WO2002095414A1

WO2002095414A1 - Methods for screening using interleukin soluble trimolecular complex

Info

Publication number: WO2002095414A1
Application number: PCT/EP2002/005271
Authority: WO
Inventors: Nicolas Gee; Jason Brown; Francois Bertelli
Original assignee: Warner-Lambert Company Llc
Priority date: 2001-05-18
Filing date: 2002-05-14
Publication date: 2002-11-28
Also published as: EP1395832A1; JP2004533612A; GB0112251D0; GB2375604A

Abstract

The present invention relates to screening methods, assays, and reagents based on the unexpected discovery that interactions between one member of the interleukin family (IL), the corresponding membrane-bound interleukin receptor (IL-R) and membrane-bound interleukin accessory protein (Il-RAcP), which occur in vivo at the cell surface, can be modelled in solution. This provides for assays suitable for high throughput screening.

Description

METHODS FOR SCREENING USING INTERLEUKIN SOLUBLE TRIMOLECULAR COMPLEX

FIELD OF INVENTION

The present invention relates to screening methods, assays and reagents based on the unexpected discovery that interactions between interleukin (IL) , the membrane-bound interleukin receptor (IL-R) and membrane- bound interleukin receptor accessory protein (IL-RAcP) , which occur in vivo at the cell surface, can be modelled in solution. This provides for assays suitable for high throughput screening.

BACKGROUND OF THE INVENTION

Interleukin-1 (IL-1) plays a central role in the mediation of immune and inflammatory responses (Dinarello, 1996) .

The IL-1 may be: IL-lα or IL-1/3 receptor agonists which have comparable biological activities, or IL-1 receptor antagonist IL-lra. These cytokines are involved in a large number of biological effects such as fever, sleep, anorexia or hypotension by binding to specific receptors on the surface of responsive cells.

Two IL-1 receptors with different pharmacological characteristics, termed type I (Sims et al . 1988) and type II (McMahan et al . 1991) have been cloned. IL-lc, IL-1/3 and IL-lra all bind with comparable affinity to the 80kDa type I receptor (IL-1RI) (Dinarello, 1996) . On the other hand, IL-1/3 binds with much higher affinity and selectivity to the 68kDa type II receptor (IL-1RII) than the IL-1 receptor type I (Colotta et al^_.993). Furthermore IL-1RI is necessary foι ' IL-1 signal transduction (Sims et al. 1993) ..ΓL-ΪRII appears to act as a decoy receptor (Colotta et al . 1993).

Another component of the receptor complex, Interleukin 1 receptor accessory protein (IL-lRAcP) , has also been cloned (Greenfeder et al . 1995) . The IL-lRAcP forms a ternary complex with IL-1RI and either IL-lα or IL-1/3, but not with IL-lra. Formation of this trimolecular complex increases the binding affinity of IL-1/3 for IL- 1RI . Although IL-1RI and IL-1RII may bind IL-1, IL- lRAcP does not bind IL-1 (Greenfeder et al . 1995). However transfection with IL-lRAcP restores IL-1 responsiveness in mammal cells not expressing IL-lRAcP

(Korherr et al . 1997). Therefore although IL-lRAcP is not required for the binding of IL-1 to IL-1RI, it is essential for IL-1 signal transduction.

Endogenous cell-surface receptor and accessory proteins are membrane-bound and contain hydrophobic transmembrane domains . Such proteins are not soluble and are not suitable for use in in vi tro high throughput screening.

SUMMARY OF THE INVENTION

The present invention provides screening methods, assays and reagents based on interactions between interleukin

(IL) , the membrane-bound interleukin receptor (IL-R) and membrane-bound interleukin receptor accessory protein

(IL-RAcP) modelled in solution. In one general aspect, the present invention provides an assay method for determining the ability of a test compound to modulate the formation of a trimolecular complex including IL, a soluble IL-R and a soluble IL- RAcP, the method comprising:

(a) bringing into contact an IL polypeptide, a soluble IL-R polypeptide, a soluble IL-RAcP polypeptide and a test compound; and (b) determining the amount of said trimolecular complex formed.

In a most preferred embodiment the present invention also provides an assay method for determining the ability of a test compound to disrupt or interfere with the formation of a trimolecular complex including IL, a soluble IL-R and a soluble IL-RAcP, the method comprising:

(a) providing a test compound; (b) bringing the test compound into contact with a defined amount of a soluble IL-R polypeptide and a soluble IL-RAcP polypeptide; (c) adding an IL polypeptide to the mixture obtained in test (b) ; and (d) determining the amount of said trimolecular complex formed.

Another general aspect of the present invention provides an assay method for determining the ability of a test compound to enhance the formation of a trimolecular complex including IL, a soluble IL-R and a soluble IL- RAcP, the method comprising:

(a) bringing into contact an IL polypeptide, a soluble IL-R polypeptide, a soluble IL antagonist, a soluble IL-RAcP polypeptide and a test compound;

(b) determining the amount of said trimolecular complex formed; and

(c) comparing the amount of said trimolecular complex formed at step (a) with the amount of said trimolecular complex formed in the absence of test compound.

A preferred embodiment of the present invention is an assay method for determining the ability of a test compound to enhance the formation of a trimolecular complex including IL, a soluble IL-R and a soluble IL- RAcP, the method comprising:

(a) providing a test compound;

(b) bringing the test compound into contact with a soluble IL-R polypeptide, a soluble IL antagonist and a soluble IL-RAcP polypeptide; (c) adding an IL polypeptide to the mixture obtained at step (b) ;

(d) determining the amount of said trimolecular complex formed; and

(e) comparing the amount of said trimolecular complex formed at step (c) with the amount of said trimolecular complex formed in the absence of test compound.

The present invention also concerns an assay method for determining the ability of a test compound to enhance the formation of a trimolecular complex including IL, a soluble IL-R and a soluble IL-RAcP, the method comprising : (a) bringing into contact an IL polypeptide , a soluble IL-R polypeptide , an IL-R antagonist , a soluble IL-RAcP polypeptide and a test compound; (b) determining the amount of said trimolecular complex formed; and

(a) providing a test compound;

(b) bringing the test compound into contact with an IL polypeptide, an IL-R antagonist and a soluble IL-RAcP polypeptide;

(c) adding a soluble IL-R polypeptide to the mixture obtained in step (b) ;

(d) determining the amount of said trimolecular complex formed; and (e) comparing the amount of said trimolecular complex formed at step (c) with the amount of said trimolecular complex formed in the absence of test compound.

A further aspect of the present invention relates to an assay method for determining the ability of a test compound to disrupt or interfere with the stability of a trimolecular complex including IL, a soluble IL-R and a soluble IL-RAcP, the method comprising:

(a) providing a defined amount of a trimolecular complex comprising an IL polypeptide, a soluble IL-R polypeptide and a soluble IL-RAcP polypeptide;

(b) contacting a test compound with said trimolecular complex; and

(c) comparing the amount of said trimolecular complex present after step (b) to the amount of said trimolecular complex initially present at step (a) .

In a preferred assay method of the invention the IL, sIL-R and sIL-RAcP are of mammalian origins.

In a most preferred assay method of the invention the IL, sIL-R and sIL-RAcP are from human, mouse or rat.

A general aspect of the invention provides an assay method wherein the IL polypeptide is in its mature form and has a barrel shaped 3 -dimensional structure composed of 12 to 13 beta- strands and a cellular activity mediated by a ternary (i.e. trimolecular) complex at the cell surface.

In a preferred embodiment of the invention, the assay method is an assay method wherein the IL polypeptide has an amino acid sequence of SEQ ID NO: 1, or SEQ ID NO: 4, or SEQ ID NO: 5, or SEQ ID NO: 6, or SEQ ID NO: 7, or SEQ ID NO: 8, or SEQ ID NO: 9, or SEQ ID NO: 10, or a fragment thereof . Another general aspect of the invention concerns an assay method wherein the soluble IL-R polypeptide comprises 3 Ig-like domains which are included in three structural domains identified as domain 1, domain 2 and domain 3 starting from the N-terminal end of the sequence, said domain 1 containing one Ig-like domain and two disulfide bonds; said domain 2 containing one Ig-like domain and two overlapping disulfide bonds; said domain 3 containing one Ig-like domain and one disulfide bond.

In a preferred embodiment of the invention, the assay method is an assay method wherein the soluble IL-R polypeptide has an amino acid sequence of SEQ ID NO: 13, or SEQ ID NO: 14, or SEQ ID NO: 15, or SEQ ID NO: 16, or SEQ ID NO: 17, or SEQ ID NO: 18, or a fragment thereof.

A further general aspect of the invention relates to an assay method wherein the soluble IL-RAcP polypeptide comprises 3 Ig-like domains which are included in three structural domains identified as domain 1, domain 2 and domain 3 starting from the N-terminal end of the sequence, said domain 1 containing one Ig-like domain and two disulfide bonds; said domain 2 containing one Ig-like domain and two overlapping disulfide bonds; said domain 3 containing one Ig-like domain and one disulfide bond.

In a preferred embodiment of the invention, the assay method is an assay method wherein the soluble IL-RAcP polypeptide has an amino acid sequence of SEQ ID NO: 21, or SEQ ID NO: 22, or SEQ ID NO: 23, or SEQ ID NO: 24, or SEQ ID NO: 25, or SEQ ID NO: 26, or a fragment thereof. The present invention also concerns a soluble trimolecular complex comprising or consisting of the following elements:

(a) an IL polypeptide

(b) a soluble IL-R polypeptide which binds said IL polypeptide, and

(c) a soluble IL-RAcP polypeptide which binds to the IL- R polypeptide/IL polypeptide binary complex.

A preferred soluble complex of the invention is a soluble complex wherein the IL, sIL-R and sIL-RAcP are of mammalian origins.

A most preferred soluble complex of the invention is a soluble complex wherein the IL, sIL-R and sIL-RAcP are from human, mouse or rat.

The most preferred soluble complex is a soluble complex wherein at least one of the IL, sIL-R or sIL-RAcP is from human.

In a particularly preferred soluble complex of the invention the IL polypeptide has an amino acid sequence of SEQ ID NO: 1, or SEQ ID NO : 4, or SEQ ID NO: 5, or SEQ ID NO: 6, or SEQ ID NO: 7, or SEQ ID NO : 8, or SEQ ID NO: 9, or SEQ ID NO: 10, or a fragment thereof.

In another particularly preferred soluble complex of the invention the soluble IL-R polypeptide has an amino acid sequence of SEQ ID NO: 13, or SEQ ID NO: 14, or SEQ ID NO: 15, or SEQ ID NO: 16, or SEQ ID NO: 17, or SEQ ID NO: 18, or a fragment thereof . In a further particularly preferred soluble complex of the invention the soluble IL-RAcP polypeptide has an amino acid sequence of SEQ ID NO: 21, or SEQ ID NO: 22, or SEQ ID NO: 23, or SEQ ID NO: 24, or SEQ ID NO: 25, or SEQ ID NO: 26, or a fragment thereof.

BRIEF DESCRIPTION OF THE DRAWINGS

Figures 1A, IB and 1C show the interaction respectively of IL-lβ (A), sIL-lRI(B) and sIL-lRAcP (C) purified proteins with separate flow cells of the BIAcore chip which have been respectively coupled with sIL-lRI, IL-lβ or sIL-lRAcP.

Figures 2A and 2B show the interaction of previously formed IL-l/3/sIL-lRI complex with an excess of IL-lβ (A) or sIL-lRI (B) being added on separate flow cells of the BIAcore chip which have been respectively coupled with sIL-RI, IL-lβ or IL-lRAcP.

Figure 3A shows the sequential injection of IL-lβ and sIL-lRAcP on separate flow cells of the BIAcore chip which have been respectively coupled with sIL-lRI, sIL- lβ or sIL-lRAcP-6His . Figure 3B shows the capture of sIL-lRAcP-6His on a NiNTA chip and injection of premixed IL-lβ/sIL-lRI binary complex or IL-lβ alone.

Figure 4 shows a comparison of the IL-1/3 and IL-lra interactions with separate flow cells of the BIAcore chip which have been respectively coupled with IL-lβ, sIL-lRAcP or sIL-lRI when sIL-RAcP is added during the dissociation process.

Figures 5A and 5B show the binding of the FEWTPGYWQPYALPL peptide (SEQ ID N° 35) on separate flow cells of the BIAcore chip which have been respectively coupled with sIL-lβ, sIL-lRAcP or sIL-lRI. In figure 5A the FEWTPGYWQPYALPL peptide is added alone. In figure 5B a premixed FEWTPGYWQPYALPL peptide/sIL-lRI complex is added with an excess of sIL-lRI.

Figure 6A shows the sequential addition of IL-lβ and sIL-lRAcP on separate flow cells of the BIAcore chip which have been respectively coupled with sIL-lR Type I, sIL-lR Type II or sIL-lRAcP. Figure 6B shows the sequential addition of sIL-lRII and sIL-lRAcP on separate BIAcore flow cells coupled respectively with

IL-lβ, sIL-lRI or sIL-lRAcP.

For figures 1 to 6 the Flow cell 1 is coupled with BSA (Bovine Serum Albumin) and this control flow cell is substracted from the other in the displayed sensorgrams .

Figure 7 shows respectively the binding kinetics of the IL-l/3/sIL-lRI binary complex (figure 7A) and IL-l/3/sIL- IRI/SIL-IRACP ternary complex (figure 7B) .

Figure 8 is a graphical representation of a direct ternary complex HTRF assay format of the invention.

Figure 9 is a graphical representation of a direct binary complex HTRF assay format of the invention.

