pathological conditions are ret i n~ lr, immunosuppressants, glucocorticoids and other steroid drugs. Estrogens are specifically employed in the preYention of postmenopausal osteoporosis.
Steroid3 ard related hormone drugs exert their therapeutic ef~ects by binding to a superfamily of intracellular receptors (IRs), which are regulators of gene tran~cription. IRs can function as activators as well as repressors of specif ic cytokine genes . The activity of IRs is controlled by hormones or other ligands that bind to the IRs.
The classical m^-h~ni rm Of transcriptional regulation by IRs involves binding of the IRs to specific response elements in the promoters of the regulated genes, for example, the binding of the estrogen receptor to its response site in the vitel l ogenin gene (Klein-Hitpass et al ., Cell 46 :1053 -1061, 1986) . More recently a different mechanism of IRs function has been described in glucocorticoid receptor mediated AP-l transcription regulation that does not Wo95131722 .~I/~J,, cC~

re~uire direct DNA-bil~ding of the IRs (Yang-Yen et al., Cell 62:1205-1215, l99D) .
Although steroid drugs have been shown to repress the level of cer~ain cytokines, a lack of tissue i3pecificity and side effects 0f the steroids may limit their use as therapeutic agents. These side effects may be reduced or completely avoided with more specifically acting compounds.
Pfahl and Karin (PCT publicatior, WO 92/07072, 1992) describes a method of screening a sample for ligands ~
which bind to a nuclear receptor to form a complex which binds or interferes with a non= rel-like protein AP-1 or an AP-1 component.
i~RY OF THE INVENTION _ __ The present invention relates to a method for identifying new therapeutic agents and for~ uslng these agents to treat diseases and conditions affected by cytokines, such as, but not limited to, osteoporosis, rheumatoid arthritis, ;nfli t;nn, psoriasis, septic shock, Kaposi's sarcoma and multiple myeloma. This method makes it possible to screen large collections of natural, semisynthetic, or synthetic compounds for -~
therapeutic agents that affect the transcription of a cytokine through an intrArpl 1111 Ar receptor mediated pathway .
By ~'cytokine" is generally meant a secreted protein which acts as a chemical mediator of cellular regulation. More ~pec; ~; r~l l y, it is meant a diverse groups of soluble polypeptides~ such as grow~h factors~
and hormones that control the growth, differentiation and function of cells, ;nrlll~;nr, but not limited to, GM-CSF, G-CSF, IL-2, I~-6, IL-B, and IL-11.
The present inventio~ relates to the determination that inhibition of interleukin 6 (IL-~ ~expresslon by estrogen-estrogen receptor complex is mediated through wo ~snl722 r~ 4 21 9~275 the control of the transcriptional activity of NFKB or closely related proteins on the IL- 6 promoter . This rnf~rhAn; ~m does not involve direct binding of ER to Il~-6 promoter but controls the DNA-binding properties of the activated NFKB and possible other members of the rel-family of proteins to their specific response elements (i.e., rel site) on the Il.-6 promoter.
Because NFKB is involved in the regulation of genes encoding various cytokines and their receptors, viral proteins, and proteins involved in the acute-phase response, the regulation of NFKB activity by estrogen and possible other hormones is of general importance (see generally saeuerle, siochemica et Biophysica Acta, 1072:63-8~, 1993, incorporated by reference herein).
For example, retinoic acid treatment, which strongly inhibits Il.-6 expression in +/+LDAll cells and other tissues (Gross, V., P. M. Villiger, B. Zhang, and M.
Lotz, 1993, nRetinoic acid inhibits interleukin-l-induced cytokine synthesis in human monocytes, " J.
I,eukoc. Biol. 54 :125-132), has the same effect as estrogen on the NFKB related complexes with the IL-6 promoter. This suggests a general pathway of transcriptional regulation involving cross-talk between members of the intrArrll-1lAr receptor family and the NFKB transcription factors.
The above det~rm;nAt;nr allows for the screening of drugs that specifically influence genes controlled by the rel-transcription factors, i.e. genes involved in inflammation, sepsis, skin and kidney disorders, osteoporosis, certain cancers, and hematopoietic dysfunctions without the side effects of known steroid drugs. The diseases listed are usually correlated with aberrant expression of cytokines such as I1J- 1, TNFo~
6, IL- 8 that are under the control of NFKB or other rel proteins WO ~)S/31722 i 2 ~ 9 0 2 7 5 F~ 4 .

Thus, the present invention features a method for identifying agents which, by activating an intr_cellular receptor, cause a significant reduction in the binding of a rel-like protein or other transcriptional protein to the rel 6ite o~ the promoter of a cytokine gene or a portion of the promoter, thereby reducing the transcription of the cytokine. : :
By ~intracellular receptor" is meant an intracellular transcription actor whose activity is regulated by binding of small molecules, including, but not limited to, estrogen receptor,~ retinoid acid receptors, retinoid X receptors, glucocorticoid~
receptor, progesterone receptors, dl~d~ (Jyt:ll receptor, thyroid hormone receptors, and vitamin D receptor.
By ~rel-like protein" is meant a protein or a protein complex of the rel family that share a homology in the rel domain and is involved in gene regulation (see Li4u and Baltimore, Current Opinion in Cell Biology, 5:477-487, 1993, incorporated by reference herein), including, but not limited to, NFI~B, Lyt-10, c-rel, and relB.
By ~transcriptional protein" is meant a cytoplasmic or nuclear protein that, when activated, bind a promoter either directly, or :indirectly through a complex of =~
proteins to modulate the transcription activity of the promoter .
By "rel site" is meant a D~A sequence that serves as a binding site for rel-like proteins or co~plexes comprising one or more rel-like proteins, including, but not limited to, I~B motifs identified in saeuerle, Biochemica et Biophysica Acta, 1072:63-80, 1993r~
incorporated by reference herein, such as the NFKs binding site on IL- 6 promoter . ~ :
By "promoter" is meant a DNA regulatory region capable of binding directly or indirectly to RNA
polymerase in a cell and initiating transcription of a Wo95/31722 21 9 0275 r~ 4 .

down6tream (3 ~ direction) coding se~uence A promoter of a DNA construct, including an oligonucleotide sequence according to the present in~ention may be - linked to a heterologous gene when the presence of the promoter influences transcription from the heterologous gene, including genes for reporter~ se~uences such as luciferase, chl~ h~n; col acetyl transferase, ,B-galactosidase =and secreted placental ~:lkAl ;n~
phosphatase In a pref erred embodiment, the assay is conducted in a whole cell ~ystem that has an intracellular receptor which is the target of the screened agent, a promoter or a portion of a promoter with a rel site and a rel-like protein or other transcription~ protein that binds to the rel site; wherein the intracellular receptor modulates the binding of the rel-like protein or the transcription protein to the rel site. The intrAr~ ll Ar receptor, the promoter or a portion of the promoter, or the protein that binds to the rel site may either be endogenous to the cell or transfected into the cell.
In another pxeferred embodiment, the assay is conducted in an extract of cell having an intrAr~ ll Ar receptor, a promoter or a portion of a promotex, with a rel site and a rel-like protein or other protein that binds to the rel site; wherein the intracellular receptor modulates the binding of the rel- like protein or the transcription protein to the rel site.
The binding of the rel-like protein or other transcription protein to the rel site may be measured by techni~[ues known to those skilled in the art, including, but not limited to, mobility shift assay, co-transfection assay, and expression ~f a reporter gene linked to the promoter.:
In a further preferred embodiment, the promoter is acti~ated by an effector, including, but not limited to, tumor necrosis factor, interleukin-1, ~iruses, Wo95/31722 21 90275 r~l~u~ c-~4 endotoxin6, phorbol esters, epidermal growth factor, leukemia inhibitor factor and cAMP agonists.
By "effector". is l~eant an agent that stimulates the expression of a cytokine to a measurable level. An effector may be endogenously produced~in~a cell= or exogenously added to a cell In another further preferred embodiment, the claimed assay iB conducted in a system including an estrogen receptor, an intèrleukin 6_promoter or a portion of an II,-6 promoter and NFI~B; wherein ER ~Lodulates the binding of NFKB or related proteins to the NFKB site on~ the I~-6 promoter. ~
The agents discovered by the above assay may either interact directly with an intrAr~ -l Ar receptox, or modulate the interaction of a ligand with the intracellular receptor. Thus, in an even further preferred ~ a ligand for the intrAr~ ,r receptor is ; nrl ll~lef~ in the assay.
While steroids and steroid analogs may exemplify agents identified by the present invention, Applicant is particularly interested in the identlfication of agents of low molecular weight (less than 10, 000 daltons, preferably less than 5, 000, and most preerably less than l, 000) which can be readily formulated as useful 2~ therapeutic agents.
Such agents can then be screened to ensure that they are specific to tissues with cytokine inflicted pathological conditions with little or no effect on healthy tissues such that the agents can be used in a therapeutic or prophylactic manner. If ~such agents have some effect on healthy tissues they may still be useful in therapeutic treatment, particularly in those diseases which are life threateningl such as Kaposi' 8 sarcoma or multiple myeloma.
Once isolated, a candidate agent can be put in pharmaceutically acceptable formulations, and used for W095/31722 2 1 9 0 2 7 ~ r~ . c e7~

specific treatment of disease~ and pathological condition6 with little or no effect on healthy tissues.
Other features and advantages of the invention will be apparent from the following detailed description of the invention, and from the claims.
BRIEF DRqCRIPTION OF TT~ RAWINGS
Figure l shows IL- 1 and TNFo! induced complex formation on the proximal IL-6 promoter.
Figure 2 skows that several distinct NF~B-related complexes induced by IL-1 and TNF~ are modulated by estrogen.
Figure 3 shows the effects of estrogen agonist and antagonist, and inhibitors of protein synthesis and protein kinase C on the formation of NFKB-related complexes .
Figure 4 shows the binding characteristics of proteins in NFI~B-related complexes with NFKB
oligonucleotides .
Figure ~i shows NFeB related proteins in complexes A, B, and C.
DETAILEI) DESt~TPTION OF 'r~R IIlVR~TION
A number of cytokines, in~ in~ IL-6, Il-8 and IL-11, have related biological effects, i.e., effects on cellular defense in response to infection by stimulating the immune and the acute-phase response and on bone metabolism by increasing bone re60rption. Aberrant expression of any of the~e cytokines results in similar pathological conditions, e.g., all cytokines listed are involved in septic shock. In another example, excessive production IL-8, like IL-6, may be involved in the pathogenesis of several types of inflammatory reactions, particularly neutrophil-~l~p~n-lont tissue damages. These cytokines have similar promoter structures, e.g., their promoters contain binding sites for ~FI~B or other rel _ _ _ _ _ . _ _ .