Figure 10 shows the kinetics of ternary complex formation in a direct ternary complex HTRF assay format of the invention.

Figure 11 shows the influence of various IL-1/3 and sIL- lRAcP-Cy5 concentrations on the signal obtained using a direct ternary complex HTRF assay format of the invention.

Figure 12 shows the influence of various concentrations of unlabelled sIL-lRI and sIL-lRAcP on the signal obtained using a direct ternary complex HTRF assay format of the invention.

Figure 13 shows the influence of various concentrations of IL-lra and the FEWTPGYWQPYALPL peptide (of SEQ ID N° 35) on the inhibition of ternary complex formation in a direct ternary HTRF assay format of the invention.

Figure 14 shows a sequence alignment of the amino acid sequences of IL-lα from human (Genbank accession number: x02531) , mouse (Genbank accession number: NM_010554) and rat (Genbank accession number: DO0403) .

Figure 15' shows a sequence alignment of the amino acid sequences of IL-1/3 from human (Genbank accession number: x02532) , mouse (Genbank accession number: NM_008361) and rat (Genbank accession number: M98820) .

Figure 16 shows a sequence alignment of the amino acid sequences of IL-18 from human (Genbank accession number: AF077611) and mouse (Genbank accession number: NM_008360) .

Figure 17 shows a sequence alignment of the amino acid sequences of IL-1R Type I from human (Genbank accession number xl6896) , mouse (Genbank accession number NM_008362) and rat (full length protein (Genbank accession number m95578) , naturally occuring soluble protein (Genbank accession number NM_013123) ) , IL-1R Type II from human (Genbank accession number NM_004633) and mouse (Genbank accession number NM_010555) , and IL- 18R from human (Genbank accession number NM_003855) and mouse (Genbank accession number NM_008365) . The underlined sequence indicates the signal sequence of the human IL-IRI protein, which is cleaved after protein maturation.

Figure 18 shows a sequence alignment of the amino acid sequences of IL-lRAcP from Human (full length protein

(Genbank accession number AF029213) , naturally occuring soluble protein (Genbank accession number AF167343) ) , mouse (Genbank accession number NM_008364) , and rat

(Genbank accession number NM_012968) and IL-18RAcP from human (Genbank accession number NM_003853) and mouse

(Genbank accession number NM_010553). The underlined sequence indicates the signal sequence of the human IL- lRAcP protein, which is cleaved after maturation of the protein.

Figure 19 shows the sequence alignment of human IL-1 (Genbank accession number xl6896) and IL-18 receptors (Genbank accession number NM_003855). Figure 20 shows the sequence alignment of human IL-1 receptor accessory protein (Genbank accession number AF029213) and IL-18 (Genbank accession number NM_003853) receptor accessory protein.

In figures 19 and 20 the bold and underlined sequence indicates the signal sequences of the proteins; the transmembrane domains are outlined by shaded boxes.

Figures 21 and 22 show hydropathy plots of the full length sequence of human IL-IRI (figure 21A) and IL-18R

(figure 21B) , IL-lRAcP (figure 22A) and IL-18RAcP

(figure 22B) .

Figure 23 is a graphical representation of the structural organization of human extracellular Interleukin-1 receptor type I and Interleukin-1 receptor accessory protein.

Figure 24 shows the sequence alignment of the human Interleukin-1 type I receptor (IL-IRI; Genbank accession number xl6896) and the human Interleukin receptor accessory protein (IL-lRAcP; Genbank accession number af029213) . The first and the last cysteine of each structural domain are indicated by shading and holding.

Underlined portions of sequence ( ) , (_ ) and ( ) indicate the localization of domains 1, 2 and 3 respectively.

Figure 25 is a graphical representation of flashplate assays formats of the method of the invention. Figure 25A shows a Nickel-flashplate format, figure 25B shows a Streptavidin-flashplate format.

DETAILED DESCRIPTION OF THE INVENTION

As used herein the term soluble when applied to a polypeptide means any polypeptide which isn't membrane- bound.

The ability of IL-1/3, IL-IRI and IL-lRAcP to interact in solution is shown for the first time in the present application. It was surprising that the binding activity of the mature membrane-bound proteins IL-1R and IL-lRAcP could be replicated by truncated proteins which lacked the cytosolic and transmembrane domains and comprise only the extracellular domains. The interaction of soluble fragments of these proteins allows the development of assays suitable for high throughput screening, which enable direct measurement of the protein binding characteristics.

The pro-inflammatory cytokines IL-lα and IL-lβ interact with the IL-1 receptor and IL-1 receptor accessory protein at the cell-surface leading to further pro- inflammatory responses via gene activation through the IRAK, NIK, NFKB signalling cascade (Dinarello, 1996) .

Current protein-based therapies include i.v. dosing of the soluble form of the IL-1 receptor and / or the IL-1 receptor antagonist which act as functional antagonists of IL activity (Rosenwasser, 1998) . Limitations of such therapies include the expense of producing the large quantities of these recombinant proteins required for sufficient biological activity and the requirement for i.v. dosing.

The identification of small molecule inhibitors of the IL trimolecular complex and hence of IL-1 biological activity using the assay method described herein may provide a significant advantage over the above routes since it may be possible to dose orally and to reduce the cost of production of such compounds compared to the production cost of recombinant proteins.

Interferon-γ inducing factor or IL-18 is a member of the T helper type I cell (Thl) -inducing family of cytokines and has many structural and functional similarities with IL-lβ. It also forms a ternary complex at the cell surface together with an IL-18-receptor (or ILlRrp; IL-1 receptor related protein also known as IL-18Rα) and an IL-18 accessory protein like receptor (AcPL) or IL-18Rβ (also known as IL-18RAcP) . As with IL-1, IL-18 signalling occurs via the IRAK, NIK, NFKB pathway leading to a pro-inflammatory stimulus (Dinarello, 1999a) .

Crohn's disease is marked by chronic inflammation of the gastrointestinal tract . The upregulation of IFN-γ in Crohn's disease due to overproduction of IL-18 may be involved in the pathogenesis of the disease (Monteleone et al . , 1999). Hence, inhibition of IL-18 activity through an appropriate antagonist represents a potential treatment for this disease. Other potential uses of an antagonist of IL-18 activity include the treatment of asthma, autoimmune demyelinating diseases, rheumatoid arthritis (Gracie et al . , 1999) and psoriasis (Dinarello, 1999b) . There are currently no therapies able to reduce IL-18 activity. Hence, there is a need for small molecule inhibitors, antagonists of IL-18 activity through its interaction with IL-18Rα and IL-18Rβ.

Conversely, potentiation of IL-18 activity may also have clinical benefits. For example, IL-18 appears to have antiangiogenic and antitumour activity on primary T241 fibrosarcomas implanted subcutaneously into C57B16/J mice IL-18 agonists could potentially - but not exclusively - be orally administered. In addition, physiological 'agonists' of IL-18 activity may also be useful therapies for protection against fungal and bacterial infections (Akira, 2000) .

Inhibitor molecules of IL activity may be obtained with the screening assay of the invention, designed and used for therapeutic purposes as described above, for example, for treatment of inflammation-related diseases such as Crohn's disease, osteoarthritis, inflammatory pain, and rheumatoid arthritis.

Activators or "agonists" of IL activity may also be identified and appropriate agents may be obtained, designed and used for therapeutic purposes as described herein above, for example in the treatment of immunodeficiency.

IL trimolecular complex

One general aspect of the present invention is a soluble trimolecular complex consisting of the following elements : (a) an IL polypeptide

(b) a soluble IL-R polypeptide which binds said IL polypeptide , and

(c) a soluble IL-RAcP polypeptide which binds to the IL-R polypeptide/lL polypeptide binary complex .

In the present application the term IL means IL polypeptide, the term sIL-R means soluble IL-R polypeptide, the term sIL-RAcP means soluble IL-RAcP polypeptide.

The three members of the IL ternary complex are highly conserved between mammalian species. As a result, in an assay of the present invention, the components of the ternary complex, IL, sIL-R and sIL-RAcP, may be derived from the same species, eg. human, mouse, rat or may be from different species. For example, human sIL-lRAcP may bind to a bimolecular complex of rat IL-l/murine sIL-lR to form a trimolecular complex.

In a preferred assay of the present invention, the components of the ternary complex, IL, sIL-R and sIL- RAcP, are derived from the same mammalian species. In a most preferred assay of the present invention, the components of the ternary complex, IL, sIL-R and sIL- RAcP, are derived from human, mouse or rat sequences. In the most preferred assay of the present invention, the components of the ternary complex, IL, sIL-R and sIL-RAcP, are derived from human sequences. IL Polypeptide

In the present invention, IL polypeptide is understood as a polypeptide whose mature form has a barrel shaped 3-dimensional structure composed of 12-14 beta-strands and whose cellular activity is mediated by a ternary (i.e. trimolecular) complex at the cell surface. IL polypeptide includes members of the IL-1 and IL-18 families, such as human (Genbank accession number: X02531) , mouse (Genbank accession number: NM_010554) and rat (Genbank accession number: D00403) immature IL-loi, human (Genbank accession number: x02532) , mouse (Genbank accession number: NM_008361) and rat (Genbank accession number: M98820) immature IL-1/3 and human (Genbank accession number: AF077611) and mouse (Genbank accession number: NM_008360) immature IL-18, human (Genbank accession number AF200492) and mouse immature IL-1H1 (Genbank accession number AF200493) , human immature IL- 1H2 (Genbank accession number AF200494) , mouse immature IL-1H3 (Genbank accession number AF200495) , human immature IL-1H4 (Genbank accession number AF200496) , human (Genbank accession number AF230377) and mouse immature IL-lδ (Genbank accession number AF230378) , human (Genbank accession number AJ242738) and mouse immature IL-1L1 (Genbank accession number AJ250429) , human immature FIL1 delta (Genbank accession number AF201830) , human immature FIL1 eta (Genbank accession number XM_002375) , human immature FIL1 zeta (Genbank accession number XM_010759) , human immature FIL1 epsilon (Genbank accession number XM_010757) , human (Genbank accession number AF206696) and mouse (Genbank accession number AF206697) immature IL-1 epsilon. The amino acid sequences of IL-lα and IL-lβ from human, rat and mouse and IL-18 from human and mouse are shown respectively on the sequence alignments of figures 14, 15 ^' and 16.

An IL polypeptide may also be an IL amino acid sequence or a fragment, derivative, analog or active portion thereof which retains the biological activity of IL. What is intended by "biological activity" is to designate the ability of the polypeptide to form a trimolecular complex with the corresponding receptor and receptor accessory protein.

Preferably IL is IL-1 or IL-18.

Preferred IL polypeptides for use in the assays of the invention are mature IL polypeptides which form a soluble ternary complex with a soluble IL receptor and a soluble IL Receptor Accessory Protein, such as human (SEQ ID No: 1), mouse (SEQ ID No: 5) and rat (SEQ ID No: 7) mature IL-lα, human (SEQ ID No: 4), mouse (SEQ ID No: 6) and rat (SEQ ID No: 8) mature IL-1/3, and human (SEQ ID No: 9) and mouse (SEQ ID No: 10) mature IL-18.

Most preferred IL polypeptides for use in the assays of the invention are the human mature IL-lβ sequence of SEQ ID N°4 and the human mature IL-18 sequence of SEQ ID N°9.

hIL indicates a human form of IL. IL-R Polypeptides

• Full length IL-R The mouse IL-1 receptor was cloned by expression cloning techniques (Sims et al . , 1988). This gene encodes a receptor containing an NH₂-extracellular domain, a transmembrane domain, and an intracellular domain. Sequence analysis further revealed the extracellular portion of the IL-1 receptor to be organised into three domains (Ig-like domains, see Figure 23), similar to those of the immunoglobulin gene superfamily. Co- crystallisation of the extracellular domain of the human IL-1 receptor with the cytokine IL-lβ confirmed the above structural prediction for this portion of the protein (Vigers et al . , 1997). X-ray crystal structure of a small antagonist peptide (ETPFTWEESNAYYWQPYALPL, SEQ ID N° 37) bound to the Type-1 Interleukin-1 Receptor was recently disclosed (Vigers et al . , 2000). This peptide, close to the sequence of SEQ ID N°35, mimicked the interaction of IL-lβ and IL-lra with domains 1 and 2 of the soluble IL-1R.

A second group who crystallised the IL-IRI with IL-lra describes three Ig-like domains (Schreuder et al., 1997) with very similar ranges to those described in Vigers et al . (1997 ) .

In the present application, IL-R polypeptide means a polypeptide comprising an intracellular domain, one single transmembrane domain and an extracellular domain. The extracellular domain comprises three Ig-like domains and allows the binding of IL-R to an IL and an IL-RAcP. The signalling form of the IL-R polypeptide is membrane bound.

IL-R polypeptides may include members of the IL-1R and IL-18R families, such as for example, human IL-IRI (Genbank accession number xl6896) , mouse IL-IRI (Genbank accession number NM_008362) , rat IL-IRI (Genbank accession number m95578) , the naturally occuring soluble rat IL-1R type 1 (Genbank accession number NM__013123) , human IL-18R (Genbank accession number NM_003855) , mouse IL-18R (Genbank accession number NM_008365) , human IL- lRrp2 (Genbank accession number AF284434) , mouse IL- lRrp2 (Genbank accession number AF284433) , human TIGIRR- 1 (Genbank accession number AF284436) , mouse TIGIRR-1 (Genbank accession number AF284437) ( Born et al, (2000), JBC, 275, 29946-29954)

Except the naturally occuring soluble rat IL-1R type 1 (Genbank accession number NM_013123) the above defined full length IL-R are not usually suitable for an assay of the invention.

Soluble IL-R Polypeptides

Soluble IL-R polypeptide (sIL-R) means any soluble polypeptide fragment of a full-length IL-R which retains the extracellular binding activity of the full length protein (i.e. the ability to form a trimolecular complex with its corresponding IL and with its corresponding receptor accessory protein which may be soluble or membrane ^' bound) .

Preferred soluble IL-R for use according to the present invention comprise or consist of the extracellular domain of the corresponding endogenous protein or a portion or fragment thereof which retains its binding capability. A suitable portion or fragment binds to the same ligands as the endogenous protein. The extracellular domain of a membrane-bound receptor or accessory protein has been identified by a hydropathy analysis of the mature sequence as described in Example 1. The extracellular domain of IL-R comprises three Ig- like domains which are included in three structural domains identified herein as domain 1, domain 2 and domain 3 starting from the N-terminal end of the sequence. Domain 1 contains one Ig-like domain and two disulfide bonds. Domain 2 contains one Ig-like domain and two overlapping disulfide bonds. Domain 3 contains one Ig-like domain and one disulfide bond. See Figure 23 for domains organization of IL-IRI.

Domain 3 of IL-IRI is critical for high-affinity binding of IL-lβ to the IL-IRI (Schreuder et al . , 1997). Thus a soluble form of the IL-IRI receptor consisting of domains 1 and 2 only has a significantly reduced affinity for IL-lβ (lOμM) .

• Preferred soluble IL-R polypeptides

For example purposes the sIL-R polypeptides described herein are human sIL-lRI and human SIL-18R but with the help of figure 17 one skilled in the art can extrapolate it to other sIL-lR polypeptides.

In one embodiment of the invention, the soluble IL-R polypeptide is a soluble IL-1 receptor polypeptide, preferably a soluble IL-1 type I receptor polypeptide.

In a preferred embodiment of the invention, a soluble receptor polypeptide is a polypeptide which retains the binding properties of the full length receptor and which has at least 60% identity, or at least 80% identity, preferably 85% identity, more preferably 90% identity, most preferably 95% identity with the amino acid sequence of SEQ ID NO: 13.

A most preferred soluble receptor polypeptide is a polypeptide which comprises or consist of an amino acid sequence of SEQ ID NO: 13 or an analog, derivative, active portion or fragment thereof which retains the binding properties of the full length protein.

In further embodiments of the invention when the IL polypeptide is IL-1, a soluble IL-R polypeptide which retains the binding properties of the full length receptor and which has at least 60% identity, or at least 80% identity, preferably 85% identity, more preferably 90% identity, most preferably 95% identity with the amino acid sequence of SEQ ID NO: 14, SEQ ID NO: 15, or SEQ ID NO: 16.

In further preferred embodiments of the invention, when the IL polypeptide is IL- 1 , a soluble IL-R polypeptide is a polypeptide which comprises or consist of an amino acid sequence of SEQ ID NO : 14 or SEQ ID NO : 15 , or SEQ

ID NO : 16 , or an analog, derivative , active portion or fragment thereof which retains the binding properties of the full length protein . In another embodiment of the invention, the soluble IL-R polypeptide is a soluble IL-18R polypeptide.

In a preferred embodiment of the invention, a soluble receptor polypeptide is a polypeptide which retains the binding properties of the full length protein and which has at least 60% identity, or at least 80% identity, preferably 85% identity, more preferably 90% identity, most preferably 95% identity with the amino acid sequence of SEQ ID NO: 17.

A most preferred soluble receptor polypeptide is a polypeptide which comprises or consist of an amino acid sequence of SEQ ID NO: 17 or an analog, derivative, active portion or fragment thereof which retains the binding properties of the full length protein.

In further embodiments of the invention, when the IL polypeptide is IL-18, a soluble IL-R polypeptide which retains the binding properties of the full length receptor and which has at least 60% identity, or at least 80% identity, preferably 85% identity, more preferably 90% identity, most preferably 95% identity with the amino acid sequence of SEQ ID NO: 18.

In further preferred embodiments of the invention, when the IL polypeptide is IL-18, a soluble IL-R polypeptide is a polypeptide which comprises or consist of an amino acid sequence of SEQ ID NO: 18, or an analog, derivative, active portion or fragment thereof which retains the binding properties of the full length protein.

Deletion ranges

The deletion of the C-terminus segment of the extracellular domain of IL-R after the final cysteine residue of domain 3 would yield a protein which is likely to retain the binding capabilities of the full length protein. Furthermore, deletions from the N- terminal portion of the extracellular domain of IL-R up to the first cysteine of domain 1 would also yield a protein which is likely to retain the binding capabilities of the full length protein.

In a preferred embodiment of the invention, deletions of between 1 and 30 amino acids, preferably between 1 and 20 amino acids, more preferably between 1 and 15 amino acids, most preferably between 1 and 10 amino acids from the C-terminus segment of the extracellular domain of IL-R yield a polypeptide which is likely to retain the binding properties of the full length protein.

A most preferred sIL-R of the invention comprises or consists of the extracellular domain (of the corresponding endogenous full length IL-R) to which no further deletion has been made.

IL-RAcP polypeptides

• Full length Interleukin receptor accessory proteins In the present application, IL-RAcP polypeptide means a polypeptide comprising an intracellular domain, one single transmembrane domain and an extracellular domain. The extracellular domain comprises three Ig-like domains and allows the binding of IL-RAcP to an IL-R/IL binary complex. The signalling form of the IL-RAcP polypeptide is membrane bound.

The prediction for the location of these three Ig-like domains in the IL-lRAcP has been made by Greenfeder et al . (Greenfeder et al . , 1995) . He defined the boundaries of the Ig-like domains of the IL-lRAcP by the location of unique Cystyl ('C') residues, which can be found close to the N- and -C termini of each domain as outlined in figure 2 . Figure 24 shows the sequence alignment of the human Interleukin-1 type I receptor (IL-IRI; Genbank accession number xl6896) and the human Interleukin receptor accessory protein (IL-lRAcP; Genbank accession number af029213) . The first and last cysteine of each domain are indicated by shading and holding. Underlined portions of sequence identified by ( ), ( ) and (

) indicate the localization of domains 1, 2 and 3 respectively.

Yoon and Dinarello (1998) showed that monoclonal antibodies directed to a segment of the third domain of IL-lRAcP were able to prevent IL-1/3 signalling.

The Ig-like domains for the IL-IRI were determined according to those presented in Greenfeder et al . , 1995 for the IL-lRAcP.

IL-RAcP polypeptides may include members of the IL- lRAcP-like family polypeptides and IL-18RAcP family like polypeptides, for example, human (Genbank accession number AF029213) , mouse (Genbank accession number NM_008364) , or rat (Genbank accession number NM_012968) immature IL-lRAcP, human (Genbank accession number NM_003853) or mouse (Genbank accession number NM_010553) immature IL-18RAcP, human IL-1R9 (Genbank accession number AF212016) (Carrie et al, Nature genetics, (1999), 23, 25-31 and Sana et al . , (2000) Genomics, 69, 252- 262) , human TIGIRR-2 (Genbank accession number AF284435) , human IL1RAPL1 (Genbank accession number NM_014271) , naturally occuring soluble human IL-lRAcP (Genbank accession number AF167343) .

Except the naturally occuring soluble human IL-lRAcP (Genbank accession number AF167343) the above defined full length IL-RAcP are not usually suitable for an assay of the invention.

• Soluble IL-RAcP Polypeptides

Soluble IL-RAcP polypeptide (sIL-RAcP) means any soluble polypeptide fragment of a full-length IL-RAcP which retains the extracellular binding activity of the full length protein (i.e. the ability to form a trimolecular complex with its corresponding IL-R/IL complex which may be soluble or membrane bound) .

Preferred soluble IL-RAcP for use according to the present invention comprises or consists of the extracellular domain of the endogenous protein or a portion or fragment thereof which retains its binding capability. A suitable portion or fragment binds to the same ligands as the endogenous protein.

The extracellular domains of IL-RAcP comprise three Ig- like domains which are included in three structural domains identified herein as domain 1, domain 2 and domain 3 starting from the N-terminal end of the sequence. Domain 1 contains one Ig-like domain and two disulfide bonds. Domain 2 contains one Ig-like domain and two overlapping disulfide bonds. Domain 3 contains one Ig-like domain and one disulfide bond. See Figure 23 for the domain organization of IL-lRAcP.

• Preferred soluble IL-RAcP polypeptides For illustration purposes, the preferred soluble IL-RAcP polypeptides described herein are human sIL-lRAcP and human sIL-18RAcP but with the help of figure 18 one skilled in the art can extrapolate it to other sIL-lRAcP polypeptides .

In one embodiment of the invention the sIL- RAcP is a soluble IL-lRAcP polypeptide.

In a preferred embodiment of the invention, the soluble IL-RAcP polypeptide is a polypeptide which retains the binding properties of the full length receptor accessory protein and which has at least 60% identity, or at least 80% identity, preferably 85% identity, more preferably 90% identity, most preferably 95% identity with the amino acid sequence of SEQ ID NO: 21.

A most preferred soluble IL-RAcP polypeptide is a polypeptide which comprises or consist of an amino acid sequence of SEQ ID NO : 21 or an analog, derivative, active portion or fragment thereof which retains the binding properties of the full length protein.

In further embodiments of the invention when the IL polypeptide is IL-1, a soluble IL-RAcP polypeptide is a polypeptide which retains the binding properties of the full length receptor accessory protein and which has at least 60% identity, or at least 80% identity, preferably 85% identity, more preferably 90% identity, most preferably 95% identity with the amino acid sequence of SEQ ID NO: 22, SEQ ID NO: 23, or SEQ ID NO: 26.

In further preferred embodiments of the invention when the IL polypeptide is IL-1, a soluble IL-RAcP polypeptide is a polypeptide which comprises or consist of an amino acid sequence of SEQ ID NO: 22, SEQ ID NO: 23, or SEQ ID NO: 26 or an analog, derivative, active portion or fragment thereof which retains the binding properties of the full length protein.

In another embodiment of the invention the sIL-RAcP is a soluble IL-18RAcP polypeptide

In another preferred embodiment of the invention the soluble IL-RAcP polypeptide is a polypeptide which retains the binding properties of the full length receptor accessory protein and which has at least 60% identity, or at least 80% identity, preferably 85% identity, more preferably 90% identity, most preferably 95% identity with the amino acid sequence of SEQ ID NO: 24.

A most preferred soluble IL-RAcP polypeptide is a polypeptide which comprises or consist of an amino acid sequence of SEQ ID NO: 24 or an analog, derivative, active portion or fragment thereof which retains the binding properties of the full length protein.

In further embodiments of the invention when the IL polypeptide is IL-18, a soluble IL-RAcP polypeptide is a polypeptide which retains the binding properties of the full length receptor accessory protein and which has at least 60% identity, or at least 80% identity, preferably 85% identity, more preferably 90% identity, most preferably 95% identity with the amino acid sequence of SEQ ID NO: 25.

In further preferred embodiments of the invention when the IL polypeptide is IL-18, a soluble IL-RAcP polypeptide is a polypeptide which comprises or consist of an amino acid sequence of SEQ ID NO: 25 or an analog, derivative, active portion or fragment thereof which retains the binding properties of the full length protein.

Deletion ranges

The deletion of the C-terminus segment of the extracellular domain of IL-RAcP after the final cysteine residue of domain 3 would yield a protein which is likely to retain the binding capabilities of the full length protein. Furthermore, deletions from the N- terminal portion of the extracellular domain of IL-RAcP up to the first cysteine of domain 1 would also yield a protein which is likely to retain the binding capabilities of the full length protein. In a preferred embodiment of the invention, deletions of between 1 and 30 amino acids, preferably between 1 and 20 amino acids, more preferably between 1 and 15 amino acids, most preferably between 1 and 10 amino acids from the C-terminus segment of the extracellular domain of IL-RAcP yield a polypeptide which is likely to retain the binding properties of the full length protein.

A most preferred sIL-RAcP of the invention comprises or consists of the extracellular domain (of the corresponding endogenous full length IL-RAcP) to which no further deletion has been made.

Family members

The three members of the IL ternary complex are highly conserved among mammalian species. This high level of sequence identity among the various species has at least two practical consequences: a) The three structural portions of the extracellular domains of IL-R and IL-RAcP can be selected from various species. It is therefore possible to construct IL-R and IL-RAcP chimeras which can include one member of the IL complex from a species and at least another member of the IL complex from another species . The chimeras can also involve a combination of nucleic acid sequences such that one generates a cross-species chimera in a single protein by reference to the cross- species sequence alignment of figures 14 to 18. See Born et al., 2000 for a selection of IL-1R family chimera. b) The prediction of the locations of the structural domains and hence the Ig-like domains in any IL-R or IL-RAcP and the determination of the portion of structural domains 1 and 3 which can be deleted from any IL-R and IL-RAcP to yield a polypeptide with appropriate binding capabilities can be determined by one skilled in the art, for example from the sequence alignment comparisons of figures 17 and 18 coupled with the data of figure 24. Figure 24 shows the sequence alignment of the human Interleukin-1 type I receptor (IL-IRI; xl6896) and the human Interleukin receptor accessory protein (IL-lRAcP; af029213) . The first and last cysteine of each domain are indicated by shading and holding. Underlined portions of sequence identi ied by ( ) , ( ) and (

__) indicate the positions of domains 1, 2 and 3 respectively.

As a consequence, the deletion ranges provided above can also be applied to other mammalian IL-R and IL-RAcP sequences such as rat or mouse sequences .