Wo 95131722 2 ~ 9 0 7 5 ~ c- ~
proteins. It i5 therefore likely that not only IL-6 ~
also the other cytokines mentioned aboYe can be targeted by drugs that modulate the binding of ~F/~s or other rel proteins to their promoter sites through the intracellular receptors. :
Interleukin 6 ana Di~eases Interleukin-6 (IL-6) ls a pleiotroplc cytokine that i8 secreted by many different cells, including monocytes, macrophages, certain B-lymphocytes and T-lymphocytes, glial cells, fibroblasts, osteoblasts, and stromal cells ~reviewed in references Hirano, T., (1992) "The biology of interleukin-6,~" Chem. T Int l. 51:153-lao.; Kishimoto, T. (1989~ ~The biology of interleukiIl-6," ~Q_ 74:1-10.; Kishimoto, T., M. Hibi, M. Murakami, M. Narazaki, M. Saito, and T. Taga (1992) IlThe molecular biology of ;ntf~rlF~llk;n 6 and its receptor, " Ciba Found.
Symp. 167:5-16; discussion 16-23; a~d Wolvekamp, M. C., and R. 1.. Marquet (1990) ~Interleukin-6: historical background, gene~ics and biological significance, Tmmllnr~l, Lett. 24 1-9) Due to its 1n~ ti~ in . . . _ . _ ~ . _ . _ . _ . _ .
response to tissue injury, ;nfl: tion and infection IL-6 ~unction is mainly associated with the host's immune and acute phase responses.
IL-6 is an important mediator of ir,tercellular communication not only under pathological conditions but also under normal physiological conditions. It is involved in neural differentiation (Satoh, T., S.
Nakamura, T. Taga, T. Matsuda, T. Hirano, T. Kishimoto, and Y. Kaziro (1988) ~Induction of neuronal _ differentiation in PC12 cells:by B-cell gt;rmll~t~ry factor 2/interleukin 6, " Mol . Cell Biol. 8 :3546-3549) and proliferation and differentiation during hematopoiesis (Ikebuchi, K., G.G. ~ong, S.C. Clark, J.N.
Ihle, Y. Hirai, and M. Ogawa (1987) "Interleukin 6 =
~nll:ln~ t of ;rtf-rleukin 3-aependent proliferation~of WO 95/31722 2 1 9 0 2 7 5 F~ s c ~4 .

multipotential hemopoietic progenitors, " Proc. Natl.
Acad. Sci. U.S.A. 84:gO35-9P39). However, elevated IL-6 expression is usually associated with disease (Yu, X . P ., - T. Bellido, N. Rice, and S.C. Manolagas (1993) .
IL-6 expression is tightly controlled by other factors. Depending on the particular cell type, it can be activated by various stimuli, including tumor necrosis factor (TNF~) and interleu~ln-1 (IL-1), viruses, endotoxin (lipopolyFArrhAr;des), phorbol esters, epidermal growth factor (EGF), 1Pl~kPm;A
inhibitor factor (hIF), and cAMP agonists.
These ef f ectors exhibit their activity through transcriptional effects on the IL-~ promoter as shown by transfection studies (Gruss, H.J., M.A. Brach, and F.
~TPrr~nn (1992) ~Involvement of nuclear factor-kappa B
in induction of the interleukin- 6 gene by leukemia inhibitory factor, " Blood 80:2563-2570; Ray, A., S.B.
Tatter, L.T. May, and P.B. Sehgal (1988) "Activation of the human "beta 2-interferon/hepatocyte-stimulating factor/; ntPrlp, lk; n 6" promoter by cytokines, viruses, and second messenger agonists, " Proc. Natl. Acad. Sci.
U.S.A. 85:6701-6705). By ser~uence comparison several potential transcriptional control elements have been ;~Pnt;f;ed in the IL-6 promoter, including a cAMP
response element, an AP-1 binding site, and binding elements for the transcription factors NF-IL6 (C/EBPB, LAP, AGP/EBP) and NFKB (Isshiki, H, S. Akira, O.
Tanabe, T. Nakajima, T. Shimamoto, T. Hirano, and T.
~r; Ah; ~to (1990) "Constitutive and interleukin-1 (IL-1) -inducible factors interact with the IL-1-responsive element in the IL-6 gene, " Mol . Cell Biol . 10: 2757-2764 ) .
Direct binding of NF-IL6 and NFI~B to the IL-6 promoter has been estAhl i ~hPd (Akira, S., H. Isshiki, T. Sugita, 0. Tanabe, S. Kinoshita, Y. Nishio, T. Nakajima, T. Hirano, and T. Kishimoto (1990) "A

WOssl31722 2 1 9 0 2 7 5 P~ )..,5,~ '7~

nuclear factor for IL-6 expression (NF-IL6) is a member of a C/EBP iamily, " Er~Bo J, 9 :1897-1906; Libermann, T.A.
and D. Baltimore, (199Q) "Activation of interleukin-6 gene expression through the NF-kappa B transcription factor, " Mol. Cell Biol. lQ:2327-2334~ . NF-IL6 belongs to the C/EBP f amily of leucine zipper protein~ It is induced by I~-1, I~-6 and lipopolysaccharide (EPS), and has been shown to interact with its binding site on the I~-6 promoter and to actiYa~:e IL-6 express~:o~ (Akira, S., H. Isshiki, T. Sugita, O. Tanabe, S. Kinoshita, Y.
Nishio, T. Nakajima, T. Hirano, and T. E~ishimoto, (1990) "A nuclear factor for I~-6 expression~ (NF-IL6) is a member of a C/EBP family, " EMB0 J. 9:1897-1906; Chang, C . J ., T . T . Chen, H . Y. Lei, D . S . Chen, and S . C . Lee (199Q), "Molecular cloning of a transcription factor, AGP/EBP, that belongs to members of the C/EBP family, "
Mol. Cell Biol. 106642-6653; Descombes, P., M.
Chojkier, S. ~ichtsteiner, E. Falvey, and U. Schibler (1990) "~AP, a novel member of the C/EBP~ gene family, encodes a liver-enriched transcriptional activator protein," Genes Dev. 4:1541-1551; Descombes, P., M. Chojkier, S. Lichtsteiner, E. Falvey, and U. Schibler (1990) "L~P, a novel member o~ the C/EBP gene family, encodes a liver-~nr; rh~ transcriptional activator protein, " Genes Dev. 4 154l-1551; Poli, V., F.P.
Mancini, and R. Cortese (199Q) "I~-6DBP, a nuclear protein involved in int~rlf~k;n-6 signal trilna~ rt;on, defines a new family of leucine zipper proteins relatea to C/BP IL-6DBP, a nuclear protein involved in 3Q interleukin-6 signal transduction, defines a ~ew family of leuci~le zipper proteins related to~C/EBP, " Cell :
63 :643-6531 . r~Fl~B i8 a transcription factor that was originally identified as a heterodime~Ic complex consisting of a 5Q kD protein ~p5Q) and a 65 kd~protein (p65 ) that binds an element in the immunoglobulin kappa light chain enhancer Both proteins reveal a high W095/3~722 2 l 9 0 2 7 5 r~ 4 -homology to the Drosophila morphogen ~or~al and to the c-rel proto-oncogeny product. The p65 subunit i8 also functionally related to c=rel (reviewed in references Baeuerle, P. A. (1991) "The ;n~ ;hle transcription activator NF-kappa B: regulation by distinct protein subunits" Bio~him~ Biol?hvs. P~ta ?,.972:63-80; Blank, V., P. Kourilsky, and A. Israel (1992) "NF-kappa B and related proteins: Rel/dorsal homologies meet ankyrin-like repeats, " Trends. Biochem. Sci. 17:135-lso; and Liou, H. C. and D. Baltimore (1993) ''Regulation of the NF-kappa B~rel transcription factor and ~-kappa B
inhibitor system, " Curr. O~in. Cell Biol. 5:477-487) .
Recently, additional proteins (p49/p52 and relB/p68) have been ;rl~nt; ~ied that are flln~ t;r~n~1 ly related to p50 and p65 (HenXel, T., T. Machleidt, I. Alkalay, M.
Kronke, Y. Ben-Neriah, and P.A. Baeuerle (1993) "Rapid proteolysis of ~ kappa B-alpha is necessary for activation of transcription factor NF-kappa B, 1I Nature 365:182-185; PerXins, N.D., R.M. Schmid, C.S. Duckett, K. Leung, N.R. Rice, and G.J. Nabel (1992) "Distinct combinations of NF-kappa B subunits determine the specificity of transcriptional activation, " Proc. Nat Acad. Sci. U.S.A, 89:1529-1533; Ryseck, R.P., P. Bull, M. Takamiya, V. Bours, ~. Siebenlist, P. Dobrzanski, and R. Bravo (1992) ~RelB, a new Rel family transcription activator that can interact with p50-NF-kappa B, " Mol.
Cell Biol, 12:674-684; Ryseck, R.P., P. Bull, M.
Takamiya, V. Bours, U. Siebe~list, P. Dobrzanski, and R.
Bravo (1992) "RelB, a new Rel family transcription activator that can interact with p50-NF-kappa B, " Mol.
Cell Biol. 12:674-684; Schmid, R.M., N.D. Perkins, C.S.
Duckett, P.C. Andrews, and G.J. Nabel (1991) "Cloning of an NF-kappa B subunit which stimulates ~rv transcription in synergy with p~5, ~ Nature 352:733-7~361 . NFKs is located in the cytosol complexes with an inhibitory p~otein Pf the IKB family. Upon inductiPn, NFKB

Wo9is/3l722 21 90275 r~ 4 .
12:
dissociates from IKB and translocates into the nucleus where it binds and activates specific promoters ~Baeuerle, P.A. and D. Baltimore (1988) "I kappa B: a speciiic inhibitor of the NF-kappa B transcription factor, " Science 242: 540-546, Ghosh, S . and D. Baltimore (1990) "Activation in vitro of NF-kappa B by phosphorylation of its inhibitor I kappa B, "
344:67~-682) . Binding of ~FKB-like factors to the consensus site of the II.-6 promoter is induced by IL-1, TNFoo LIF, ~PS and phorbol esters, vary-ing with the particular cell type (Gruss, ~ J., M.A. Brach, and F. Herrmann (1992) "I1lvolvement of nuclear factQr-kappa B in induction of the interleukin-b gene by leukemia ;nh;h;~rry factor,~ Blood 80.2563-2570; Libermann, T~.
and D. Baltimore (1990) "Activation of interleukin-6 gene expression through the NF-kappa B transcription factor," Mol. Cell Biol. 19:232? 2334; i~himizu, H., K.
Mitomo, T. Watanabe, S. Okamoto, and K. Yamamoto (1990) "Involvement of a NF-kappa B-like transc~iption factor in the activation of the interleukin-6 gene by ;nfl tory ly, _hnk;nP~ Mol. Cell Biol. 10:561-568;
Zhang, Y.H., ;r.x. Lin, and J. Vilcek (1990) "Interleukin-6 induction by tumor necrosis factor and interleukin-1 irl human fibrQblasts involves activation of a nuclear factor binding to a kappa B-like seSIuence, "
MQl . Cell Biol . 10 :3818-3823) .
Unregulated expressio~of IL-6 is linked to a number of diseases (Bauer, J. and F. TTorrr~nn (1991) ~Interleukin-6 in rl;n~ri~l monl;r;no,l~ Ann. HematQl.
62:203-210; Hirano, T. (1992) "Inter~eukin-6 and its relation to ;nfli tion and di8ease, ~ Clin. Immunol.
T ~nrni~thol. 62:S60-S65) such as postmenopausal_ osteoporosis after 1088 of ovarian function (Roodrran, G.D. (1992) "Interleukin-6: an osteotrQpic factor?" ~.
Bone Miner. Res. ?-475-478). Ex vi~o culture~.~i bone marrow from ovariectomized mice show an increase of wog5,3l,2~ 2190275 r~l,.) .r- ~4 osteoclastogenesis compared with cultures from sham-operated animals. This increase in osteoclast development can be prevented by i~jection of an anti-IL-- 6 antibody or by administration of ;:estrogen (Jilka, R.L., G. Xangoc, G. Girasole, G. Passeri, D.C. Williams, J.S. Abrams, B. Boyce~, H Broxmeyer, and S.C. Manolagas (1992) "Increased osteoclast development after estrogen loss- I l;Atinn by interleukin-6, " Science 257:88-91) .
In mice that carry a null mutation for IL-6, ovariectomy does not affect bone volume or osteoclast number as seen with normal mice (Balena, R., F. Costantini, M.
Yamamoto, A. Markatos, R. Cortese, G.A. Rodan, and V.
Poli (1993) "Mice with IL-6 gene knock-out do not lose cancellous bone after ovariectomy, " J. Bone M;nlor. Res.
8 :S130 [Abstract] ) .
Requlation of Interleukin 6 bv Estxoqen Estrogen ~as been found to inhibit IL-6 expression in bone-derived stromal cell lines and osteoblastic cells from rats and mice aA well as in nontransformed human bone cells (Girasole, G., R.L. Jilka, G. Passeri, S . Boswell, G . Boder, D . C . Williams, and S . C . Manolagas (1992) "17 beta-estradiol inhibits interleukin-6 production by bone marrow-derived stromal cellæ and osteoblasts in vitro: a potential mechanism for the antiosteoporotic effect of estrogens, " J. Clin. Tnvest.
89:883-891). This effect of estrogen on IL-6 expression is not restricted to bone tissue but has also been shown for uterine cells (Jacobs, A.L., P.B. Sehgal, J. Julian, and D.D. Carson (1992) "Secretion and hl ,l regulation of -interleukin-6 production by mouse uterine stromal and polarized epithelial cells cultured in vitro,~ Endocrinoloqv 131:1037-1046; Tabibzadeh, S.S., U. SAnthAnAm, P.B. Sehgal, and L.T. May (1989) "Cytokine-induced production of IFN-beta 2/IL-6 by freshly explanted human endometrial stromal cells.
Modulation by estradiol-17 beta, " J. Immunol. 142:3134-2 1 9027 ~
Wo 9S/31722 - F~ n. r-~74 .