Signal sequence for peptide production

A secretory signal sequence is defined as a sequence at the N-terminus of a protein sequence which permits the protein to translocate outside the cell membrane via the endoplasmic reticulum and golgi apparatus . Through which the secretory signal sequence is specifically cleaved by enzymes, ultimately yielding a secreted and processed protein. Signal sequence and cleavage sites vary with proteins . General features of signal sequences are the hydropathy of their composition (F, L, I, V, M, A, C and P residues are generally present) . The average length of a signal sequence is twenty amino-acid residues. This appears to be the length necessary for the protein to go through the membrane and be secreted. In order to produce an extracellular protein, a skilled person can fuse a mature protein and effective signal sequences to obtain a fully secreted protein. In the context of the present invention any signal sequence can be used in order to produce a mature IL polypeptide or/and an sIL-R polypeptide or/and an sIL-RAcP polypeptide for use in the invention. Preferred signal sequences are known signal sequences such as Melittin sp

(MKFLVNVALVFMWYISYIYA, SEQ ID N°38) , Human pancreatic Lipase sp (MWLLLTMASLISVLGTTHG, SEQ ID N°39) , human IL- IRI sp (MKVLLRLICFIALLISSLEA, SEQ ID N°40) or human IL- lRAcP sp (MTLLWCWSLYFYGILQSDA, SEQ ID N°41) . By inserting the mature gene of a target protein downstream from the DNA sequence coding region of the signal in a eukaryotic expression system (e.g. the insect- cell/baculovirus system) , a secreted protein may be produced.

Screening assays

The main advantage of using soluble forms of the proteins is the ease with which these reagents enable one to format and run High Throughput Screening (HTS) assays. Homogenous and soluble assays are also very amenable to miniaturisation which can aid HTS formatting. Such assays often give more reliable data due to their homogenous nature. This is very important in HTS modes as one wishes to be able to detect with confidence the small number of compounds which may be active in the assay (often around 0.3% of the total number of compounds screened in a typical HTS) .

IL-R polypeptides used in the assays of the present invention are soluble. Endogenous active functional IL/IL-R/lL-RAcP complex involves membrane bound IL-R which contains a hydrophobic transmembrane domain and a cytoplasmic domain. Such membrane-bound IL- R polypeptides are not soluble and are not usually suitable for use in the present invention.

IL-RAcP polypeptides used in assays of the present invention are soluble. Endogenous active functional IL/IL-R/lL-RAcP complex involves membrane bound IL-RAcP which contains a hydrophobic transmembrane domain. Such membrane-bound IL-RAcP polypeptides are not soluble and are not usually suitable for use in the present invention.

In one particular embodiment of the invention the assay can be carried out with soluble IL-R and IL-RAcP polypeptides still comprising a signal sequence. A preferred signal sequence is the naturally occuring signal sequence of the protein.

However most preferred sIL-R and sIL-RAcP polypeptides used in the assay of the present invention do not contain a signal sequence .

In another embodiment of the present invention, soluble polypeptides may be bound to a support for use in screening. This support could be through anchorage to a cell membrane or anchorage to a BIAcore sensor chip. In a further embodiment of the invention the IL polypeptide may contain an ATG Methionine start codon in replacement of the naturally occurring signal sequence.

The present invention concerns the use of a trimolecular IL/SIL-R/SIL-RACP complex of the invention in screening methods and assays for agents which modulate the interaction between IL and IL-R, and/or the interaction between IL-RAcP and the IL-R/IL bimolecular complex, thereby modulating the formation and/or stability of the trimolecular complex including IL, IL-R and IL-RAcP.

Methods of obtaining agents able to modulate the interaction between IL and sIL-R, or the interaction between sIL-RAcP and the sIL-R/IL bimolecular complex include methods wherein a suitable end-point is used to assess the interaction of the soluble components in the presence and absence of a test substance. Such assay systems may be used in an assay of the present invention to determine IL binding to sIL-R, sIL-RAcP binding to the sIL-R/IL bimolecular complex or disruption of an already formed trimolecular complex. Generally of most interest is modulation of the formation of the ternary complex including IL, sIL-R and sIL-RAcP.

Various types of screening assay technologies which can be used in the present invention are briefly described below.

• Scintillation Proximity Assay (SPA)

In a scintillation proximity assay, a biotinylated protein fragment may be bound to streptavidin coated scintillant - impregnated beads (e.g. as produced by Amersham) . Binding of a radiolabelled polypeptide is then measured by determination of radioactivity induced scintillation as the radioactive peptide binds to the immobilized fragment. Agents which cause a modulation (i.e. reduction or enhancement) in the measured scintillation are thus modulators (inhibitors, activators) of the interaction.

• Flashplate™ technology (NEN Life Sciences)

Flashplate™ technology (NEN Life Sciences) may also be used as an end point determination method. The surfaces of wells in a Flashplate™ microtitre plate contain scintillant and may also be coated with a material such as Ni-NTA or streptavidin which enables a molecule to be anchored to the surface of the well, for example via a 6-His tag or a biotin group. If a [³H] labelled molecule interacts with a molecule anchored to the surface, the proximity of the [³H] to the scintillant leads to the activation of the scintillant and the emission of light. This emission, which is directly related to the amount of interaction of the molecules, can be measured in a microplate scintillation counter. Inhibitors or enhancers of the interaction reduce or enhance respectively the amount of

[³H] which is in proximity with the scintillant and hence reduce or enhance respectively the signal.

Test compounds may be assayed by measuring the effect on the cpm (counts per minute) output of the test assay when compared to that under identical conditions without the test compound.

In one preferred embodiment of the present invention, biotin-sIL-lR is anchored to a streptavidin flashplate, free sIL-lRAcP, and free ¹²⁵I IL-lβ (or more preferably

³H IL-lβ) are then added.

In another preferred configuration, sIL-lRAcP-6His is anchored to a Ni-flashplate via the 6his tag and a binary complex comprising [³H] labelled sIL-lR associated with IL-1/3 is then added.

• Homogeneous Time Resolved Fluorescence (HTRF)

Homogeneous time resolved fluorescence (HTRF) is a dual label fluorescent technique which is particularly suitable for end point determination in assays of the present invention. The technique involves labelling each of the interacting polypeptides or complexes with a fluorescent label, such that a donor label on one molecule is brought into proximity with an acceptor label when the molecules interact.

A problem with many dual-label fluorescent technologies is the background fluorescence of the assay components. Signal from an energy acceptor molecule proximal to an energy donor is partially obscured by signal from other components of the assay, such as acceptor molecules distant from a donor molecule.

HTRF is based on the discovery that lanthanide cryptate molecules, such as europium cryptate fluoresce over a long time period when excited by light of a particular wavelength (Kolb et al 1996, Prat et al 1995) . When a lanthanide cryptate such as europium cryptate is used as an energy donor label in a bioassay, energy is transferred to acceptor molecules over a long time period. Short-lived signal from distant acceptors and other assay components rapidly dies away after excitation, so signal measured a short time after excitation will be almost entirely produced by acceptors which are in close proximity to a donor molecule and are still being excited by the long lived fluorescence of the donor. Thus the assay can be used to determine the proximity of two species of biomolecules labelled with the donor and acceptor molecules respectively. Suitable energy donor labels include europium cryptate, and suitable energy acceptor labels include Cyanine 5 (Cy-5) and XL665™ (obtainable from CIS-Bio international, France) .

Thus, sIL-lRAcP is for example labelled with an energy acceptor such as Cy-5 and sIL-lRI may be labelled with an energy donor such as europium cryptate. The proximity of the receptor and accessory protein can thereby be determined by measuring the delayed fluorescent emission at 665nm (for Cy-5) after excitation at 337 nm. The close proximity of the receptor and accessory protein is indicative of the presence of a trimolecular complex of IL-1, sIL-lRI and sIL-lRAcP.

• Origen^t Technology (Igen)

Origen™ technology (Igen) may also be used as an end point method in assays of the present invention. One molecule of an interacting pair is attached to a magnetic bead, for example by Ni-NTA/6-His or biotin/streptavidin linkage. A second molecule labelled with ruthenium (II) tris- (bipyridine) is added and the interaction between the molecules brings the ruthenium (II) tris- (bipyridine) label into proximity with a stimulating electrode. The electrochemical stimulus causes the label to chemio-luminesce. The light output is related to the extent of interaction between the molecules .

For example, in one suitable configuration, sIL-lRAcP- 6His is anchored to a Ni-NTA magnetic bead via the 6His tag and a binary complex comprising ruthenium (II) tris- (bipyridine) labelled sIL-lR associated with IL-1/3 is then added. Light output from the label is then measured following electrochemical stimulation.

The amount of test substance or compound which may be added to an assay of the invention will normally be determined by trial and error depending upon the type of compound used. Typically, concentrations from about 0.001 nM to 1 mM of putative inhibitor compound may be used. Prefered concentrations are for example from 0.01 nM to lOOμM, most preferably 0.1 to 50 μM, and particularly about 10 μM. Greater concentrations may be used when a peptide is the test substance.

It has to be noted that the use of BIAcore (surface plasmon resonance) or Flashplate is more physiological than the use of HTRF. Use of the C-terminal 6His tag to anchor sIL-lRAcP to either a Ni-NTA BIAcore sensor chip or flashplate surface orientates the protein in a manner similar to that it would be predicted to take when anchored to a membrane in its membrane-anchored full- length form. This makes such assays slightly more representative of the physiological situation compared to the assay of the proteins in free solution.

Primary screening assays

• Screening assay for IL modulators

¹Modulation' as used herein means enhancement, disruption or interference with the formation and/or stability of a trimolecular complex consisting of IL, sIL-R and sIL-RAcP. A modulator may activate, enhance, disrupt, reduce, interfere with or wholly or partially abolish interaction between IL and sIL-R, thereby affecting the formation of the sIL-R/IL bimolecular complex and/or activate, enhance, disrupt, reduce, interfere with or wholly or partially abolish the interaction between the sIL-R/IL bimolecular complex and sIL-RAcP, thereby affecting the formation of the trimolecular complex including IL, sIL-R and sIL-RAcP. A modulator may also disrupt an already formed trimolecular complex including IL, sIL-R and sIL-RAcP.

In one general aspect, the present invention provides an assay method for determining the ability of a test compound to modulate the formation of a trimolecular complex including IL, a soluble IL-R and a soluble IL- RAcP, the method comprising:

In a most preferred embodiment, the present invention is an assay method for determining the ability of a test compound to modulate the formation of a trimolecular complex including IL, a soluble IL-R and a soluble IL- RAcP, the method comprising: (a) providing a test compound;

(b) bringing the test compound into contact with a defined amount of a soluble IL-R polypeptide and a soluble IL-RAcP polypeptide;

(c) adding an IL polypeptide to the mixture obtained in step (b) ; and

(d) determining the amount of said trimolecular complex formed.

In another aspect, the present invention is an assay method for determining the ability of a test compound to disrupt or interfere with the formation of a trimolecular complex including IL, a soluble IL-R and a soluble IL-RAcP, the method comprising:

(a) bringing into contact an IL polypeptide, a soluble IL-R polypeptide, a soluble IL-RAcP polypeptide and a test compound; and

(b) determining the amount of said trimolecular complex formed.

It is important to note that the order in which IL, the sIL-R, the sIL-RAcP and the test compounds are contacted in step (a) can vary. In other words, any order of addition of the various components is possible as long as there is no trimolecular complex formation before the test compound is added. Furthermore, any preformation of bimolecular complexes is also acceptable as long as no trimolecular complex is preformed before the incorporation of the test compound.

(a) providing a test compound;

(c) adding an IL polypeptide to the mixture obtained in test (b) ; and

(d) determining the amount of said trimolecular complex formed.

In most preferred embodiments of the invention an IL used in the assay is a human mature IL-1 or a human mature IL-18, sIL-R used in the assay is a human sIL-lRI or a human SIL-18R, sIL-RAcP used in the assay is a human sIL-lRAcP or a human sIL-18RAcP.

In the most preferred embodiment of the invention IL is the human mature IL-lβ, sIL-R is a human sIL-lRI, sIL- RAcP is a human sIL-lRAcP. • Screening assay for IL activators using an IL antagonist

In a reaction medium comprising IL, an IL antagonist, sIL-R and sIL-RAcP, the presence of an entity which binds the IL antagonist freeing IL from its interaction with the IL antagonist leads to an IL activator like effect. In fact IL can then bind sIL-R, leading to an IL/sIL-R/sIL-RAcP ternary complex formation.

The present invention therefore also concerns an assay method for determining the ability of a test compound to enhance the formation of a trimolecular complex including IL, a soluble IL-R and a soluble IL-RAcP, the method comprising:

(a) bringing into contact an IL polypeptide, a soluble IL-R polypeptide, a soluble IL antagonist, a soluble IL-RAcP polypeptide and a test compound; (b) determining the amount of said trimolecular complex formed; and (c) comparing the amount of said trimolecular complex formed at step (a) with the amount of said trimolecular complex formed in the absence of test compound.

(a) providing a test compound;

(b) bringing the test compound into contact with a soluble IL-R polypeptide, a soluble IL antagonist and a soluble IL-RAcP polypeptide;

(c) adding an IL polypeptide to the mixture obtained at step (b) ; (d) determining the amount of said trimolecular complex formed; and (e) comparing the amount of said trimolecular complex formed at step (c) with the amount of said trimolecular complex formed in the absence of test compound.

A most preferred embodiment of the present invention is an assay method for determining the ability of a test compound to enhance the formation of a trimolecular complex including IL, a soluble IL-R and a soluble IL- RAcP, the method comprising:

(a) bringing into contact a test compound, an IL polypeptide and a soluble IL antagonist;

(b) adding a soluble IL-R polypeptide and a soluble IL-RAcP polypeptide;

(c) determining the amount of said trimolecular complex formed; and

(d) comparing the amount of said trimolecular complex formed at step (b) with the amount of said trimolecular complex formed in the absence of test compound.

In preferred embodiments of this screening assay the soluble IL antagonist is a soluble IL antagonist polypeptide.

A most preferred soluble IL antagonist polypeptide is soluble IL-1RII or IL-18BP. • IL-1RII

Suitable soluble IL-IRII polypeptides include polypeptides which retain the binding properties of the full length IL-RII and which have at least 60% identity, or at least 80% identity, preferably 85% identity, more preferably 90% identity, most preferably 95% identity with the amino acid sequence of SEQ ID NO: 42.

A most preferred soluble receptor type II polypeptide for use in those assays is the polypeptide of SEQ ID NO: 42 or an analog, derivative, active portion or fragment of said sequences which retains the binding capabilities of the full length protein.

In other embodiments of the invention another signal sequence as mentionned above may be added to the IL-RII polypeptide .

IL-18 binding protein (IL-18BP)

A suitable IL-18BP polypeptide is a polypeptide which retains the binding properties of the full length IL- 18BP and which has at least 60% identity, or at least 80% identity, preferably 85% identity, more preferably 90% identity, most preferably 95% identity with the amino acid sequence of SEQ ID NO: 43.

A most preferred soluble IL-18BP polypeptides for use in this assay is the polypeptide of SEQ ID NO: 43 or an analog, derivative, active portion or fragment of said sequences which retains the binding capabilities of the full length protein.

• Screening assay for IL activators using an IL-R antagonist

In a reaction medium comprising IL, an IL-R antagonist, IL-R and IL-RAcP, the presence of an entity which binds the IL-R antagonist freeing IL-R from its interaction with the IL-R antagonist leads to an IL activator like effect. In fact sIL-R can then bind IL, leading to an IL/sIL-R/sIL-RAcP ternary complex formation.

(a) bringing into contact an IL polypeptide, a soluble IL-R polypeptide, an IL-R antagonist, a soluble IL-RAcP polypeptide and a test compound;

(b) determining the amount of said trimolecular complex formed; and (c) comparing the amount of said trimolecular complex formed at step (a) with the amount of said trimolecular complex formed in the absence of test compound.

(a) providing a test compound;

(c) adding a soluble IL-R polypeptide to the mixture obtained in step (b) ;

In preferred embodiments of this screening assay the IL- R antagonist is an IL-R antagonist polypeptide. A most preferred IL-R antagonist polypeptide is IL-ra.

IL-lra

Suitable IL-lra polypeptides are polypeptides which retain the binding properties of the full length IL-lra and which have at least 60% identity, or at least 80% identity, preferably 85% identity, more preferably 90% identity, most preferably 95% identity with the amino acid sequence of SEQ ID NO: 44.

One of the most preferred soluble IL-1 ra polypeptides for use in this assay is the polypeptide of SEQ ID NO: 44 or an analog, derivative, active portion or fragment of said sequence which retains the binding capabilities of the full length protein. Other suitable IL-lra polypeptides include human IL1HY1 (Genbank accession number AF186094) , mouse IL1HY1 (Genbank accession number NM_019451) and mouse IL-lrn (Genbank accession number M74294) .

The assay methods set forth above are considered by the inventors as preferably being primary assays for the screening of test compounds capable of disrupting, interfering with or enhancing the formation of a ternary complex including IL, IL-R and IL-RAcP.

An agent identified using one or more of the primary screens as having ability to modulate the formation of the trimolecular complex may be further assessed using one or more of the secondary screens described below.

• Secondary Screening assays

A secondary screen assay method according to the invention involves testing for the effect of the test compound that was found to be positive in one of the primary screening assays on the bimolecular complex between IL and the soluble IL-R polypeptide. This allows the interaction which is affected by the test compound to be identified as either the IL interaction with sIL-R or the sIL-RAcP interaction with the sIL-R/IL bimolecular complex .

In particular embodiments of the invention, a secondary screening assay method comprises:

(a) bringing into contact an IL polypeptide, a soluble IL-R polypeptide, and a test compound that was found to be positive in one of the primary screens described above; and, (b) determining the formation of a bimolecular complex consisting of the IL polypeptide and the soluble IL-R polypeptide.

For example, in the case of a test compound being an inhibitor, the interaction disrupted by the test compound can then be determined by comparing the formation of bimolecular complex with that of trimolecular complex. Where only trimolecular complex formation is inhibited, the compound disrupts the sIL- RAcP interaction with the sIL-R/IL bimolecular complex. Where the formation of both complexes is inhibited, either both interactions are inhibited or the interaction between IL and sIL-R is inhibited.

A particular embodiment of the invention concerns a secondary screening assay method comprising: (a) bringing into contact an IL polypeptide, a soluble IL-R polypeptide, and a test compound; and,

(b) assaying the presence of a bimolecular complex consisting of the IL polypeptide and the soluble IL-R polypeptide in the reaction mixture of step (a) .

A further aspect of the present invention relates to a secondary screening assay method for determining the ability of a test compound to disrupt or interfere with the stability of a trimolecular complex including IL, a soluble IL-R and a soluble IL-RAcP, the method comprising: (a) providing a defined amount of a trimolecular complex comprising an IL polypeptide, a soluble IL-R polypeptide and a soluble IL-RAcP polypeptide; (b) contacting a test compound with said trimolecular complex; and

(c) comparing the amount of said trimolecular complex present after step (b) to the amount of trimolecular complex initially present at step (a) .

In the most preferred embodiment of the invention IL is the human mature IL-1, sIL-R is a human sIL-lRI, sIL- RAcP is a human sIL-lRAcP.

Secondary assays may also include testing for induction of downstream IL effectors. For example, the activity of a NFkB activation gene may be assayed after IL-1 stimulation using a commercially available NFKB driven- β-lac reporter (Aurora technology) .

Agents able to modulate the interaction between IL and sIL-R, or the interaction between sIL-RAcP and the sIL- R/IL bimolecular complex as described above may also be used to modulate the interaction between IL and the membrane bound IL-RAcP and the membrane bound IL-R/IL bimolecular complex.

The present invention is further illustrated, but in no case limited, by the figures and the examples below.

EXAMPLES

MATERIALS

Oligonucleotide primers were synthesized using an ABI 392 DNA Synthesizer or obtained directly from PE Applied Biosystems . Human brain Quick Clone cDNAs were obtained from Clontech. A synthetic gene, in plasmid pDR540, encoding the mature form of human IL-1/3 with optimal E. coli codon usage was employed. Restriction enzymes and other DNA modifying enzymes were obtained from either Boehringer Mannheim or Stratagene. Recombinant human IL- lra, IL-1/3, and soluble IL-1 receptor type II (sIL-lRII) were purchased from R&D Systems. Peptides were synthesised by Abachem Ltd. Tissue culture media and reagents were obtained from Life Technologies Ltd. A BIAcore^" 2000 was used to measure binding kinetics and affinity constants for the interactions between proteins. CM5 and NTA-chips, N-hydroxysuccinimide, N- ethyl-N' - (3-diethylaminopropyl) carbodiimide, and ethanolamine coupling reagents (BIAcore) were used to immobilize proteins to the sensor surface using a standard amine-coupling procedure (Jonsson et al . 1991). EXAMPLE 1: Identifica ion of IL-1R and IL-lRAcP extracellular domains

The extracellular domain of a membrane-bound receptor or accessory protein has been identified by a hydropathy analysis of the mature sequence. Trans-membrane regions are indicated in a hydropathy plot as regions with an average of twenty continuous amino acids residues with a high hydropathy index.

The hydropathy index provides indication of the transmembrane region. Following this analysis, the precise boundaries of the region are then mapped using standard techniques known to a skilled person. Figures 21 and 22 show a hydropathy plot of the full- length sequence of human IL-IRI and IL-18R (Figure 21) and human IL-lRAcP and IL-18RAcP (Figure 22) . The TMpred program makes a prediction of membrane- spanning regions and their orientation. The algorithm is based on the statistical analysis of TMbase, a database of naturally occuring transmembrane proteins. The prediction is made using a combination of several weight-matrices for scoring (K. Hofmann & W. Stoffel, 1993) .

The vertical dashed lines delimit the region that is predicted to be the single transmembrane domain of the protein (as can be seen from the strongly positive hydropathy index in this region of the protein sequence) . The residue immediately prior to the start of the transmembrane domain of the human IL-1R and human IL-lRAcP was thus selected to be the final residue encoding the soluble form of these proteins (and indeed these C-terminally deleted proteins were secreted upon expression in the insect-cell/baculovirus system) . These C-terminally deleted proteins could also be expressed in bacteria or yeast . Figures 17 and 18 show a sequence alignment of the amino acid sequence of respectively IL-1R type I and IL-1R type II, IL-lRAcP and IL-18RAcP from human, rat and mouse .

In Figures 17 and 18, the predicted transmembrane domains of the Human Interleukin I receptor type I and II, Human Interleukin I receptor accessory protein, Human Interleukin 18 receptor and Human Interleukin 18 receptor accessory protein are outlined in a transparent box which contains the appropriate bolded residues in accordance with the transmembrane domain determination obtained by plotting the hydropathy.

Figures 17 and 18 show a conserved location in the primary sequence of the transmembrane domain (boxed) and thus it is very likely that similar C-terminally deleted forms of other full length IL-R and full length IL-RAcP, including other members of the IL-1 family like receptor polypeptide and IL-1 family like receptor accessory protein polypeptide are also secreted upon expression in the insect-cell/baculovirus system.

EXAMPLE 2 : Cloning, Expression, and Purification of Recombinant Proteins

1) Methods 1 . 1 ) IL- 1/3

The synthetic gene (SEQ ID N° 2) encoding IL-1/3 (SEQ ID N° 3) was modified by PCR using Pfu DNA polymerase to introduce a Nco I restriction site at the 5 ' start codon (Primers: 5 ' -GTCCCATGGCACCGGTTAGATCTCTG-3 ' , with the Nco I site shown in bold (SEQ ID No: 27) and 5'- CAGCTTATCGGCGTAGAGGAT-3' , which corresponds to a region of pDR540 distal to the IL-1/3 gene (SEQ ID No: 28) . The product was digested with Nco I and Bam HI (a site present immediately after 2 tandem stop codons in the gene) and subcloned into pQE-60 (Qiagen) . The resulting construct (pQE-hrIL-1/3) , which was sequenced for confirmation, is designed for expression of the mature form of the IL-1/3 protein (sequence of SEQ ID N° 4) with the addition of an N-terminal methionine residue. pQE~ hrIL-1/3 was transformed into M15 E. coli containing the plasmid pREP4, which encodes the lac repressor for tight regulation of IL-1/3 expression (Qiagen QIAexpress system) .

Transformed M15 E. coli were cultured using standard methods and then induced with Isopropyl-1-thio-beta-D- galactopyranoside (Sigma) at a final concentration of 2mM. After 4h of growth, cells were harvested by centrifugation, 4000g (Beckman 21, JA14 rotor, 5500rpm) for 30 minutes at 4°C. The supernatant was discarded and pellets were stored at -70°C until required. One litre of culture produced approximately 4g of cells (wet weight) . The cell pellet was resuspended in 50ml of 50mM Tris-HCl, pH 8.0, 5mM EDTA, 0. ImM PMSF and 250μg lysozyme/ml (buffer A) . The suspension was mixed gently and incubated at 25°C for 20 minutes. Five ml of Buffer B (1.5M NaCl, 0.1M CaCl₂, 0.1M MgCl₂, and ImM PMSF) containing 25μl of DNAse I (Stratagene) were added. The mixture was incubated until the cell lysate showed low viscosity.

The mixture was centrifuged at 17,000 g for 1 hour at 4°C and the supernatant recovered and diluted with 400ml of lOmM sodium acetate, pH 5.1 (buffer C) . After centrifugation at 17,000 g for 30 minutes at 4°C (to remove the precipitate) the supernatant was loaded onto an SP-Sepharose Fast-Flow (Pharmacia Biotech Inc.) column (5cm i.d. 2.5cm) equilibrated with buffer C at a flow rate of 250ml/h. Elution was performed with 220mM sodium acetate, pH 5.1. The eluate was concentrated to 3ml and loaded at 30ml/h onto a G75 Sephadex column (2.5x90cm) equilibrated with buffer (PBS, pH 7.4 or lOOmM NaHC03, pH 8.3). Fractions containing IL-1/3 were pooled and concentrated to 5ml . The last peak fraction from the G75-Sephadex yielded a single band on a 4-20 % SDS-PAGE gel under reducing conditions at the expected M_r of hIL-1/3, which is ~17,400Da.

An IL-1/3 affinity column was generated by cross-linking human recombinant IL-1/3 (20mg) purified as above to lg of CNBr-activated Sepharose-4B matrix (Pharmacia) as described by the manufacturer.

1.2) sIL-lRI

A full-length cDNA (SEQ ID N°ll) encoding the human IL- 1RI (SEQ ID N°12) (Hammond et al . 1999) was modified by PCR using Pfu DNA polymerase to introduce an EcoRI site prior to the start codon, and a stop codon in place of the codon for His³³⁶ (the first residue in the predicted transmembrane region) followed by an .Xho I site.