3139). There are only a few other genes known to be negatively regulated by estrogen agonists ~Adler, S., M.L. Waterman, X. He, and M.G. RosRnfeld ~1988) ~Steroid receptor-l ~; Atf"l inhibition of rat prolactin gene expression~does not require the receptor DNA-binding domain,~ Cell 52:6a5-695; Ree,~A.H., B.F. Landmark, W.
Eskild, FØ Levy, H. Lahooti, T. Jahnsen, A. Aakvaag, and V. Hansson ~1989) ~Autologous down-regu~lation of messenger ribonucleic acid and protein level6 for estrogen receptors in MCF-7 cells: an inverse correlation to progesterone receptor levels, "
~nllncr; nnloqv 124: 2577 -2583 ) . _ _ _ To investigate the mf~rh~n; sm of the e6trogen effect, Applicant perf ormed a series of DNA-hinding experiments using the human IL-6 promoter. Co-transfection studies showed that the ~roximal 225 bps of ~the IL-6 promoter mediate both the induction of the reporter gene by IL-1 and TNFcY as well as the repression by estradio~ The repression by estradiol also required the expression of the estrogen receptor (ER).
Using gel retardation assays, no specific binding of the ER to the proximal 225 bp could be detected.
However, nuclear extracts from +/+~DAll bone marrow stromal cells that revealed II~-6 regulation by IL-1, TNF~, and estradiol showed an induced complex with a -225 to -52 promoter fragment when the cell were treated with IL-1 and TNF~. Induction of the complex was fast (10 minutes) but transient Pretreatment of the cells with estradiol increased the intensity as well as the mobility of the complex.
To iaentify the proteins involved in the formation of the complex, antiboay supershift experiments were carried out using antibodies again~t factors with potential binding sites in this promoter fragment 35 ;n~ l;n~ c-jun, NF-IL6, c-rel, and ~FKB p50 and p65 WO gS~l722 2 1 q 0 2 7 5 r~ c ''74 .

proteins. Only anti-p50 and anti-p65 had an effect and abolished the formation of the induced complex.
An oligonucleotide ccvering the potential NFKB 6ite - of the IL- 6 promoter competed against the induced binding to this fragment, while an oligonucleotide covering the NF-IL6 Eite was ineffe~ctive. When the NPKs oligonucleotide was used as probe, three IL-1/TNF~-induced complexes were observed.
Pretreatment with estradiol decreased the intensity of the slowest complex and strongly increased the intensity of the faste6t migrating complex. The three bands were differentially supershifted (i.e., further decrease in the mobility of the complex due to binding of the antibody) by anti-p50 and anti-p65 antibodies, while none of several other antibodies tested, including anti-ER antibody, had any effect. Methylation interference assays showed identical DNA contact sites for all three complexes.
Ray, et al., J. Biol. Chem., 269(17) :12940-946 (1994), not admitted to be prior art, describe that activation of the IL-6 promoter, elicited by a combination of NF-II.6 and the p65 subunit of NFKB, can be inhibited by the wt E~ but not by an ER nnllti~inirl~ a mutation in its DNA binding domain. Furthermore, the repression of the IL- 6 promoter by a combination ER and l~,B-estradiol did not appear to be mediated via high af f inity binding of the receptor to the promoter .
These data suggest that negative regulation by estrogen is mediated through the IL-6 promoter and is estrogen receptor dependent. Inhibition of IL-6 expression by estrogen is mediated through control of the transcriptional activity of NFKB o~ closely related - proteins on the I~- 6 promoter .
Mukaida, et al., J. Biol. Chem., 269(18) :13289-295 (1994), not admitted to be prior art, describe that a glucocorticoid, dexamethasone, inhibited IL- 8 production Wo95131722 2 ~ 902 75 at the transcriptional level . = Mutation of either the AP-l or rirF-IL6 b~nding site on the IL-a promoter did not abolish I~-8 gene repression by dexamethasone, suggesting that these sites were not targets for ~ h~ n~ Yet dexamethasone ~imini qhPfi the IL-l induced formation of l~FKB complexes. ~= ~
The invention will now be described in greater detail by reference to. t~e following non-limiti~g examples regarding the regulation of: interleukin 6 transcription by estrogen receptor.
ExamPles A candidate agent will be screened by either A) direct evaluation of protein binding to rel-sites, or B) indirect evaluation of binding to rel-sites.
A) Direct evaluatiDrl Qf ~rQtein bindinq to rel-sites Cells selected :Eor expression of the necessary c~ nr)n,-nts will be treated with the agent or vehicle control and an inducer (e.g., phorbol ester, cytokines, lipopolysaccharides). Cellular extracts prepared from those cells (e.g., whole cell, ~cytosolic, or nuclear extracts) will be analyzed for their DlIa-bir,dirg using cytokine promoter fragments~ or various rel-sites as probes. Binding will be analyzed qualitatively (i.e.t comparing pattern~ and ~uantitatively comparing extracts from cells treated with vehicle or the agent.
B) Indirect evaluation of bindinq to rel-sites by l) Measurinq endoqenous cvtokine ex~ression.
Cells selected for expression of the necessary components and their productior. of cytokine will be treated with the agent or vehicle control and an inducer (phorbol ester, cytokines, lipopolysaccharides).
ActiYity of the agent will be qua~titatively assessed by 2l 9~27~
Wo ss/3 1 722 1 ~ . ' 'C ' ''~

measuring of cytokine using standard assays known to those skilled in the art.
2 ) Measurinq the exA,ression of a re~orter - introduced into the cell.
By means of transfection a reporter construct will be introduced into the cells that expresses an eaeily measurable protein under the control of a cytokine promoter or fragments thereof or isolated rel-sites.
The other n~ri~AR;lry components are~either expressed 10 endogenously by the cells or provlded by cotransfection of expression vectors for the particular, ~ ^nt Cells will be treated with the agent or vehicle control and an effector (phorbol ester, cytokines, lipopolysaccharides ) . The activity of the agent will be 15 analyzed ~uantitatively by measuring the expression of the reporter protein.
Agents will also be tested for their binding to IRs by traditional binding assays as well as for their activity to effect the classical mechanism of gene 20 regulation by IRs. An agent that binds to IRs and regulates binding of rel proteins to cytokine promoters but does not activate the classical r~ n; Fm Of IR
action is a potential drug candidate for the specific treatment of diseases associated with aberrant 25 expression of cytokines.
Experimental procedures employed in the examples described herein are set forth below:
Tr;lnAient transfections and ~ n ex~res5ion 3 constructs Construction of the pERE-tk-Luc reporter plasmid and the vector expressing ~Rgly ~pRShER) has been described - (Tzukerman, M., A. Esty, D. Santiso-Mere, P. Danielian, M.G. Parker, R.B. Stein, J.W. Pike, and D.P. McDonnell 35 (1994) ~uman estrogen receptor transactivational capacity is determined by both cellular and promoter W0 95l3~722 2 1 9 0 2 7 5 ~ 'C ''74 conte~t and mediated by two fu~ctionally distinct intramolecular regions, n Mol . Endocrinol . 8 :21-~30, incorporated by re~ereIlce herein) . The pIL6 [-225] Luc ~
reporter construct was derived irom the parental pIL6 [- -1200] Luc by excision of a NheI-BamHI fragment and religation of the vector fragment after blunt ending with ~Clenow DNA-polymerase. The parental pIL6[-1200]Luc was constructed by clQning the 1.2 kb IL-6 promoter insert excised with BamHI and KpnI frDm pCAT-M54-IL6 (-) into the corresponding sites of the luciferase Yector Lucpl.
C3HlQT1/2 cel 1 s were seeded in phenol-red-free DMEM
supplemented with 109~ FBS at 80, 000 cells per well (12-well plates). The c~lls were transfected by calcium phosphate precipitation (Peterson, J.L. and O.W. McBride (1980) ~Cotransfer of linked eukaryotic genes and efficient transfer Qf hypn~ ntll;nf~
phosphoribosyltransferase by DNA-mediated gene transfer," Proc. Natl. Açad. 9ci. IJ.S.A. 77:1583-l587) with 0.5 mg pIL6[-225]Luc alone or together with 0.05 mg pRShER or 0.1 mg HE0 using pGEM as carrier to adjust to 2 mg total DNA in the transfection mix. After 4 h at 37 C the cells were treated with 79~ DMS0 for 30 min followed by a medium change and addition of hormones.
The folIowing aay the cells were induced with T~F~ and II,-lb (1 nM each) for 24 h. After a brief wash with PBS
the cells were lysed in 200 ml lysis bu~fer (25 mM Tris [pH 7.8], 2 mM DTT, 2 mM 1,2-~ m;nncyrlnl~ n~-N,N,N' ,N' -tetraacetic acid, 10~6 glycerol, 1~ Triton X-lO0) . To 20 ml of each ~ctract 100 ml of reagent (20 mM
Tricine [pH 7.8], 1.07 mM (MgC03)~Mg(OH~2, 2-67 mM MgSoi, 0.1 mM EDTA, 33 3 mM DTT, 279 mM coenzyme A, 470 mM
luciferin, 530 mM ATP) was added ard lllr; ~r~re activity war measured immediately with a Dynatech lllminometer in cycle mode wo gsl3l722 2 1 9 0 2 7 5 r~ ,JI~ ~74 .

Antibodies. IL-6 Fr,r~, and ER assaY
Peptides used to raise the f ollowing antibodies in rabbits correspond to amino acld resdues 91-105 of murine c-jlm (Ryder and Nathans (1988) 'iInductlon of protoonco~ene ~- jun by serum growth factors, " Proc.
Natl . Acad. Sci . USA 85 : 8464-8467), 278-296 of murine NF-IL6 (Chang, C.J., T.T. Chen, H.Y. Lei, D.S. Chen, and S. C. Lee (199Q) "Molecular cloning of a transcription factor, AGP/EBP, that belongs to members of the C/EBP
family, " Mol. Cell Biol. 10:6642-6653), 15a-176 of murine c-rel IBUll, P., K.L. Morley, M.F. Hoekstra, T.
Hunter, and I M. Verma (1990) "The mouse c-rel protein has an N-terminal regulatory domain and a C-terminal transcriptional transactivation domain, " Mol. Cell Biol.
10:5473-5485; Inoue, J., L.D. Kerr, L.J. Ransone, E.
Bengal, T. Hunter, and I M. Verma (1991) ~c-rel activates but v-rel suppresses transcription from kappa B sites, " Proc. Natl. Acad Sci. U.S.A. 8:3715-3719), 347-361 of murine p50 (Ghosh, S., A.M. Gifford, L.R.
Riviere, P. Tempst, G.P. Nolan, and D. Baltimore (1990) "Cloning of the p50 DNA binding subunit of NF-kappa B:
homology to rel and dorsal, " Cell 62:1019-1029), and 3-19 of human p65 (88g6 homology with murine p65) (Nolan, G.P., S. Ghosh, H.C. Liou, P. Tempst, and D. Baltimore (1991) "DNA binding and I kappa B inhibition of the cloned p65 subunit of NF-kap~a B, a :re~l-related polypeptide, " Cell 64:961-969; Ruben, S.M., P.J. Dillon, R. Schreck, T. E~enkel, C.H. Chen, M. Maher, P.A.
Baeuerle, and C.A. Rosen (1991) "Isolation of a rel-related human cDNA that potentially encodes the 65-kD
subunit of NF-kappa B [letter], " Science 254 :11) . All the r,o~f~r~n~ ~s m~n~ od above= are incorporated by reference herein~ All antibodies listed above were obtained affinity purified at a concentration o~ 1 mg/ml from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA).
Anti-TBP was a~protein-A purified serum preparation from Wo95131722 21 ~0275 r~ c ?1 a rabbit immunized with t~e iull length human re~ ' ,; n~nt protein and reported to react with TBP from mou~e, rat, and human oriyin (Santa Cruz Biotechnology, Inc. ) . Anti-ER i8 a mouse monocl~al antibody (IgG2a) raised against a peptide corresponding to amino acid residues ~-22 o~ the murine ER. I~-6 concentration ir.
tissue culture supernatants was determined by use of an I1.-6 E~ISA kit (Endogen, Inc., Boston, MA) using murine II.- 6 as standard. ~ =
ER in +/+LDA~l cells was measured ir whole cell extractF:. After washing and counting, cells were homogenized in bu~fer crnt~;nlnr 50 mM Tris [pH 7.5], 30~ glycerol, 500 mM KCl, 1 mM EDTA, l mM PMSF, and 5 mM
DTT. After 30 min on ice the homogenate was centrifuged (lO0, 000 g, 4 C, l h) . The supernatant was taken as whole cell extract, adjusted to 0.5~6 CHAPS, and incubated with 5 nM [3H] estradiol in the absence or presence oi a 200-fold excess of DES oYernight at 4 C.
After incubation with anti-ER antibody, the complexes iormed were preclpitated with protein-A sepharose (Pharmacia), washed three times with lO mM Tri6 [pH
7 . 5] /o . 536 CHAPS, and measured by li~r,uid 8r; nt; 1 1 ;3t; rn count ing .
Electro~horetic mobilitv shift assa~ (EMSA) and methylation interference ~ Y
DNA binding studies were carrie out with nuclear extracts ~rom +/+I~DAll cells, extracts ~rom yeast expressing recombinant human ERg1y, and purified p50 and p49 proteins. +~+LDAll cells were m~;n~lnP-l under conditions as described (Girasole, G., R.~. Jilka, G.
Passeri, S. Boswell, G. Boder, D.C. Williams, and S.C.
Manolagas ( l9 92 ) n 17 beta - estradiol inhibits interleukin-6 pr~duction by bone marro~derived stromal cells and osteoblasts in vitrs: a potential mechanism for the antiosteoporotic eifect Rf ~estrogens, '~