Primer sequences : 5 ' -

TGCGAATTCATGAAAGTGTTACTCAGACTTATTTG-3' (Eco RI site in bold) (SEQ ID NO: 29) and 5'- TGACTCGAGTTACTTCTGGAAATTAGTGACTGG-3 ' (Xho I site in bold) (SEQ ID NO: 30) . The PCR product was cloned into pBluescript II SK (+) and sequenced on both strands. The Eco RI to Xho I fragment was then subcloned into pFastBacl and used to generate recombinant baculovirus bacmid DNA by site specific transposition in DHlOBac (Bac-to-Bac System, Life Technologies) . The resulting bacmid DNA from several clones was purified and used to transfect sf9 cells using Cellfectin.

For optimal expression of the Human Interleukin I receptor protein in insect cells, PCR was utilized as described herein to insert the following Kozak translation initiation consensus sequence immediately prior to the ATG start codon: "GCCACC".

Variations on this consensus sequence may also effect efficient expression, indeed other sequences may be even more effective at directing high-level translation.

Cells were routinely maintained between 1 and 5xl0⁶ cells/ml in 1 litre polycarbonate shake flasks (Corning) containing 400ml of SF-900 II SFM serum-free insect cell media. Culturing conditions were 27°C with rotation at llOrpm in a shaking incubator. For protein production, cells were grown to a density of 5xl0⁶ cells/ml then diluted to 2xl0⁶ cells/ml immediately prior to infection with a 1:100 dilution of the appropriate baculovirus P3 stock. Cells infected with the sIL-lRI expressing baculovirus were cultured for a further four days and then filtered serially through 100, 70 and 50mm nylon sieves prior to dilution in three volumes of 0.2M glucose. A cocktail of protease inhibitors containing lmM PMSF, lmg/ml Leupeptin, lmg/ml Aprotinin, lOmg/ml phosphoramidon, lmg/ml E64, and lmM EDTA was added into the mixture .

The crude broth was diluted 1/3 with lOmM MES pH 6.0 and applied at 25ml/min to an SP-Sepharose XL Streamline-50

(Amersham Pharmacia Biotech.) expanded-bed chromatography column (i . d. 2.5cm x h. 100cm, 20cm packed bed) pre-equilibrated in lOmM MES pH 6.0. The column was washed before elution with 350mM NaCl/lOmM HEPES pH 7.5. The eluted material containing the soluble

IL-IRI was applied to an IL-l/3-Sepharose column (2.5mg

IL-l/3/ml of gel) , equilibrated in PBS at a flow rate of

5ml/h. The column was then washed extensively with PBS and eluted with lOOmM Glycine-HCl, 150mM NaCl, pH 3.0. Fractions (2ml) were neutralised immediately with 250μl of 1M Tris-HCl, pH 8.0.

Aliquots of each step were analysed by SDS-PAGE and Western blotting using an IL-RI (N-20) polyclonal antibody (Santa Cruz Biotechnology Inc.). Glycine-HCl pH 3.0 elution fractions yielded a single band on a 4-20 % SDS-PAGE gel under reducing conditions at -44,000 Da due to the glycosylation of the receptor. The expected M_r of non-glycosylated human sIL-lRI is -36,381 Da.

Protein concentrations were measured by U.V. spectrophotometry using a theoretical molar extinction coefficient of 50,810 M^"1.cm^"1 confirmed by the Bradford assay using bovine serum albumin as standard.

The sIL-lRI produced has the sequence of SEQ ID N°13

1.3) sIL-lRAcP

Many tags may be introduced either to the C- terminus or N-terminus of the protein using standard molecular biological techniques. Examples of purification tags that may be used include, c-myc, FLAG, Hemagluttinin A, V5 ,E-tag, BirA-tag and 6His.

For the human soluble IL-lRAcP (generated from GenBank AF029213, nucleic sequence of SEQ ID N°20, amino acid sequence of SEQ ID N° 19) , a 6His tag was utilized. Thus the following sequence was added after the final GAA (glutamic acid) triplet codon sequence:

CAT CAC CAT CAC CAT CAC (SEQ ID NO: 36)

The final CAC was followed by a 'TGA' STOP codon.

The following oligonucleotide primers were used in PCR reactions to amplify human IL-lRAcP sequences. The primers were based on the published IL-lRAcP cDNA sequence (Genbank accession number AF029213, (Huang et al. 1997)), 5 ' -GGATGACACTTCTGTGGTGTG-3 ' (SEQ ID NO: 31) and 5 ' -TCCTTTTCATTATTCCTTTCATACA-3 ' (SEQ ID NO: 32). PCR products were amplified from human cortex cDNA (Clontech) using Taq polymerase . The resulting products were cloned into pCR-Script (Stratagene) and a number of clones characterized by DNA sequencing. Three mismatches were found in the full-length clone, two of which lead to changes in the encoded amino acids . These were corrected by site directed mutagenesis.

Using the corrected clone as the template, the extracellular portion of the IL-lRAcP (residues 1-359) was generated by PCR using Pfu DNA polymerase and the primers 5' -TCGCCACCATGGACACTTCTGTGGTGTG-3 * (5' primer)

(SEQ ID NO: 33) and 5'- TCGGAATTCCTCAGTGATGGTGATGGTGATGTTCCACTGTGTATCTTGGAGC-3 '

(3' primer) (SEQ ID NO: 34) . The 5' primer includes a

Kozak consensus sequence flanking the start codon while the 3' primer encodes six Histidine residues, a stop codon and an EcoRI site following the final extracellular residue (Glu³⁵⁹) of the IL-lRAcP. The PCR product was 5' phosphorylated with polynucleotide kinase, purified then blunt-end cloned into the Stul site of the baculovirus transfer vector pFastBac 1 (Life Technologies) . A clone was isolated with the gene ligated in the positive orientation with respect to the polyhedrin promoter and confirmed by sequencing. The resulting bacmid DNA from several clones were purified and used to transfect sf9 cells using Cellfectin as described by the manufacturer (Life Technologies) .

For optimal expression of the Human Interleukin I receptor accessory protein in insect cells, PCR was utilized as described herein to insert the following Kozak translation initiation consensus sequence immediately prior to the ATG start codon: "GCCACC".

The soluble IL-lRAcP (sIL-lRAcP) was produced in insect cells essentially as described above for sIL-lR except that the cells were cultured for five days after infection. Prior to chromatography on the streamline column the crude broth was diluted two-fold with glucose and elution was performed with 0.5M KCl/lOmM Imidazole/20mM Tris-HCl pH 8.0 at lOml/min. This material was loaded at 0.5ml/min onto a Ni²⁺ charged NTA

(Qiagen) column (1cm i.d. x 4cm h.) . The column was then washed sequentially at 4°C with 0.5M KCl/20mM Tris-HCl pH 8.0, 20mM Imidazole, then IM KCl/20mM Tris-HCl pH 8.0, and finally 0.5M KCl/20mM Tris-HCl pH 8.0, 20mM Imidazole until a stable A₂₈o_nm baseline was reached. The 6-His tagged sIL-lRAcP was eluted with 0.5M Imidazole/lOOmM KCl/20mM Tris-HCl pH 8.0 and dialysed extensively against PBS.

The protein concentration was determined using a theoretical molar extinction coefficient of 59,630 M^" ¹. cm^"1 and confirmed by Bradford protein assay.

2) Results:

SDS-PAGE gels (4-20%) of the samples coupled with colloidal blue staining and Western blotting and hybridization with the appropriate commercially available antibodies were performed to check the purity and identity of the proteins respectively from the purification schemes. Molecular weights were verified by Western blotting. The biological activity of the purified IL-1/3 was checked by microphysiometry as described (Hammond et al . , 1999) and was similar to material from commercial sources . The yield of recombinant IL-1/3 was ~5mg/litre of culture with a purity of >95% as determined by SDS-PAGE. Mass spectrometry of IL-1/3 revealed partial processing of the initiating methionine.

Aliquots from each step of the sIL-lRI purification were analysed by SDS-PAGE and Western blotting using an IL-RI (N-20) polyclonal antibody. The ~44,000Da band confirms the presence of the soluble glycosylated IL-IRI. The predicted molecular weight of the receptor lacking the signal sequence is 36,381 Da.

Using Penta-His monoclonal antibodies for detection (Qiagen) sIL-lRAcP gave a molecular weight of ~45,000Da indicating a significant glycosylation of the protein. The predicted molecular weight of sIL-lRAcP with the 6- His tag is 39,935Da. A yield of 1.5mg/L culture supernatant was generally obtained for both proteins.

The sIL-lRAcP produced has the sequence of SEQ ID N°21.

EXAMPLE 3 : Surface Plasmon Resonance

1) Methods:

1.1) Immobilization of proteins

The BIAcore running buffer was lOmM HEPES, pH 7.4, 150mM NaCl, lmM EDTA, and 0.005% P20 surfactant (Polyoxyethylenesorbitan) (HBS, BIAcore) . Equal volumes of 0.1M N-hydroxysuccinimide and 0. IM N-ethyl-N' - (3- diethylaminopropyl) carbodiimide were mixed, and 35μl were injected over the surface of the sensor chip to activate the carboxymethylated dextran at 5μl/min. hlL- 1/3 50μg/ml in lOmM acetate, pH 5, sIL-lRI lOμg/ml in lOmM acetate, pH 5.5, and sIL-lRAcP lOμg/ml in lOmM acetate, pH 4.5, were coupled respectively in flow cells 2, 3, and 4. Each coupling was followed by 35μl of ethanolamine to block remaining active carboxyl groups . The immobilization procedure was carried out at 25 °C and at a constant flow rate of 5μl/min. Flow cell 1, immobilization control, has been coupled with BSA lOμg/ml in lOmM acetate, pH 5.0.

1.2) Kinetic Assays on the BIAcore

All experiments were carried out at 25°C with a constant flow rate of 20 μl/min in HBS buffer. 40μl of the analyte were injected for 2 min (association phase), followed by HBS for 5-min (dissociation phase) . Equal volumes of each protein dilution were also injected over a mock blocked surface to serve as a blank sensorgram for subtraction of bulk refractive index background and non-specific binding of the analyte.

All kinetic assays were followed by an injection of 20 μl of lOO M HC1 to dissociate any remaining bound ligand. Experiments were carried out with a low level of immobilization, high flow rate (20μl/min) , and appropriate concentrations of analyte to limit mass transport effects. Furthermore, mass transport limitations were checked using the BIAsimulation program. Curves derived from these assays were used to generate kinetic constants. Five concentrations of ligand (in duplicate) were injected over the four flow cells in a random sequence.

All protein surfaces displayed a high capacity for ligands relative to surface RU (resonance units) density as well as long stability. Typically, an average of more than 85% binding capacity was preserved after 100 cycles of ligand binding and regeneration with HC1.

The amount of IL-1/3 on flow cell 2 was 763 RU. The amount of sIL-lRI on flow cell 3 was 2413 RU, and the amount of sIL-lRAcP on flow cell 4 was 1092 RU.

1.3) Data Analysis

Sensorgrams were analyzed by non-linear least-squares curve fitting using the BIAevaluation program 3.0 (BIAcore) . Kinetic constants were generated from the association and dissociation curves from the BIAcore experiments by fitting to a single-site binding model (A + B = AB) . This model gave a single exponential fit with a χ²<0.5. Comparison fitting with more complicated models did not give a better interpretation of the data. The equation (1)

& = Ro.e-^{fe (Wo)} (Eq. 1)

was used for the dissociation phase, where Rt was the amount of ligand remaining bound in RU at time t and t₀ was the beginning of dissociation phase. The final dissociation rate constant, k_off, was calculated from the mean of the values obtained from a series of injections. To analyze the association phase, the equation (2)

R_t = R e_q. (l -e -^ks <^t-^t0> ) (Eq. 2)

was employed where R_eq was the amount of bound ligand (in RU) at equilibrium, t₀ was the starting time of injection, and k_s = k_on.C +k₀__f, where C was the concentration of analyte injected over the sensor chip surface. The association rate constant, k_on, was determined from the slope of a plot of k_s versus C. The apparent equilibrium dissociation constant K_D was determined from the ratio of these two kinetic constants

Data were first subtracted from controls and zeroed using BIAevaluation 3.0 software (BIAcore) before global fitting to a bimolecular reaction model . Rate equations were generated and then numerically integrated for the entire data set simultaneously. Association (k_orι) and dissociation (k_Qf_f) rate constants were obtained for the entire data set along with residual S.D. (standard deviation) values, replication S.D. values, and correlation coefficients. Fits were not improved by using a mass transport model .

2) Results:

2.1) Interactions between the members of the IL trimolecular complex Interaction of IL-1/3 and sIL-lRI on the sensor chip

Channels 2, 3, and 4 coupled with IL-1/3, sIL-lRI, and sIL-lRAcP are represented respectively with ( ) , ( ) , and ( __ _ ) lines. The concentrations of proteins used for free ligands are 50μg/ml, lOμg/ml, and lOμg/ml respectively for IL-1/3, sIL-lRI, and sIL-lRAcP. Figure

1A, IB, and 1C represent interactions of IL-1/3, sIL-lRI, and sIL-lRAcP respectively.

Figure 1A-C indicates the ability of each ligand to bind specifically to its cognate receptor on the chip. Thus, hIL-13 recognises sIL-lRI immobilized on flow cell 3, while free sIL-lRI recognises IL-1/3 on flow cell 2.

Neither protein interacts with immobilized sIL-lRAcP.

Moreover, free sIL-lRAcP does not recognize immobilized IL-1/3 or sIL-lRI (Flow cells 2 and 3, respectively), though it can interact to some extent with itself (flow cell 4) .

Formation of an IL-l/3/sIL-lRI/sIL-RAcP ternary complex Figures 2A and 2B show the interaction of previously formed IL-l3/sIL-lRI complex with an excess of IL-lβ (A) or sIL-lRI (B) being added on separate flow cells of the BIAcore chip which have been respectively coupled with sIL-RI, IL-lβ or IL-lRAcP. The two proteins were incubated at room temperature for 15 minutes in HBS buffer (lOmM HEPES, 150mM NaCl, pH 7.4) at different ratios. Figures 2A and 2B represent the complex interaction with the sensor chip when an excess (approximately 2:1) of IL or IL-R was applied respectively ( ) , ( ) , and ( ) lines represent channels 2, 3, and 4 coupled with IL-1/3, sIL-lRI, and sIL-lRAcP respectively.

As shown in figure 2, the pre-formed IL-l /sIL-lRI binary complex interacts with immobilized sIL-lRAcP to generate a ternary complex.

In order to demonstrate the sequential mechanism of complex formation, IL-1/3, sIL-lRI, and sIL-lRAcP were serially injected. With both methods, ternary complexes were obtained on the IL-1/3 and sIL-IRI chip surfaces, thus mimicking the predicted interaction of these molecules at the cell surface: i.e. first IL-lβ binds to sIL-lRI and sIL-RAcP binds to the IL-lβ/lL-RI complex.

Figure 3 shows the formation of the ternary complex using two different methods . Figure 3A shows the sequential injection of IL-lβ and sIL-lRAcP on separate flow cells of the BIAcore chip which have been respectively coupled with sIL-lRI, sIL-lβ or sIL-lRAcP- 6His. IL-1/3 and sIL-lRI (not shown) were bound to their respective specific molecule on the chip to form a binary complex and then sIL-lRAcP was injected before the dissociation phase to form the ternary complex (coinject command on the BIAcore) . Concentrations of IL- 1/3, sIL-lRI, and sIL-lRAcP used were lOμg/ml. The analysis of the binding of sIL-lRAcP on the binary complex gave an affinity constant (K_D) of 1.37nM and 4.28nM respectively for IL-1/3 on sIL-lRIchip (figure 3A) and sIL-lRI on IL-lβ chip (figure not shown) . Figure 3B shows the capture of sIL-lRAcP_6His on a NiNTA chip followed by the addition of binary complex IL-lβ/sIL-lRI ( ) or IL-lβ alone ( ) .

Figure 3B examined whether immobilization via the C- terminal 6-His tag prevents the formation of the ternary complex. When sIL-lRAcP was captured on a Ni-NTA (Ni²⁺- Nitriloacetic acid) sensor chip and pre-formed IL- 1/3/sIL-lRI binary complex was injected, the formation of a ternary complex was observed, while no signal was observed if IL-1/3 or sIL-lRI (not shown) were injected alone.

Interactions with IL-lra

The IL-lra inhibits the biological effects of IL-lα and IL-1/3 by competing with these agents for cell surface receptors. IL-1/3, IL-lα, and IL-lra all bind with comparable affinity to the IL-IRI (Dinarello, 1996) . The IL-lRAcP forms a ternary complex with IL-IRI and either IL-lα or IL-1/3, but not with IL-lra (Greenfeder et al . 1995) . This was confirmed in figure 4 where the inability of the IL-lra/sIL-lRI to form a ternary complex with the sIL-lRAcP is shown.

Figure 4 shows a comparison of the IL-1/3 and IL-lra interactions with separate flow cells . of the BIAcore chip which have been respectively coupled with sIL-lβ, sIL-lRAcP or sIL-lRI when sIL-RAcP is added during the dissociation process.

Concentrations of cytokine were 0.5μg/ml and lμg/ml for IL-1/3 and IL-lra respectively. Concentration of sIL- lRAcP was lOμg/ml. ( ), ( ), and ( ) lines represent channels 2, 3, and 4 coupled respectively with IL-1/3, sIL-lRI, and sIL-lRAcP.

In the same way, the addition of sIL-lRAcP during the dissociation phase of the IL-lra/sIL-lRI complex has no effect on the rate of dissociation. However, the dissociation rate (k_off) is much slower than that of the IL-l/3/sIL-lRI complex (see Table 2) .

Interaction of the peptide of SEQ ID NO: 35 with sIL-lRI

The peptide FEWTPGYWQPYALPL (AF11377) SEQ ID No: 35, which was identified by phage display, is able to form a tight complex with the sIL-lRI (Yanofsky et al . 1996).

Figures 5A and 5B show the binding of FEWTPGYWQPYALPL peptide on a BIAcore sensor chip. Figure 5A represents the interaction of the peptide with the different channels. Figure 5B shows the interaction of a pre-mixed complex between the peptide and sIL-lRI in excess. (

), ( ), and ( ) lines represent channels 2, 3, and

4 coupled with IL-13, sIL-lRI, and sIL-lRAcP respectively.

Figure 5 A confirms the interaction between the sIL-lRI and the peptide of SEQ ID N° 35. A pre-formed Peptide/sIL-lRI complex was unable to interact with immobilized IL-1/3 (Figure 5B) nor was it able to form a ternary complex with sIL-lRAcP.

Interaction with the soluble IL-IR Type II

Figure 6 shows the involvement of the soluble form of the decoy IL-IR type II in the IL-IR complex. Figure 6A shows a comparison of binding of IL-lβ on a sIL-lRII flow cell ( ) , on a sIL-lRI flow cell ( ) and on a sIL-lRAcP flow cell ( ) , with a coinjection of sIL- lRAcP following cytokine injection. Figure 6B uses the same chip as the previous experiment and shows the interaction of sIL-lRII with coupled IL-lβ, sIL-lRI and sIL-lRAcP. This shows the ability of sIL-lRII to form a binary complex with IL-lβ, but this binary complex with soluble receptor was unable to form a ternary complex with sIL-lRAcP. It has to be noted that although the k_on of sIL-lRII for IL-lβ is slower than for sIL-lRI, the k_0ff of sIL-lRII for IL-lβ is much slower than that for sIL-lRI. In physiological terms, this decoy receptor type II is slower than IL-IRI to capture circulating IL- lβ but once type II binary complex is made, this complex is very slow to dissociate and there is no need for an accessory protein to stabilize it.

2.2) Kinetic studies

Figure 7 shows binding kinetics of the IL-l/3/sIL-lRI binary complex (figure 7A) and IL-l/3/sIL-lRl/sIL-lRAcP (figure 7B) . In figure 7A, free sIL-lRI was run on the IL-1/3 chip using five different concentrations (119, 59.5, 29.8, 14.9, and 7.4nM, upper to lower curve respectively) . In figure 7B, binding kinetics of sIL- IRAcP on the IL-l/3/sIL-IRI binary complex is shown. Figure 7B represents a range of sIL-lRAcP concentrations on the receptor chip following an IL-1/3 capture step with free IL-1/3 (50μg/ml) . The range of sIL-lRAcP concentrations was 119nM, 29.8nM, 14.9nM, and 7.4nM from upper to lower curves respectively. The sensorgrams shown (corresponding to figures 7A and 7B) were subtracted from controls. See table 2 for results.

IL-l/3/sIL-lRI binary complex Figure 7A presents kinetic data for the interaction of the sIL-lRI (concentration range 7.4-119 nM) with the IL-1/3 chip. The K_D observed was 1.53 nM. A similar value was observed in the reverse experiment with free IL-1/3 on the receptor chip (concentration range 0.12-50 nM) .

IL-l/3/sIL-lRI/sIL-lRAcP ternary complex Two protocols were used to determine the kinetic constants of the sIL-lRAcP on the IL-l/3/sIL-lRI binary complex. First, the pre-formed binary complex (2:1 molar excess of IL-1/3) was tested over a range of concentrations on the sIL-lRAcP chip. Second, IL-1/3 was captured on the receptor chip prior to the application of the sIL-lRAcP (Figure 7B) . The receptor chip was preferred over the IL-1/3 chip in this kinetic analysis since the formation of the ternary complex was more pronounced when the receptor was immobilized. Kinetics of molecules involved in IL-1 signalling

Data obtained from the SPR experiments using the BIAcore sensor chip are shown in table 2. The two first rows (experiments 1 and 2) indicate rate constants (K_on and Ko_ff) and equilibrium binding constants (K_D) for the simple kinetics between IL or sIL-R as shown in figure 7A. Rows 3 & 4 indicate the same kinetics in the presence of a constant concentration of sIL-lRAcP (10 μg/ml ~250nM) in each sample of free IL-1/3 or sIL-R. Rows 5 , and 6 represent binding constants of kinetics of a pre-formed binary complex (protein in excess in bold) on the sIL-lRAcP chip. Ranges of concentrations of the binary complex were 7.4nM-119nM and 2.9-45.8nM for experiments 5 and 6 respectively. Experiment 7 (figure 7B) represents a two-step kinetic with a first capture of a constant concentration of IL-1/3 (5μg/ml) on the sIL-R chip followed by the injection of a range of concentrations of sIL-lRAcP (7.4nM-119nM) . Binding analyses of molecules interacting with the IL-l/3/sIL- lRI/sIL-lRAcP complex: Row 8 represents binding data for the Interleukin-1 receptor antagonist on the sIL-R chip. Concentration range applied was 1.82-29.2nM. Row 9 represents binding data for the soluble IL-1 receptor type II on the IL-1/3 chip. Concentration range applied was 4.2-67.8nM. Row 10 represents binding data for AF11733 peptide (FEWTPGYWQPYALPL-OH, Seq NO: 35) on the sIL-R chip. Concentration range applied was 240-7760nM. All constants were calculated as described herein. The standard deviation was less than 0.1 and the χ² parameter is lower than 0.1 except for the experiment 10

(row 10 of table 2) . TABLE 2

Table 2 shows the results of binding analysis of the IL- 1/3/sIL-lRI/sIL-lRAcP interaction using surface plasmon resonance and with molecules involved in this interaction. • IL-lra

IL-lra interacts with the sIL-lRI but does not form a ternary complex with the sIL-lRAcP (Fig.4) . IL-lra over the concentration range 1.82-29.2nM was applied to the sIL-R chip. Kinetic differences were apparent between IL-lra and IL-1/3. The major difference is due to the dissociation process which is significantly slower in the case of IL-lra compared to IL-1/3 (K_off of antagonist 1.9xl0^~4 s-¹ compared to agonist 2.5xl0^"3 s^"1, see table 2, compare rows 1 and 8) . However, the affinity of IL-1/3 for the sIL-R in the presence of sIL-lRAcP is similar to that of the IL-lra/sIL-lRI binary complex (table 2) .

• sIL-lRII

IL-1RII interacts with IL-1/3 but no signalling results from the formation of this complex in cells (Liu et al . 1996) , due essentially to the lack of a cytoplasmic domain on IL-1RII. A range of soluble IL-1RII concentrations (4.2-67.8nM) was applied to the IL-1/3 chip. This enables the comparison of the interaction of IL-1/3 with these two receptor types and its implications for the signal transduction mechanism. The interaction of IL-1/3 with sIL-lRII is about 10-fold weaker (K_D = 13.5 nM) than with the sIL-lRI. This is consistent with the fact that the type II receptor acts as a decoy receptor competing weakly with type I for agonist binding.

• Peptide of sequence SEQ ID NO: 35

The peptide FEWTPGYWQPYALPL (Genbank accession number: AF11733) , which was obtained by phage display against immobilized sIL-lRI (Yanofsky et al . 1996), was applied to the receptor chip (240-7760nM) . As observed in Figure 5, this peptide inhibits binary complex formation with IL-1/3 (see Table .2) . Binding constants for the peptide/sIL-lRI and IL-l/3/sIL-lRI complexes were similar. The lower K_D for the IL-lra/sIL-lRI complex is due primarily to a slower off rate (table 2) .

In the present application, the ability of recombinant extracellular domains involved in the IL-1/3/IL-1RI/IL- lRAcP complex to interact and form stable binary and ternary complexes in solution, or on a sensor chip

(Figures 1-3) has been demonstrated. As reported previously, for full length proteins (Greenfeder et al . (1995), Yoon and Dinarello, (1998)), there was no interaction between IL-1/3 and sIL-lRAcP, and no direct interaction between sIL-lRI and sIL-lRAcP. A weak interaction was shown between free sIL-lRAcP and immobilized sIL-lRAcP (Figure 1C) .

The ternary complex could be generated after sequential injection of IL-1/3 and sIL-lRAcP over the sIL-lRI chip or after injection of sIL-lRI and sIL-lRAcP over the IL- 1/3 chip. In the former case, the formation of the ternary complex was more pronounced (figure 3B) , although the chip surfaces had the same level of immobilization in molar terms. This probably reflects coupling of IL-1/3 in an unfavourable orientation when IL-lβ is anchored on the chip. A ternary complex was also obtained after addition of a pre-made IL-l/3/sIL-lRI binary complex to the sIL-lRAcP chip (Figure 2) . The K_D for the IL-1/3/ IL-IRI binary complex in cells lacking accessory protein was 2 nM (Greenfeder et al . 1995; Laye et al . 1998). A significant stabilization of the complex was observed in cells co-expressing the IL- IRI and sIL-lRAcP, with a five-fold increase in affinity (K_ ~ 0.4 nM) (Greenfeder et al . 1995). Kinetic data from our SPR experiments with sIL-lRI and sIL-lRAcP yielded a K_D of 1.53nM and 3.61nM for the free sIL-lRI on anchored-IL-1/3 and free IL-1/3 on anchored-sIL-lRI respectively. When the analytes were loaded in the presence of sIL-lRAcP the apparent K_D values were 1.1 nM and 2.32 nM, thus confirming the stabilizing effect of the accessory protein on the binary complex.