21 9027~
WO 9!j131722 r~ .'C ''~4 J, Cl;n~ Tnvegt. 89:883-891). To prepare nuclear extracts the cells were eeeded in phenol-red-free McCoy' s medium supplemented with 101; FBS and pretreated - with hormone for 24 h if not indicated otherwise. After adjusting the medium to 2~ FBS, the cells were induced with T~F~ and ~-lb (1 nM eachl for varying periods. In cases where cycloheximide (10 mg/ml) or the kinase inhibitor H7 (50 mM) were included, those compounds were added 5 min before induction. Incubation was stopped by two washes with ice cold PBS and cells were lysed in situ in cold buffer A (10 mM HEPES rpH 7.9], 1.5 mM
MgCl~, 10 mM KCl, 0.5 mM DTT, 0.296 Nonidet P-40) .
Lysates were transferred into microfuge tubes, nuclei pelleted (8000 rpm, 1 min) and resuspended in buffer C
(20 mM HEPES [pH 7.9], 1.5 mM MgClj, 420 mM NaCl, 25~
glycerol, 0.2 mM EDTA, 0.5 mM DTT, 0.5 mM PMSF). After 40 min rocking at 4 C, samples were centrifuged (15, 000 rpm, 10 min) and supernatants taken as nuclear extracts.
Bradford protein assays ~Bradford, M.M. (19i6) "A rapid and sensitive method for the quantitation of microgram quantities of protein llt; 1 i 7;ng the principle of protein-dye binding, " Anal. Biochem. 72 248-254) showed only minimal variations in protein concentrations which did not correlate with hormone or cytokine treatment.
Extracts of yeast recombinantly expressing ERgly were prepared from the BJ2168 strain transformed with YEpE10 as described (Tzukerman, M., A. Esty, D. Santiso-Mere, P. Danielian, M.G. Parker, R.B. Stein, J.W. Pike, and D.P. McDonnell (1994) "Humar3 estrogen receptor traneactivational capacity is det~rmin~rl by both cellular and promoter context and mediated by two functionally distinct intr~rnn~ ~ regions, ~ Mol.
Endocrinol. 8:21-30). Purifiedl E6cherichia coli expressed human p50 and p49 proteins were purchased from Promega (Madison, WI).

wo95l3~722 2 1 9 0 2 7 ~ r~l~u~ s c~-74 For EMSA, 2ml of the ~ tr~rt,q uere preincubated with 2 mg poly[dI-dC] in binding buffer ad~ustea to 20 mM _ XEPES [p~ 7.9], 40 mM NaCl, 20 mM KCl, 2.5 mM MgCl2, 109~
glycerol, 0.1 mg/ml BSA, and 1 mM DTT. When the -225 to -52 I~-E promoter 1~r;ls nt was use~L as probe 0.5 mg of Bluescript plasmid (Stratagene, ~a ~oll~, CA) was also included. After 20 min on ice, the probe was added and the incubation rr,nt;nl~f3rl for 20 min at room temperature.
When antibodies were; nrl ~ , 1 mg was added 20 min =
after the probe and tlle ;nrl-hi~t;nn continued for 40 min at 4 C. The complexes ~ormed were analyzed on ~on- ~
denaturing polyacrylamide gels (4gs acrylamide/0 . 0596 BIS;
2x200 mm) at 4 C and 15 V/cm in 0.25xTBE. _ PrDhes were either double stranded oligonucleotides corr~qprnfl1ng to the regions -82 to -47 (ATCAaATGTGG(3ATTTTCCCATGAGTCTt~T~TT~) and -172 to -131 (CTA~AGGACGTCACATTGCACAaTCTTAATA~GGTTTCC~AT) of the human II--6 promoter and to the ERE of the vitellogenin promoter ~Tzukerman, M., X.K. Zhang, T. Hermann, K.N.
Wills, G. Graupner, and M. Pfahl (1990) "The human estrogen receptor has transcriptional activator and repressor functions in the absence of ligand, " New Biol .
2:613-620) or the -225 to =52 NheI-SspI II,-6 promoter fragment. All probes were either labeled with [y32P]ATP
using T4-polynucleotide kinase or with ta~3~P] dATP using Klenow polymerase and subsequently purified by polyacrylamide gel electrophoresis. ~
~or methylation interference assays, the -82 to -47 probe labeled with Ly32P] ATP either on the upper or the lower strand was subjected to limited DMS-methylation (Maxam, A.M. and W. Gilbert ~1980) ~Seguencing end-labeled DNA with base-6pecific rhl~rn;r;:l cleavages,"
Methods Enzvmol . 65: 499-56D) ~ EMSA was performed as described above, scaled up 10--fold. Gels were blotted onto NA45 anion exchange membranes (Schleicher &
Schuell) in 0.5xTBE for 30 min at 30 V ~Singh, H., J.X.

W0 95~31722 2 ~ 9 0 2 7 5 r~ c ~4 LeBowitz, A.S. saldwin, Jr., and P A. Sharp (1988) "Molecular cloning of an enhancer binding protein:
isolation by screening of an~ expression library with a - recognition site DNA, " Cell 52 :415-423) . After autoradiography, the DNA corresponding to the various P~ and the unretarded probe was eluted (10 min at 65 C in 20 mM Tris [pH 8 . 0], 1 M NaCl, 0 . 1 mM EDTA) and purified by phenol/chloroform extraction and ethanol precipitation. After strand cleavage in 1 M piperidine (30 min at 9o C) the fragments were resolved on denaturing polyacrylamide gels (12% acrylamide/0.6%
BIS ) .
r le 1. Screeninq for ER mediated inhibition of IL-6 DromOter activity It has been-shown that I~-6 repression is regulated by estradiol at the mRNA level (Girasole, G., R.L.
Jilka, G. Passeri, S. Boswell, G. Boder, D.C. Williams, and S.C. ~ n~lag~2 (1992) "17 beta-estradiol inhibits interleukin-6 production by bone marrow-derived stromal cells and osteoblasts in vitro: a potential mechanism for the antiosteoporotic effect of egtrogens, ~ J. Cl ;n .
Invest. 89:883-891; Jacob~, A.L., P.B. Sehgal, J.
Julian, and D D. Carson (1992) "Secretion and hormonal regulation oi interleukin-6 production by mouse uterine stromal and polarized epithelial cells cultured in vitro, " EndocrinoloqY 131:1037-1046) . To determine if estrogen or a candidate agent act6 directly on IL-6 transcription, we transfected a reporter construct, expres~ing the firefly luciferase under the control of the human IL- 6 promoter region f rom - 225 to +14, into the murine fibroblast cell line C3~10T1/2. These cells can be considered as pre-osteoblasts since they differentiate into osteogenic cells ~in response to bone morphogenic protein-2 (Katagiri, T., A. Yamaguchi, T.
Ikeda, S. Yoshiki, J.M. Wozney, V. Rosen, E.A. Wang, H.

W0 951317Z2 p~ 4 .

Tanaka, S. Omura, and T. Suda (1990~ "The non-osteogenic mouse pluripotent cell line, C3HlOT1,/2, is induced to differentiate into osteoblastic cells by recombinant human bone morphogenetic proi~ein-2, " Biochem. Bio~hvs .
Res. C~lmml1n, 172:295-299) . = = ~
Therefore, C31~1QT1/2 cells were transfected:with a i f~r~Re expression vector under the control of the proximal human I~-6 promoter (pIL6 [-225] ~uc) alone or :~
together with the expression vector for the wild-ty~e human ERgly ~pRShER). After pretreatment with varying concentrations of estradiol for 24 hoursl the cultures were induced with 1 nM each of T~F(Y and I~-l or ~ef t n;n~ and 24 h later cells were harvested and extracts analyzed f or l - l ~; f ~r~ Ri~ act ivity .
Treatment of transfected cells with I~-l and TNF~
induced a 5-fold increase in luciferase activity over basal levels. Without cotransfection oi a plasmid expressing the estrogen recepto~, treatment with estradiol had no effect. ~owever, with the expression of estrogen receptor by cotransfection, treatment with estradiol resulted in a strong, dose-dependent repression of luciferase activity.
Repression was observed with the wild-type human ER
(ERgly) as well as with an ER variant r~mt:~inin~ a glycine to valine point mutation in the hormone binding domain ~ER~,,l) (Tora, ~., A. Mullick, D. Metzger, M.
Ponglikitmongkol, I. Park, and ~. Chambon (1989l "The cloned human oestrogen receptor contains a mutation which alters its hormone binding properties, n EMBO J. _ 8:1981-198~ While E~,,l required a higher estradiol concentration, it exhibited a stronger repression. This is consistent with the f inding that in induction experiments ERgly responds at lower hormone concentrations but has considerable basal activity (Tzukerman, M., X.K.
Zhang, T. Hermann, K.N. Wills, G. Graupner, and M. Pfahl (1990~ "The human estrogen receptor has transcriptional Wo95J31722 1'~~ 'C '?4 2s activator and repressor functions in the absence of ligand,~ New Biol. 2:6~3-620).
The dependence of the estrogen effect on cotransfected ER suggested that C3HlOTl/2 cells do not express functional endogenous ER. This was confirmed by transfecting the cell with a luciferase reporter under the control of the vitellogenin estrogen response element (ERE) (Klein-~Iitpass, L., M. Schorpp, U. Wagner, and G.U. Ryffel (~986) ~An estro~en-responsive element derived from the 5' flanking region of the Xenopus vitellogenin A2 gene functions in transfected human cells, ~ Cell 46:1053-1061) .
Therefore, C3HlOT1/2 cells were transfected with a lucif erase expression vector under the control of the minimal thymidine kinase promoter and the~:~Fitellogenin estrogen response element (pERE-tk-Luc) alone or together with pRShER. 24 h after treatment with 10 nM
estradiol or vehicle cells were harvested and extracts analyzed for luciferase activity. Induction of luciferase actlvity by estradiol was only observed in the presence of cotransfected ER.
In addition, C3HlOT1/2 cells were incubated with or without lO nM estradiol. After 24 h the cultures were induced with TNFo! and I~-l (l nM each~ or left llninr~ rl for additional 24 h. Il-6 in the supernatants was assayed by an ELISA specific for mur~ine IL-6. C3HlCTl/2 cells responded to IL-1 and TNF~ treatment with strongly increased production of endogenous IL-~, but unlike other osteogenic or stromal cells cnnt~;n;nr~ endogenous ER (Girasole, G., R.L. Jilka, G. Passeri, S. Boswell, G.
Boder, D.C. WiIliams, and S.C. Manolagas (1992) "1~
beta-estradiol inhibits interleukin-6 production by bone marrow-derived stromal cells and osteoblasts in vitro: a potential m~ h;~n; rm for the antiosteoporotic effect of estrogens," J, Clin. Invest. 39:883-891), ~L-6 levels were not decreased by estradiol. These data suggest Wo9~131722 2 1 ~ 0 2 7 ~ P~IIIJ.. ,_.~ ?4 .

that the ;nh;hjt;nn of I~-6 expressio~:is at the transcriptional level and meaiated through the E~.
sy cotransfection studies using tke preoste=oblastic cell line C3EI10T1~2, we showed that IL-~lTNF~-inducPd activation of the proximal II.- 6 promoter region could be inhibited by estrogen. This inhibition was estrogen receptor dependent and was observed with both the wild-type human ER (ERgly) and the ER,"l variant. Similar results have been obt~ined by others in both HeLa cells cotransfected with ER~l/ and in MBA13 cells, a preosteoblastic cell line expressing endogenous ER.
Together with the described effects of ~strogen on I~-6 mRNA (Girasole, G., R.l.. Jilka, G. Paæseri,= S. Boswell, G. Boder, D.C. Williams, and S.C. Manolagas (1992) "17 beta-estradiol inhibits int~rlP~ ;n-6 production by bone marrow-derived stromal cells and osteoblasts in vitro. a potential mechanism for the antiosteoporotic effect of estrogens, ~ J. Clin. Inveæt. 89 .883-891i Jacobs, A.L., P.B. Sehgal, J. Julian, and D.D. Carson (1992) ~Secretion and hormonal requlation of interleukin-6 production by mouse uterine stromal and polarized epithelial cells cultured in vitro, " EndocrinoloqY ::
131 : 1037-1046), these results suggest a transcriE)tiQnal mechanism of estrogen-induced ;nh;h;t;nn, A ~-~n~ te agent can be scre~ned using the above assay, replacing estradiol with said agent.
le 2. Screeninq aqents that modulates bindinq of NFI~B related ~roteins to the ~roximal Il.-6 ~romoter A cell line that ex~resses ER (+/+:LDA11) _ ~n exemplary assay system ~is a cell l~e that expressed all the necessary component3 endogenously, including the ER. The bone marrow derived murine stromal cell line ~/+LDA11 has been shown to respond to I~-1 and TNF~Y treatment with strongly increased Wo 95/31722 , ~ " ~J~,5 o(C ~4 secretion of I~.-6 Treatment with estradiol inhibits this induction of I~-6 as shown for the protein and its mRNA (Girasole, G., R.L. Jilka, G. Passeri, S. Boswe G. Boder, D.C. Williams, and S.C. Manolagas (1992) "17 beta-estradiol inhibits interleukin-6 production by bone marrow-derived stromal cells and osteoblasts in vitro: a potential mechanism for the antiosteoporotic effect of estrogens, " J. Clin. Invest. 89:883-891~
rO verify that ER is actually present in +/+I DA11 cellg, hormone binding studies were carried out.
Initial experiments showed a low number of specif ic estradiol binding sites in high salt extracts from these cells. Using a monoclonal antibody directed against the amino terminus of the ER, specifically bound [3~] estradiol was immunoprecipitated confirming that the binding sites represented ESR. From those studies we calculated that +/+I DA11 cell~ contain approximately 1000 ER molecules per cell.
However, when using electrophoretic mobility shift assays (EMSA~ in com~ination with the vitellogenin ERE
as a probe, ER-specific D~A binding activity could not be detected in nuclear extracts from +/+LDA11 treated with estradiol and/or IL-1 and TNFom ~uclear extracts of +/+LDA11 cells pretreated with estradiol (10 nM~ and TNF~ and III-l (lnM each for 40 min) as i~dicated or yeast extract rnnt;linln3 recombinantly expressed human wild-type ERgly were incubated with the vitellogenin ERE
as probe in the absence or presence of anti-ER antibody.
Complexes formed were analyzea by EMSA.
The complexes detected are unrelated to the ER since they were not significantly affected by anti-ER
antibody. Controls using ER nnntF31n~ng extracts obtained from a yeast expression system gave rise to two slowly migrating complexes that were s~ecifically shifted with the anti-ER antibody. These data suggest _ wo 95131722 2 1 9 0 2 7 5 PCT/US95106524 -that ER is present in +/+LDA11 ~cells but at concentrations tqo low to be detected by EMSA.
DNA-bin~;nq activitv of nuclear extracts to the IL-6 _ ~romoter ~ _ To study the molecular mechanism of IL-6 induction, and its repression by estrogen, nuclear extracts from +/+LDAll cells were analyzed for DNA-binding activity to the IL-6 ~promoter region that mediated cytokine induction and estrogen suppressiorL in the cotransfection experiments. Since this DNA fragment showed a high background binding with nuclear extracts the most proximal region r~n~;n;ng the TATA box was removed leaving a fragment from -225 to -52 upstream of the transcriptional start site. This regiqn of the promoter contain3 consensus binding sites for several transcription factors ;n~ lin~ a core sequence of the cAMP response element tCRE), a binding site for the leucine zipper protein NF-IL6, and a NFI~B site (Isshiki, H., S. Akira, O. Tanabe, T. Nakajima, T. Shimamoto, T
Hirano, and T. Kishimoto (1990~ "Constitutive and interleukin-l (IL-l)-;nr~ ihle factors i~teract with the IL-l-responsive Pl ement in the IL-6 gene, " Mol . Cell BiQl. 10: 2757-2764) .
sinding of NF-IL6 and ~FKB-like pr~teins to these sites has been demonstrated lAkira, S., H. Isshiki, T.
Sugita, 0. Tanabe, S. Kinoshita, Y. Nishio, T. Nakajima, T. ~irano, and T. Kishimoto (1990) "A nuclear iactor for IL-6 expressio~ (~F-IL6) is a member of a C/EBP family, "
EM80 J. 9 :1897-1906; Libermann, T.A. and D. Baltimore (1990) "Activation of interleukin-6 gene expression through the ~F-kappa B transcription_factor, " Mol. Cell ~Q~L 10:2327-23:3a~). This fragment was incubated with nuclear extracts from +/+LDA11 cells that had been treated with IL-1 and TNFIY for various times. +/+LDA11 cells were pretreated with 10 nM estradiol as indicated.

WO gsr3~r22 After 24 h the cells were induced with TNF~ and IL-1 (1 nM each) for variou6 periods of time. Induction was stopped by cell lysis and nuclear ;extracts were analyzed by EMSA using the -225 to -52 IL-6 promo~er fragment as probe (Flgure la).
Complexes formed were analyzed by EMSA. After treatment with the cytokines an inducible complex was observed. ~he intensity of the complex was maximal already after 1o min treatment with IL-1 and TNFo! and decrea6ed gradually over time . Af ter 2 hours of induction the intensity of the complex was signif icantly reduced .
Pretreatment of the cells with estradiol had no effect on the binding capacity of extracts from lln;nrl~ P~l cells. However, estradiol pretreatment resulted in a marked increase of the induced complex with induction intervals from lo min to 40 min but only a slight effect on the complex after 2 h of induction.
In addition to the increased intensity, pretreatment with estradiol also caused a qualitative change, increasing the mobility of the complex~
Detect;nq thP com~osition sf the DNA-bindinq com~lex To i~vestigate the nature of the complex and the proteins potentially involved, we incubated the binding reactions with antibodies directed against several potential binding factors. Nuclear extract from +/+LDAll cells ~reated with estradiol (10 nM) and TNF~
and IL-l (lnM each for lo min) as indicated were incubated with the -225 to -52 probe in the absence or presence of various antibodies. Complexes formed were analyzed by EMSA.
Fig. lb shows that none of the antibodies tested affected DNA binding of extracts from uninduced cells.
Neither anti-c-jun, nor anti-c-rel, nor anti-NF-IL6 _ _ Wo 95/31722 r~l" /C-'74 .

antibodies had any effect on the cytokine induced complexes .
~oweYer, anti-p50 and anti-p65, antibodies directed against the two proteins in the NFI~B complex, ;Ihrl; qh~
formation of the complex (lanes 10-13). This was observed:with extracts derived from cells treated or ~:
untreated with estradiol, over ~ the whole period of cytokine induction (Fig. lb de~icts the results at 10 minutes after the induction). With longer exposures, a very weak complex of low mobility was detected, probably resulting f rom a supershif t of ~: the induced complex by anti-p50 and anti=p65.
Although EFc binding activity to the vitellogenin ERE
was not detectable in +/+hDAll extracts, we tested whether the ER was involved in complex formation on the II,-6 promoter fragment. Nuclear extracts of +/+L~All =
cells treated with estradiol (10 nM) and TNF~ and I~-l ~lnM each for 40 min) as indicated or yeast extract rrnt~;n;nr~ recombinantly expressed human wild-type ER
were incubated with the -225 to -52 probe in the absence or presence of anti-ER antibody. Complexes formed were analyzed by EMSA.
Fig. lc shows that independent of cytokine induction or ~estradiol treatment, addition of the a~ti-ER antibody did not ~ignificantly affect a~ny of the complexes, induced or constitutive. In ~ 1tirln/ when yeast extracts rr,nt~;n;ng high concentratiors of recombinant ER were incubated with the II,-6 promoter fragment no ~pecific binding of the ER was ~t~rt~ lanes 7 and 8).
The weak bands observed are urErelated to the ER, since~
they were not affected by the anti-ER antibody.
The result6 from the antibody gel shift experiments were further supported by oligonucleotide competition studies. Nuclear extract ~rom +/+I.DA11 cells t--eated ~
with estradiol (10 n~) and: TNF~ a~d II=1 ~lnM each for 10 min) as indicated were incubated with the -225 to -52 WO95/3172~ _ 2 1 9;2 5 ~ ,5;-r~4 probe in the ab6ence or presence of a ~00-fold molar excess of oligonucleotides corr~esponding to the regions of -82 to -47 and -172 to -131 of the human IL-6 promoter. Complexes formed were analyzed by EMSA.
The arrows in Fig. ld indicate the complexes formed upon induction with TNFc: and IL-1. Inclusion of an oligonucleotide covering the NF-IL6 site, the CRE, and an adjacent CCAAT-box of the IL-6 promoter (-172 to -131) in tllf' h;n~1;ng reaction in 40t~=fold excess over the labeled -225 to -52 fragment did not affect any of the complexes, constitutive or cytokine-induced (lanes 4 and 7). However, an oligonucleotide covering the putative NFKB site and adjacent se~uences (-82 to -47) specifically abolished the formation of the cytokine induced complexes (lanes 3 and 6).
The antibody experiments and the ol; ~,~n~ l eotide competition studies suggested that IL-1 and TNF~
specifically activated NFI~B or related proteins. No binding of c- jun (AP-1), NF-IL6, c-rel, or ~ER was detected.
The lack of ~F-IL6 binding is surprising, since induction and binding of this transcription factor in response to IL-1 has been reported for other cells (Akira, S ., H. Isshiki , T . Sugita, O . Tanabe, S .
Kinoshita, Y. Nighio, T. Nakajima, T. Hirano, and T.
Kishimoto (1990) "A nuclear factor for IL-6 expression (NF-IL6) is a member of a C/EBP family, " BMBO J. 9 :18g7-1906; Inoue, J., L.D. Rerr, L.J. Ransone, E. sengal, T.
Hunter, and I .M. Verma (1 991) '~c-rel activates but v-rel suppresses transcription from kappa B sites, ~ Proc.
Natl. Acad. Sci. ~SA BB:3715-3719) . Our DNA binding experiments show that in the bone marrow derived +~+LDA11 cell IL-1 and TNFtY induce the binding of NFI~B
or closely related proteins to the IL- 6 promoter .
Similar results have been obtained in different~cell types (~. Shimizu, K. Mitomo, T. Watanabe, S. Okamoto, Wo 9s/3l722 r~ r74 and K. Yamamoto (1990) "Involvement of a NF-kappa B-like transcription factor in the activation of the ~ _ interleukin-6 gene by inflammatory 1y~rphr~k;n~q~ ~ Mol .
Cell Biol. 10:561-~68; Zhang, Y.H., J.X~ Lin, and J.
Vilcek (1990) "Interleukin-6 induction by tumor necrosis factor and interleukin-1 in human fihr~hlilqtc involves activation of a nuclear factor binding to a kappa B-like se~uence, 1I Mol . Cell Biol. 1Q :3818-3B23~ Neither =
induced nor uninduced binding of æeveral other f actors with potential binding sites in the proximal IL- 6 promoter=fragment~ could be detected, including AP-1.
This transcription factor is one of the paradigms for direct inhibition by intracellular receptor in~ lllrl;ng GR, RAR and TR The mechanism of AP-1 inhibition can involve protein-protein interaction and/or competition for DNA binding depending on the particular gene (Diamond, M. I ., J.N. Miner, S.K. Yoshinaga, and ~.R.
Yamamoto (1990) "Tranæcription factor interactions:
selectors of positive or negatlve regulation from a single DNA element," Science 249:1266-1272; Schule, R., K. Umesono, D.J. Mangelsdorf, J. Bolado, J.W. Pike, and R.M. Evans (1990) "Jun-~os and receptors for vitamins A
and D recognize a common response element in the human osteocalcin gene, " Cell 61:497-504; Yang-Yen, H.F., J.C.
Chambard, Y.L. Sun, T. Smeal, T.J. Schmidt, J. Drouin, and M. Karin (1990) "Transcriptional interferenc~
between c-Jun and the glucocorticoid receptor: mutual inhibition of DNA binding due to direc~ protein-protein interaction," Cell 62:1205-1215; Ya~g-Yen, H.P., X.K.
Zhang, G. Graupner, M. Tzukerman, B. Sakamoto, M. Karin, and M. Pfahl (1991) "Antagonism between retinoic acid -receptors and AP-1: implications for tumor promotion and ;nfli tion," New Biol. 3-12Q6-1219; Zhang, X.K., K.N
Wills, M. Husmann, T. Hermann, and M. Pfahl ~1991) ~~
"Novel pathway for thyroid hormone receptor action through interaction with 3 un and f 08 oncogene Wo 9~/31722 1 ~ ,.S:C '~
.

activitiç~ Mol. Cell Biol. 11:6016-6025) . Several reports al80 suggest a cross-talk between ER and AP-l, however there is no evidence ~or estrogen dependent i~hibition of AP-l activity (Gaub, M. P., M. Bellard, I .
Scheuer, P. Chambon, and P. Sa8sone-Corsi (1990) "Activation of the ovalbumin gene by the estrogen receptor involves the fos- jun complex, " Cell 63 : 1267-1276; Tzukerman, M., X.K. Zhang, and M. Pfahl (1991) ~ Inhibition o~ estrogen receptor activity by the tumor promoter 12-0-tetradeconylphorbol-13-acetate: a molecular analysis, " Mol. Endocrinol. 5:1933-1992) .
Taken together, it is highly unlikely that AP-l plays a role in the negative regulation of IL-6 expression by estrogen .
F~ le 3. Scn~n;nq aqents that differentiallv a~fect dist;nrt com~l,çxes with the IL-6 ~romoter Di8t;nrtive com~l~xes with the IL-6 ~romoter Treatment o~ +/+LDAll cells with IL-l and TNF~
induced binding of NFKB or related proteins to the IL- 6 promoter.= Since pretreatment with estradiol not only increased the intensity but also the mobility of the complexes, we investigated the binding of +~'+~DAll nuclear extractS to the oligonucleotide covering the putative NFKB site (-82 to -47).
+/+LDAll cells were pretreated with 10 nM estradiol as indicated. After 24 h the cells were induced with TNF~ and IL-l (l nM each) for various periods of time.
~nrlllrt;~n was stopped by cell lysis and nuclear extracts were analyzed by EMSA using the -82 to -47 IL-6 promoter fragment as probe. Fig. 2a shows that extracts from cells treated with IL-l and TNF~Y exhibited 3 induced complexes (A,B,C) when compared with extracts from untreated cells.

WO95/31722 2 1 9 027 5 ~ 4 .

Over the course of induction (10-120 min) in particular the ~tastest migrating complex tC) decreased in intensity. Interestingly, estradiol pretreatment reduced the intensity of the slowest migrating complex (A) while strongly increasing the intensity of the fastest band (C) . This corresponds to the pattern obtained with the -225 to -52 fragment where complexes seemed to migrate ~aster with estradiol treatment (Fig.
3) .
It is likely that with both fragments analogous complexes were formed; however, only with the shorter oligonucleotide were they completely resolved : ~uclea~
extract from +/+LDAll cells treated with estradiol (10 nM) and TNF~ and IL-~ (lnM each for 4D min) as indicated were ;nrllh~t/~ with the -82 to -47 probe in the absence or presence of a 100 fold molar excess of ~ ~
oligonucleotides corresponding to the regions of -82 to -47 and -172 to -131 of the human IL-6 promoter or the vitellogenin ERE . Complexes f ormed were analyzed by 2 0 EMSA .
All three induced complexes (A,B,C) were specific since their formation was abolished by inclusion oi a 100-fold excess of the unlabeled probe (Fig. 2b, lanes 6 and 10), while the same molar excess of the ~F-IL6 oli~onucleotide (-172 to -131) or the vitellogenin ERE
had no effect (Fig. 2b, lanes 7, 8, 11, and 12).
Fig. 2c shows that when binding of the extracts to the oligonucleotide covering the NF-IL6 si~e (-172 to -131) was investigated, several complexes were detected.
~uclear extract from +/+LDAll~cells treated with estradiol ~10 nM) and TNFol a~d IL-1 ~TnM each for 40 =
min) as inf~ t ~d were incubated with the -172 to -131 probe in the absence or presence of a 100-fold`molar excess the llnl ~h,~l ~rl oligonucleotide. Complexès formed were analyzed by EMSA. The arrows in Fig. 2c indicate 2 1 9027r Wo95131722 ~ r~ J., c~c '74 the complexes A, B, and C formed upon induction with TNF~ and IL~
Two of the ~ Cf~S were specific, since they could be competed with an excess of the unlabeled oligonucleotide (lanes 2, 4, and 6). However, all of the complexes were formed constitutively, ;nrlrrf~n~1~nt of cytokine induction or estradiol treatment, suggesting that they were unrelated to the regulation of IL - 6 expression by IL-l, TNF~, and estrogen.

Screeninq for com~ounds that aE~ect the foxmation of dist i nrt com~lexe~
In more detailed studies we analyze~ the effects of other compounds on the formation of complexes A, B, and C (Fig. 3) . +/+LDAll cells were pretreated with cyclohP~c;m;-l~ (CHX) or the kinase inhibitor ~I7 for 5 min or with estradiol 510 nM~ and/or ICI 164,384 ~loO nM) for 24 h or 60 min before indnction with TNFo! and IL-l (lnM each for 30 min). Treatment was stopped by cell lysis and nuclear extracts were analyzed by EMSA using the -82 to -47 fragment as probe. Arrows indicate the induced complexes A, B, and C. ~ ~
Pretreatment of the cells with the protein synthesis inhibitor cyclrhr-r; m; ~lr before addition of cytokines did not interfere with complex formation (lane 8). This is consistent with the fast induction of binding and has been shown before for the activation of NFKB (Xenkel, T ., T . Machleidt , I . Alkalay, M . Kronke , Y . Ben-Neriah, and P.A. Baeuerle (1993) "Rapid proteolysis of I kappa B-alpha is necessary for activation of transcription factor NF-kappa s, ~ Nature 365:182-185; Sen, R. and D.
saltimore (1986) "Tn~ r;h;l;ty of kappa immunoglobulin enhancer-binding protein Nf-kappa B by a posttranslational m~rh~n;~," Cell 47:921-928) . It has been reported that cyclrh~; m; ~ treatment activates NF K B binding (Sen, R. and D. Ba1timore (1986) WO9~31722 2 1 qO275 P~ ,C '~I

~l rn~ hi l i ty of kappa immunoglobulin enhancer-~indi~g protein Nf-kappa B by a posttranslational mechanism, ~
Cell 47:921-928; Zhang, Y.H, ~r.x. Lin, and J. Vilcek (1990) "Interleukin-6 induction by tumor necrosis factPr and int~rlF~--kin-l in human fibroblasts involves activation of a ruclear factor bi~ding to a kap~a B-like sequence, " Mol. Cell Biol. 10 38I8-3823~_ Elowever, in +/+~DAll cells we did not observe any; n~ t 1 rln by cy~-lr)h~-Yim;de (lane 7) . ~;t;r~ns~lly, pretreatment with H-7, a potent protein kinase C (PKC) i~hibitor (Kawamoto, S . and: H. Xid~ka (1984) "1- (5-I~ioquinolinesulfonyl)-2-methylpiperazine (II-7) i5 a selective inhibitor of protein ki~a~e C in rabbit platelet~:, " Biochem. Bio~hvs. Res. Commun. l25 25a-264)J
did not interfere with induction o~ complex formation by I~-l and TNF~r (lane 10). This suggests that induction o~ complexes A, B, and C ls mediated through a PKC-;n~p~n~ nt pathway and iE co~si5tent with the findiIlg that II,-1 and T~FcY activate NFKB and induce IL-6 independently of PKC (11, 55, 99) .
Activation of NFKB by phorbol esters is~ probably mediated by PKC (Baeuerle, P.A.- and D. Baltimore (1988) ~I kappa B: a specific inhibitor o~ the D~F-kappa B
transcription factor, " Science 242:54Q-546)_ E~owever, treatment of +/+~DA11 cells with 12-O-tetradeconylphorbol-~ t~ (TPA) did not signif icantly induce production o E IL- 6, nor did it induce complex formation with the -82 to -47 fr;33 The estradiol e~fect, i.e. decreasing complex A and increasing compleY C (Fig. 3, compare lanes~ 2 and 3), was not seen with a short estradiol pretreatment (60 min) before induction (lane 6~ addition, the pure anti-estrogen ICI 164,384 (Wakeling, A.E. and J. Bowler (1988) "Biology and mode of action of pure~ _ ~
antioestrogens," J. Steroid siochem. 30:141-147) did not affect the complex pattern (lane 4) . However, when ICI

WO 95/3l722 2 1 9 0 2 7 5 , ~u~ C~c ~~4 164,3~4 was added in combination with estradiol it prevented the e+fect mediated by the estrogen (lane 5).
Since ICI 164, 384 ac~s as an alltagonist via binding to the ER, these results further support the hypothesis that the effects of estradiol on the induced complexes are receptor mediated. E~owever, the mechanism of estrogen action is probably indirect, as indicated by the lack of response to short term estradiol treatment.
Screen~n~ for a~ents th~t ~ffect bin~l;n~ ~-h~racteristics of the ~rotf~inS in ~i~Stin~t com~lexes To inv~sti~te the binding characteristics of the proteins in complexes A, B, and C with the NFKB
oligonucleotide, methylation interference experiments were carried out (Fig. 4). Nuclear +/+~DAll extracts from cells induced with TNFcY and I~-1 (lnM each) were incubated with -~2 to -47 probe that had been labeled either on the upper or the lower strand and sub; ected to limited DMS-methylation. After preparative EMSA, DNA
from complexes A, B, and C and from the unretarded probe (F) was isolated, cleaved with piperidine, and electrophoresed on a 129G denaturin~ gel. The sequence corresponding to the NFKB consensus site is shown boxed, a cryptic AP-1 site is shaded.
On both strands N-7-methylation of the guanine bases within the ~FKB site (-73 to -63, boxed) interfered with complex formation, while methylation of gua~ines fl~nk;n~ the consengus ~site had no observable effect.
The interference pattern for all three complexes (A,B,C) was identical.
The observation that methylation of guanines just outside of the NFKB site (-75, -60, -5~) did not affect the formation of even the largest complex (A) suggests that in all three complexes DNA contacts are made within the same core region. In addition, the lack of D~A
binding interference with methylation of guanines -60, -wo 9~131~22 r~l" ",.~
2 1 qO27~

58 l and -56 strongly argues against any cytokine induced binding of factors to the noncollsensus (TGAGTCT,_ shaded) AP-l site (Tanabe, O., S. Akira, T. Kamiya, G.G. Wong, T. Hirano, and T. Kishimoto (1988) "Genomic structure of the murine IL- 6 gene . High degree conservation of potential regulatory sequences between mouse and humanl "
1. 141:3875-3881) in this region (-61 to -55~.
... . , . . = . ~. , .
Our studies ~how that estrogen affects the formation oE complexes with the IL-6 promoter that involve NFKB
p50 and p65 or ve-ry closely related proteins. Treatment of +/+LDAl~ cells with Il,-1 and TNF~ specifically induced the formation of at least three distinct complexes~ with the NFKB consensus site in the IL-6 promoter. ~1 th~ of various size/ in all three complexes the DNA contacts are restricted to the core se~uence of the NEKB site. The corresponding core sequence of other NFI~B elements is protected by p50 and p65 (Baldwinl A.S. I Jr. and P A. Sharp (1988) "Two transcription factors, NF-kappa B and H2TF1, interact with a single regulatory sequence in the class I major~
histocompatibility complex promoter, " Proc. ~atl . Acad.
Sci. USA 85:723-727; Kieranl M, ~1. Blank, F. Logeat, ,J.
Vandekerckhove, F. IJottspeich/ O. Le Bail, ~.B- - Urbanl p, Rollri 1 cky, P.A. Baeuerle~ and A. Israel (1990) "The DNA binding subunit of NF-kappa B is identical to factor KBFl and homologous to the rel oncogene product, " Cell 62:10Q7-1018; Sen, R. and D. Baltimore (1986) I'Multiple nuclear factors interact with the immunoglobulin enhancer se~uences, " Cell 46:705-716), both of which have been shown to directly interact with DNA (Nolan, G.
P., S. Ghosh, H.C. Liou, P. Tempst, and D Baltimore (1991)_ "DNA binding and I kappa B ;nilihit;r-n of the cloned p65 subunit of NF-kappa~ B, a rel-related polypeptide," Cell 64:961-969;-Urban, M-B., R- Schreck and P.A. Baeuerle (1991) "NF-kappa B contacts Dl~rA by a heterodimer of t~Le p50 and p65 subunit, " EMBO J.

Wo9~/31722 ~ I 9 02, 5 P~ c ~4 10:1817-1825):_ C-rel homodimers and heterodimers with p~O have been shown to bind the NFKB site in the II,-6 promoter (Nakayama, K., H. Shimizu, K. Mitomo, T.
Watanabe, S. Okamoto, and K. Yamamoto (1992) "A lymphoid eell-specific nuelear factor eontaining e-Rel-like proteins preferentially interaete with interleukin-6 kappa B-related motifs whose aetivities are repressed in lymphoid eells, ~ Mol. Cell Biol. 12:1736-1746) .
In addition, this study showed that in lymphoid eells e-r~l or_an immunologically related faetor iB a eomponent of a larger eomplex that binds the NFKB site in the Il.-6 promoter and funetions as a eonstitutive repressor. In +/+LDA11 eells, we eould not deteet any c-rel specific binding activity. A number of other NFKB
1~ unrelated proteins have been shown to bind to NFKB
consensus sites. Those include aA-CRYBP1 (N~l d, T., D.M. Donovan, K. Hamada, C.M. Sax, B. Norman, J.R.
Flanagan, K. Ozato, H. Westphal, and J. Piatigorsky (1990) "Regulation of the mouse alpha A-crystallin gene:
isolation of a cDNA encoding a protein that binds to a cis seque~ree motif ehared with the ma~or histocompatibility complex class I gene and other genes, " Mol. Cell Biol. 10:3700-3708), MBP-1/PRDII-BFI
(Baldwin, A.S., Jr., K.P. ~eClair, H. Singh, and P.A.
Sharp (1990) "A large protein cr~n~;n;n~ zinc finger domains binds to related se~uence elements in the enhancers of the class I major histocompatibility complex and kappa immunoglobulin genes, " Mol . Cell Biol.
10:1406-1414; Fan, C.M. and T. Maniatis (1990) "A D~A-3 0 binding protein containing two widely separated zinc finger motifs that recognize the same DNI~ sequence, ~' Genes Dev. 4:29-42), and AGIE-BP1 (Ron, D., A.R.
Brasier, and J.F. Habener (199I~ "Angiotensinogen gene-inducible enhancer-binding protein 1, a member of a new 3~ family of large nuclear proteins that recognize nuclear Wo 95131722 2 1 q 0 2 7 5 ~ 4 .

factor kappa B-binding sites through a zinc finger ~
motif," Mol. Cell Biol- 11:2837-2395? - _ _ _ It has been shown that C/EBP-like proteins attenuate NFKB mediated transactivation of the angiotensinogen gene acute-phase response element (Brasier, A.R., D.
Ron, J.E. Tate, and J.F. Babener (1990) "A family of constitutive C/EB~-like DNA binding proteins attenuate the IL-1 alpha induced, NF kappa B mediated trans-activation of the angiotensinogen gene acute-phase response element, " EMso J. 9:3933-3944~ . Currently, we_ cannot exclude that those proteins or others are ~part of the obser~red complexes or are invDlved in the ;n71ihitiDn of I~-6 expression by estrogen. A recent study suggested ~hat in uterine cells, estradiol induced complex formation with an NFKB element (Shyamala, G. and M.C. Guiot (1992) "Activation of kappa B-specific proteins by estradiol, ~' Proc. Natl. Acad. Sci . ~SA
89 :10628-10632) . The induced complex did not contain p50 or p65 and therefore may represent other factors.
F le 4. Screeninq for aqents that affects the hin~1inq of ~65 to ~L-6 ~romoter ~n~lvzinq com~osition of the com~lexes formed wi~h the NFKB site As with the larger promoter~fragment, we analyzed the nature of the complexes formed with the NFKB ~:
oligonucleotide (-82 to -47) by antibody shift experiments (Fig. 5a). Nuclear extract from +/+LDA11 cells treated with estradiol (10 nM) and TNFo~ and IL-l (lnM each for 10 min) as indicated were incubated with the -82 to=-47 probe in the absence or presence of various antibodies. Complexes formed were analyzed by EMS~ .
We observed atrong ef~ects on the induced complexes when anti-p50 or ~nti-p65 (lanes 17-20~ were included in WO 95/31722 2 ~ 9 0 2 75 P~

the binding reactions. Interestingly, anti-p5Q
specifically abolished the formation of complexes B and C, seemed to leave complex A unaffected, and caused the appearance of a single super~hifted band (Sl~. Anti-p65, however, inhibited the formation of all three induced complexes and produced two supershif ted bands (Sl, S2). This suggested that p65 or an immunologically closely related protein is part of all three induced complexes, while p50 or a related protein is only present in complexes B and C.
It has been reported that recombinantly expressed c-rel binds to the NFI~B site in the lL-6 promoter a~
heterodimer with p50 andr with particular high affinity, as h, - ~ir-~r (Nakayama, K., H. Shimizu, K. Mitomo, T.
Watanabe, S. Okamoto, and K. Yamamoto (1992) "A lymphoid cell-specific nuclear factor cnnt;~;n;n~ c-Rel-like proteins preferentially interacts with interleukin-6 kappa B-related motifs whose activities are repressed in lymphoid cell~," Mol. Cell Biol. 12:1736-1746). When anti-c-rel was included in the binding reactions with the +/+LDAI1 extracts the antibody did not inhibit any of the induced complexeæ. On longer exposures a weak supershif ted complex was detectable . This complex migrated at the same position aæ the supershift observed with anti-p50 suggesting that it did not contain the larger c-rel protein. Since the peptide used to raise the anti-c-rel a~tibody has a 56~ homology to the analogoul3 p50 ~equence (Ghosh, S., A.M. Gifford, L.R.
Riviere, P. Tempst, G.P. ~olan, and D. Baltimore (1990) "Cloning of the p50 DNA binding subunit of NF-kappa B:
homology to rel and dorsal, ~' Cell 6Z:1019-lC29; Inoue, J., L.D. Kerr, L.J. Ransone, E~. Bengal, T. Hunter, and I.M. Verma (1991) "c-rel activates but v-rel suppresses transcription from kappa B sites, " Proc. Natl. Acad.
Sci. USA 88 :3715-3719), it is likely that this weak band is the result of a cross-reactivity and u~related to c-Woss~3l722 2l qn~75 ~ '0-~4 _ rel. As with the larger promoter fragment, anti-c-jun did not affect complex formation (lanes 13 and 14~.
Results shown depict the 10 min induction time point and are essentially identical with longer cytokine treatments.
None of the antibodies tested had a marked efi'ect on complex formation with the lln~nfll~r,~ extracts. Only anti-p50 produced a very weak supershifted complex migrating at the S1 position as observed with the induced extracts ~only visible on longer exposures).
This weak binding activity, only detectable when supershifted, could either reslllt from a cytosollc rr,nt~m;niltion or repregent ba6al activation under the culture conditions.
In additio~al EMSA experiments we included a purified preparation of r~l ' ;n~nt p50 as well as the yeast expressed ER (Fig 5b~. Nuclear +/+LDA11 extracts from cells pretreated with estradiol (10 nM) ~nrl ; ndllr~,1 with TNFa and IC-1 (lnM each for 30 min) as indicated as well as purified human p50 protein and yeast extract rt~nt~;n;nrj recombinantly expregged human ER were incubated with -~2 tQ -47 probe in the absence or presence of various ~nt; hnrl; ~ Complexes ~ormed were analyzed by EMSA. Arrows indicate complexes A, B, and C
lnduced by cytokine treatment and the complexes S1 and S2 resulting from the antibody supershifts.
As expected, neither did yeast expressed ER bind to the NFKB -82 to -47 fragment (lanes 5, 9, 13, and 17) nor was ER involved in the formation of the induced ,~
complexes as indicated by the lack of any anti-ER
antibody effect (lanes 14-17). Purified p50 bound to this fragment and was specifically supershifted by anti-p50 but not by anti-p65 (compare lanes 4, 8, and 12).
Surprisingly, the complex~ formed with purified r~ 1 ;n~nt p50 migrated slower than_complexes B and C
~sing lower concentrations of purified p50 did llot W095/31722 21 qO275 r~ .,,s,t e~4 .

affect the migration, 6uggesting that the band represented p50 h' ~;r-rs and not higher order complexes (Duckett, C.S., N.D. Perkins, T.F. Kowalik, R.M. Schmid, E.S. Huang, A.S. Baldwin, Jr., and G.J.
Nabel ~1993) "Dimerization Qf NF-KB2 with RelA~p65) regulates DNA binding, transcriptional activation, and inhibition by an I kappa B-alpha (MAp-3~, " Mol . Cell Biol. 13:1315-1322). However, the antibody shift experiments suggested that both NFKB proteins, p50 and p65, are part of complexes B and C (lanes 6, 7, 11, and 12 ) and consequently both complexes should migrate slower than p50 homodimers (Urban, M.B., R. Schreck, and P.A. BaeuerIe (1991) "NF-kappa B contacts DNA by a heterodimer of the p50 and p6S subunit, " EMB0 ;r.
10:1817-1825).
It is possible that the proteins in complexes B and C are only immunologically related to p50 and p65 but actually of smaller size. Speculations that the p50 homologue p4 9 is part of the induced complexes a~d is responsible for the faster migration could not be ~ nf; ~1. Although purified p49 bound the -82 to -47 IB-6 fragment ana strongly cross-reacted with the anti-p50 antibody, the complex formed migrated even more slowly than the p50 complex. This corresponds to results obtained with other NFKB binding sites (Duckett, C.S., ~.D. Perkins, T.F. Kowalik, R.M. Schmid, E.S.
Huang, A.S. Baldwin, Jr., and G.J. Nabel 11993) "Dimerization of NF-KB2 with RelA(p65) regulates DNA
binding, transcriptional activation, and inhibition by an I kappa B-alpha (MAD-3), " Mol . Cell Biol. 13 :1315-13 22 ) .
However, the finding that the anti-p50 antikody strongly cross-reacts with p49 indicates that the antibodies used may detect other NFKB r~lated proteins in the complexes. Alternatively, the migration of complexes B and C could be higher than the migration _ _ _ _ _ _ . _ _ . .. . _ . . . .. . .

WO95/31722 2 1 90275 r~ elc~
I~

observed with reco.mbinant p50 due to conformational differences resulting from post-translational I ~; f; r:~tion. This would corre-=late with the observation that the ~clusion of anti-p50, abolishing complexes B
and C, produced a supershifted=complex (Sl) migrating more slowly than the supershif ted complex obtained with recombinant p50 (Fig. 5b, lanes 6-8). .
A closer inspection of the band shift results indicated that an-ti-p50 also affected complex A._As discussed before, treatment with estradiol not only increased the ir,t~nsity of complex C but also decreased the intensity of complex A (lanes 2 and 3). We consiste~tly observed that ;nrll~;nn of anti-p50 had a very similar effect: the antibody sprr;f;r~1ly decreased the intersity of complex A for~ned with the extracts from cells not treated with estradiol, resulting in e~ual intensity of this band using extracts from estradiol treated or untre~ted cells (Fig. 5b, compare lanes 2 and 7) . These results suggest that band A induced by IL-1 and TNF~ in the àbsence of estradi~l is composed of two differe~t unresolved complexes, one (Al) that is also induced ir, the presence of estradioL and does not contain p50, and another complex (A2) rnntA;n;n~ p50 (or _n immunologically related protein).
Treatment with estradiol may increase the intensity of complex C at the expense of complex A2. If complex A2 represents a transcriptionally more active state, this could explain the inhibitory effect of estradiol Dn IL-6 expression. Recently it has been shown that the TATA ~iIlding protein (TBP or TFIIDt) directly interacts with ~FI~B (Kerr, L.D., L.J. Ransone, P. Wamsley, M.J.
Schmitt, T.G. Boyer, Q. Zhou, A.J. Berk, and I M. Verma (1993) "Association between proto-oncDprotein Rel and =
TATA-binding protein mediates ~transcriptional activation by NF-kappa B, " Nature 365:412-419) . Therefore, we were interested if TB was involved in the formation of the -Wo 9513l722 21 9 ~ ~ 7 ~ P~~ v c~ ~1 induced complexes. However, using an anti-TBP antibody we could not detect any participation of TBP in complexes A, B, or C.
Our antibody gel shift experiments suggested that p65 is a componen~ of all three observed complexes.
Thi3 particular protein is the NFKB component cnn~i:l;nin~
the tran6activation domain (Schmitz, M.L. and P.A.
Baeuerle ~199I) "The p65 subunit is responsible for the strong transcription activating potential of NF-kappa B, " EMBO J. 10:3~305-3v17) . Within the different complexes the transactivation function may be differentially active. The antibody shift experiments suggest that estradiol ~1; m; ni ~ the A2 complex while increasing complex C. The slow migrating A2 complex may contain other factor (8) involved in the transactivation process. The TATA-binding protein TBP, part of the TFIID complex has been reported to interact strongly with c-rel and p65, but not with p50 or p49 (Kerr, ~.D., L J. Ransone, P. Wamsley, M.J. Schmitt, T.G. Boyer, Q.
Zhou, A.J. Berk, and I.M. Verma (1993) "Association between proto-oncoprotein Rel and TATA-binding protein mediates transcriptional activation by NF-kappa B, ~
365:412-419). Xowever, using a TBP-specific antibody we could not detect TBP as part of any of the complexes formed with the NFKB site in the ITv-6 promoter .
Examrle 5. Ef~icacy-testinq of Putative CYtnk;n Modulators Methods for testing the efficacy of putative cytokine modulators are provided. Each candidate compound is tested for its efficacy in modulating cytokine expression in cell lines, in animal models, and in controlled clinical studies using methods known to those skilled in the art and approved by the Food and Drug Administration, such as, but not limited to, those WO 95131722 2 1 9 ~ 2 7 5 r~ s ~ ~4 promulgated in The Federal Register 47 ~no._~6): 1255a-12564, March 23, 1982 ~lr;lmnle 6. ToxicitY-testinq of Putative Cytokine Modulato~s:
Methods are provided for determining whether; an agent active in any of the methods listed above has little or no effect on healthy :cells. Such agents are then formulated in a~ pharmaceutically acceptable buffer or in buifers useful for standard animal tests.
By "pharmaceuticall~Y acceptable buifer~ is meant any buffer ~hich can be used in a ?harmaceutical ~
composition prepar~ed for storage and subsequent adminis-tration, which comprise a pharmaceutically effective amount oi an agent as described herein in a pharmaceuti-cally acceptable carrier or diluent. Acceptable carriers or diluents for therapeutic use are well known in the pharmaceutical art, and are deæcribed, f or example, in Remi~qtsn' s Pharmaceutical Sciences, - Mack Publishing Co. (A.R. 13ennaro edit. 1985). Preserva-tives, stabilizers, dyes and even flavoring agen~s may be provided in the pharmaceutical composition. For example, sodium benzoate, sorbic acid-and esters of p-hydroxybenzoic acid may be added as preservatives. Id.
at 1449. In addition, ilnti~n~ ntq and suspendi~g agents may be used. Id.
. Additional sC~ nq for Toxicitv: Method 1 Agents identi~ied as having cytokine modulating activity are assessed ~or toxicity to cultured human cell6. This assessment is based on the ability of liv-ing cells to reduce 2, 3, -bis [2-methoxy-4-nitro-5-6ulpho- -nylphenyl] -~- [ (phenylamino) carbonyl] -2~-tetrazolium hydroxide] otherwise referred to as XTT_~Paull et al.,=
J. HeterocYl. Chem. 25763-767 (1987), Weislow et al., (1989), J. Natl. Canc. Inst, 81:577). Viable ~ n cells are capable of reductive cleav3ge o~ an N-N bond Wo 9Sl31722 2 1 9 0 2 7 5 r~ c r~4 in the tetrazole ring of XTT to ~orm XTT formazan. Dead - cells or cells with impaired energy metabolism are inca-pable of this cleavage reaction. The extent Qf the - cleavage is directly proportional to the number of liv-ing cells tested. Cells from a human cell line such as HeLa cells are~-6eeded at 103 per well in 0.1 ml of cell culture medium (Dulbecco' 8 modified minimal essential medium supplemented with 10~ fetal calf serum) in the wells of a 96 well microtiter plate. Cells are allowed to adhere to the plate by culture at 37 C in an Atmr,crhl~re of 9596 air, 596 CO2. After overnight culture, solutions of test substances are added in duplicate to wells at rr,nrrntrations that represent eight half-decade log dilutions. In parallel, the solvent used to dis-solve the test substance is added in duplicate to other wells. The culture of the cells is cnnt;nll~ri for a peYiod of time, typically 24 hours. At the end of that time, a solution of XTT and a coupler (methylphenazonium sulfate) is added to each of the test wells and the in~rllh~t;nn is cnntinll~rl for an additional 4 hours before the optical density i~ each of the wells is determined at 450 nm in an automated plate reader. Substances that kill 1 i;ln cells, or impair their energy metabolism, or slow their growth are detected by a reduction in the optical density at 450 nm in a well as compared to a well which recei~red no test substance.
3. Additional screens for Tn~;citY: Method 2 Cytokine modulators are tested f or cytotoxic effects on cultured human cell iines using incorporation of ~ 35S methionine into protein as an indicator of cell viability. ~eLa cells are grown in 96 well plates in Dulbecco' s minimal essential medium supplemented with lO9~i fetal calf serum and 50,~1g/ml penicillin and streptomycin. Cells are initially seeded at 103 cells/well, 0.1 ml/well. Cells are grown for 48 hrs without exposure to the cytokine modulator, then medium _ WO 9~i131722 2 1 9 0 ~ 7 5 r~l,L~ ~ c '?4 is removed and varying dilutions of the cytokine mn,~ tnr prepared in completé medium are added to each well, with control wells receiving no cytokine modulator. Cells~ are incubated for an additio~al 48-72 hrs. Medium is changed every 24 hrs and replaced with fresh medium cnnt~;n;ng the same concentration of the cytokine modulators. Medium is then removed and replaced with complete medium without antifungal. Cells are incubated for 24 hr in ~he absence of cytokine .
modulator, then viability is estimated by the incor-poration of 35S into protein Medium is removed, re-placed with complete medium without methionine, and ~_ incubated for 3=0 min. Medium is again removed, and replaced with complete medium without met~ nin~ but ,-.mt~;n;ng 0.1 ~Ci/ml 35S methionine. Cells are incubat-ed ior 3 hrs . Wells are washed 3 times ~ in PBS, then cells are permeabilized by adding 10096 methanol for 10 min. Ice cold lD~-trichloroacetic acid (TCA) is added to f ill wells; plates are incubated on ice ior 5 min .
This TCA wash is repeated two more times. Wells are again washed in methanol, then air dried. 50~1 of scin-tillation cocktail are added to each well and dried onto the wells by centrifugation. Plates are used to expose X ray film. Densitometer scanning of the autor~ r~m, including wells without antifungal, is used to determine the dosage at which 50~6 of ce~ls are not vlable (IDso) ~
(Culture of Animal Cells. A manual of basic technique.
(1987). R. Ian Freshney. John Wiley ~ Sons, Inc., New York) .
F~mnle 7. Administration of ~ytokine Modulators The invention features novel cytokine I ~ tnrc:
discovered by the methods described 3bove. It also includes novel pharmaceutical compositions which i~clude cytokine modulato~s discovered as described above formu-lated in pharmaceutically acc~ptable fo~mu1ations.

Wo 9S/31722 2 1 9 0 2 7 5 r~ .c - ~4 Furthermore, the invention features a method for treating~a subject inflicted with a pathological condition affected by the level of a cytokine by administering to that subj ect a therapeutically effective amount of a cytokine modulator. Such ad-ministration can be by any method known to those skilled in the art, ~or example, by topical application or by systemic administration.
~y ~therapeutically effective amount" is meant an amount that relieves (to some extent) one or more symptoms of the disease or condition in the patient.
Additionally, by "therapeutically effective amount'~ is meant an amount that returns to normal, either partially or completely, physiological or biochemical parameters associated with or causative of a mycotic disease or condition . (~ n~ l l y, it is an amount between about 1 nmole and 1 ~mole of the molecule, dependent on its ECso and on the age, size, and disease associated with the patient .
Other embodiments of this invention are disclosed in the following claims.

Claims (24)

WHAT IS CLAIMED IS

1. Method for screening for a therapeutic agent for treatment of a pathological condition affected by the level of a cytokine, comprising the steps of:
contacting a potential therapeutic agent with a system comprising an intracellular receptor, a promoter or a portion of said promoter with a rel site, and a protein that binds to said rel site on said promoter;
measuring the binding of said protein to said rel site on said promoter; wherein a reduction in the binding of said protein to the rel site on the promoter compared to the binding of said protein in the absence of said agent is an indication that said agent is potentially useful for treatment of said condition.

2. The method of claim 1, wherein said protein is a rel-like protein.

3. The method of claim 2, wherein said rel-like protein is NF?B.

4. The method of claim 1, wherein said system further comprises a ligand for said intracellular receptor.

5. The method of claim 1, wherein said condition is osteoporosis.

6. The method of claim 1, wherein said condition is rheumatoid arthritis.

7. The method of claim 1, wherein said condition is inflammation.

8. The method of claim 1, wherein said condition is psoriasis.

9. The method of claim 1, wherein said condition is Kaposi's sarcoma.

10. The method of claim 1, wherein said condition is septic shock.

11. The method of claim 1, wherein said condition is multiple myeloma.

12. The method of claim 1, wherein said intracellular receptor is a steroid receptor.

13. The method of claim 1, wherein said intracellular receptor is an estrogen receptor.

14. The method of claim 1, wherein said intracellular receptor is selected from the group consisting of retinoid acid receptors, retinoid X
receptors, glucocorticoid receptor, progesterone receptors, androgen receptor, thyroid hormone receptors, and vitamin D receptor.

15. The method of claim 1, wherein said measuring comprises determining the expression level of a cytokine or an acute phase protein.

16. The method of claim 1, wherein said measuring comprises determining the expression level of a reporter gene linked to said promoter.

17. The method of claim 1, wherein said system further comprises an effector of said promoter.

18. The method of claim 17, wherein said effector is selected from the group consisting of tumor necrosis factor, interleukin-1, viruses, endotoxin, phorbol esters, epidermal growth factor, leukemia inhibitor factor and cAMP agonists.

19. The method of claim 1, wherein said cytokine is interleukin 6.

20. The method of claim 1, wherein said cytokine is interleukin 8.

21. The method of claim 1, wherein said system is a cell .

22. The method of claim 1, wherein said system is an extract of a cell.

23. The method of claim 21, wherein said intracellular receptor is expressed from a transfected vector .

24. The method of claim 21, wherein said promoter or said portion of said promoter is transfected into said cell.