The interaction of pre-formed IL-l/3/sIL-lRI complex with the sIL-lRAcP sensor chip gave a significantly higher affinity (K_D = 0.85 nM) . To further study the formation of the ternary complex, we used a two-step protocol. IL- 1/3 was first captured on the receptor chip and varying concentrations of sIL-lRAcP were immediately applied (see Table 2, row 7). The K_D obtained, 0.77nM, correlates well with values obtained in different cell lines expressing the IL-IRI and IL-lRAcP by Greenfeder et al. 1995.

The IL-lra interacts strongly with IL-IRI but is unable to induce signal transduction and acts as a competitive antagonist of IL-1. Crystallographic studies have demonstrated that IL-lra interacts with the first two domains of the receptor in the same manner as IL-1/3 and IL-lα (Schreuder et al . , 1997). However, unlike the agonists, it is unable to interact with the third domain, a crucial step for the formation of the ternary complex with accessory protein. As expected (Figure 4) , IL-lra interacts strongly (K_D = 0.75nM) with the soluble receptor type I, however a ternary complex with IL-lRAcP cannot be formed.

The IL-IRII is known as a decoy receptor, it does not contain a significant intracellular domain, crucial for signal transduction. Our data provides evidence that there is no interaction between the IL-l/3/sIL-lRII binary complex and sIL-lRAcP (figure 6) . Using conditions identical to those adopted previously for sIL-lRI, which gave a ternary complex with the sIL-lRAcP chip, no signal was observed when a pre-formed sIL-lRIl/lL-1/3 complex was injected. To confirm this result, sIL-lRII was passed over the IL-1/3 chip and IL-lβ on sIL-lRII chip, both gave a binary complex, and injection of sIL- lRAcP gave no further signal. Lang et al . (1998) showed that chimeric IL-lRIIextra-membranar/IL-lRIcytoplasmic receptor was able to form a ternary complex with IL- lRAcP and induces signal after IL-lβ stimulation. Malinowsky et al . (1998) suggested that the presence of IL-lRAcP was required for binding of an agonist to the IL-1RII. In contrast, the present application provides evidence that soluble IL-1RII binds IL-1/3, but does not form a ternary complex with sIL-lRAcP. A recent publication (Neumann et al . , (2000)) demonstrates how a cell-surface IL-lRIl/lL-lβ binary complex is not able to form a ternary complex with the IL-lRAcP thus resulting in no intracellular signalling. This finding supports the hypothesis that the role of the IL-IRII is as an IL- 1 'scavenger' - competitor of the IL-IR - leading ultimately to the prevention of IL-lβ to form a ternary complex (and hence signal) with the IL-IRI and the IL- lRAcP.

Our results confirm that sIL-lRII has the potential to modulate IL-1 signalling by recruiting agonists to form a non-signalling complex with a 9-fold higher K_D

(13.5nM) compared to that for the sIL-lR/lL-1/3 binary complex (K_D 1.53nM). However, some type II receptors have shown a higher affinity with a preference for IL-1/3 over IL-lα and IL-lra (Liu et al . 1996) . But the k_off of the sIL-lRIl/lL-lβ binary complex is significantly slower than that of a type I binary complex, indicating a physiological role for the type II receptor.

EXAMPLE 4 : Optimization of the labelling of the proteins for the HTRF Assays

1) Method

The interactions characterized above were adapted and developed for use in an assay based on the HTRF^® techno1ogy.

Proteins

Proteins at the following concentrations were used in the development of assays. - IL-1/3 (Mw 17,400)

Concentrations .42mg/ml - Extracellular domain of IL-IR type I (sIL-lRI) (MW 44,000)

Concentrations .49mg/ml

- Extracellular domain of IL-lRAcP (sIL-lRAcP) with a C- terminal 6-His tag (MW 45,000) Concentrations .58mg/ml

- Interleukin-1 receptor antagonist (IL-lra) (MW 17,100)

Concentration=20μg/ml

- Peptide AF11377 (Seq No: 35) (MW 1,858) Concentration=lmM

Labelling of Reagents

The reagents were labelled with activated-Europium Cryptate, Activated-XL665 or Activated-CY5 through amine groups using heterobifunctional reagents and purified using gel filtration chromatography to remove unreacted reagents, according to standard protocols for amine coupling via EDC/NHS activation.

Cryptate labelling

- sIL-1 Receptor type I

Labelled sIL-1 Receptor at a concentration of 0.130 mg/ml and a Rmf=2.05 K/Receptor (Ratio of Dye Cy5/Protein) . The product was stored in Phosphate buffer lOOmM pH 7.0 with 0.1% BSA and 0.1% Tween 20 and stored at -80°C under aliquots of lOμl. CY5 labelling

- sIL-1 Receptor Accessory protein:

Labelled sIL-lRAcP at a concentration of 0.065mg/ml and Rmf=2.6 CY5/sIL-lRAcP was stored in Phosphate buffer lOOmM pH 7.0 with 0.1% BSA and stored at -80°C in lOμl aliquots .

1.1) Semi direct assay

The semi-direct system consisted of tagging the sIL- lRAcP with a 6His affinity-tag. However, this protein was unable to bind to its anti6His-K conjugate.

1.2) Direct assays

Then several direct assay formats were tested.

In the first direct assay to be tested the sIL- 1RI is directly labelled with cryptate and the sIL- IRAcP is labelled with XL665.

However, when the trimolecular complex was formed, maximum signal obtained with XL665 was very low (around 30%) . The low signal being attributed to high steric hindrance .

In the second assay system the sIL-lRAcP was directly labelled with cryptate and the sIL-lR was labelled with XL665. In these conditions no signal was detected.

The third assay system to be tested corresponds to the first direct assay system in which XL665 is replaced by CY5. In these conditions a maximum specific signal of around 1200% was obtained.

This third assay system has been choosen.

Example 5 : HTRF Direct Assay System

A direct assay format as shown in Figure 8 was designed involving sIL-lRI directly labelled with Europium cryptate and sIL-lRAcP directly labelled with CY5.

The interaction reaction may be carried out in a one step or a two step procedure: binary complex formation followed by ternary complex formation. After testing as described below, the two procedures were shown to give equivalent results so only the one step procedure was used subsequently.

All the reagents were diluted in HEPES buffer lOmM, pH 7.4, 0.1%BSA, 0.005% Tween 20 and 0.2M KF.

1) Protocol

The assay was set up as follows: 50μl IL-1/3 (or buffer in the negative test) 50μl Cryptate conjugate 50μl CY5 conjugate 50μl buffer The reaction was incubated at room temperature for at least 15 minutes, the equilibrium being reached after about 1 hour (see figure 10) . Readings were taken on a Discovery instrument (Packard) under standards conditions for HTRF measurement (delay 50μs, gate 400μs) .

2) Results

The results were expressed as DeltaF : R=ratio= (F_66Ξnm/F₆₂on ) xlO⁴

DeltaF ( s) = [ (Rpositive^_Rnegative)

XlOO

Negative=all reagents except IL-1/3 ligand Positive=all reagents with IL-1/3 ligand

Evaluation of the Assay Performances

The first experiment was performed in the following conditions :

IL-1/3 was used at a final concentration of 50 nM. sIL-lRI-Eur was used at final concentration of 1.25 or 5 nM. sIL-lRAcP-CY5 was used at final concentration of 10 or

50 nM.

The results obtained are given in table 3 below.

Table 3 :

As expected we observe that in the absence of IL-1/3 ligand, sIL-lRI-Eur and sIL-lRAcP-CY5 do not interact with each other thus no energy transfer signal was generated in the negative.

In presence of the ligand, the formed ternary complex gives rise to a high-energy transfer signal. A signal of around 1200% was observed after a lOmin incubation.

Specificity of the signal

In the absence of the ligand no energy transfer signal was obtained.

We also investigated the absence of signal by replacing either sIL-lRI-Eur by an antibody-Eur conjugate or sIL- 1RACP-CY5 by an antibody-CY5 conjugate both in negative

(without ligand) and positive (with ligand) wells.

These reagents were used at about the same concentration in cryptate or CY5 than the reference reagents we are using in the assay.

No non-specific signal was detected using antibody-Eur instead of sIL-lRI-Eur. No non-specific signal was detected using antibody-CY5 instead of sIL-lRAcP.

Kinetics and stability of the signal

sIL-1/3 was used at a final concentration of 10 nM sIL-lRI-Eur was used at a final concentration of 1.25 nM sIL-lRAcP-CY5 was used at a final concentration of 10 nM Reading after 10 mins, 30 mins, 3 hours and overnight.

The results obtained are shown in Figure 10. The equilibrium is reached after 10 mins at room temperature. After an overnight incubation at room temperature less than 10% of the signal is lost.

Influence of the KF concentration

IL-1/3 was used at a final concentration of 10 nM sIL-lRI-Eur was used at a final concentration of 1.25 nM sIL-lRAcP-CY5 was used at a final concentration of 10 nM All reagents were diluted in HEPES lOmM, pH 7.0, 0.1% BSA, 0.005% Tween 20 and KF 0. IM, 0.2M or 0.4M.

The results obtained are shown in the table 4 below.

Table 4

KF concentration between 0. IM and 0.4M has no effect on the signal .

Binding Analysis

Influence of IL-lβ concentration

ILlβ was used at increasing final concentrations from

0.156 to 10 nM sIL-lRI-Eur was used at a final concentration of 1.25 nM

(data not shown) . sIL-lRAcP-CY5 was used at a final concentration of 10 nM The results obtained are shown in Figure 11.

There is a direct relationship between the Delta F obtained and IL-1/3 concentration. The signal plateau is reached at 5 nM final concentration of IL-1/3. From the saturation binding curve the K_d for IL-1/3 was estimated at around 1.0 nM.

Influence of sIL-lRAcP-CY5 concentration

IL-1/3 was used at a final concentration of 10 nM sIL-lRI-Eur was used at a final concentration of 1.25 nM sIL-lRAcP-CY5 was used at increasing final concentrations from 0.078 to 10 nM

The results obtained are shown in Figure 11.

There is a direct relationship between the Delta F obtained and sIL-lRAcP-CY5 concentration. A plateau is obtained from a 10 nM final concentration of sIL-lRAcP- CY5. From the saturation binding curve the Kd for sIL- lRAcP was estimated at around 1.5 nM.

Influence of the sIL-lRI-Eur concentration

IL-1/3 was used at a final concentration of 10 nM. sIL-lRI-Eur was used at increasing final concentrations from 0.312 to 20 nM. sIL-lRAcP-CY5 was used at a final concentration of 10 nM.

There is a direct relationship between the specific signal (at 665 nm) obtained and sIL-lRI-Eur concentration. A plateau is obtained from lOnM final concentration of sIL-lRI-Eur. From the saturation binding curve the Kd for sIL-lRI was estimated at around 3.5 nM.

The maximum delta F% was obtained using sIL-lRI-Eur from 1.25 to 2.5 nM as shown in the table 5 below.

Table 5 :

Influence of unlabelled sIL-lRI concentration

ILl-beta was used at a final concentration of 10 nM. sIL-lRI-Eur was used at a final concentration of 1.25 nM. sIL-lRAcP-CY5 was used at a final concentration of 10 nM.

Unlabelled sIL-lRI was used at increasing final concentrations from 0.098 to 50 nM. The reaction was set up as follows:

50μl sIL-lRI-Eur

50μl unlabelled sIL-lRI 50μl SIL-1RACP-CY5

50μl IL-1/3 (or buffer in the negative test) The results are shown in Figure 12.

50% inhibition is measured with an unlabelled sIL-lRI concentration of 25 nM.

Influence of unlabelled sIL-lRAcP concentration

IL-1/3 was used at a final concentration of 10 nM sIL-lRI-Eur was used at a final concentration of 1.25 nM sIL-lRAcP-CY5 was used at a final concentration of 2 nM Unlabelled sIL-lRAcP was used at increasing final concentrations from 0.78 to 100 nM

The reaction was set up as follows:

50μl IL-1/3 (or buffer in the negative test) 50μl unlabelled sIL-lRAcP 50μl SIL-1RACP-CY5 50μl sIL-lRI-Eur

Results are shown in figure 12.

50% inhibition is measured with an unlabelled sIL-lRAcP concentration of 10 nM.

Effect of ILl-ra

IL-1/3 was used at a final concentration of 2 nM sIL-lRI-Eur was used at a final concentration of 1.25 nM sIL-lRAcP-CY5 was used at a final concentration of 10 nM IL-l-ra antagonist was used at increasing final concentrations from 0.078 to 20 nM The reaction was set up as follows:

50μl IL-1/3 (or buffer in the negative test)

50μl IL-lra

50μl SIL-1RACP-CY5

50μl sIL-lRI-Eur Readings were taken after 15 mins, 1, 3 and 5 hours.

Equilibrium is reached after about 3 hours. At the equilibrium, 50% inhibition is measured with a IL-lra concentration of around InM.

These results are shown on figure 13.

Effect of AF11377 inhibitor peptide

IL-1/3 was used at a final concentration of 2nM sIL-lRI-Eur was used at a final concentration of 1.25nM sIL-lRAcP-CY5 was used at a final concentration of lOnM

AF11377 inhibitor peptide (Seq No: 35) was used at increasing final 'concentrations from 0.78 to 20 nM

The reaction was set up as follows:

50μl IL-1/3 (or buffer in the negative test) 50μl AF11377 50μl SIL-1RACP-CY5

50μl sIL-lRI-Eur

Readings were taken after 15 mins, 1, 3 and 5 hours. Equilibrium is reached after about 2 hours. At the equilibrium, 50% inhibition is measured with an AF11377 peptide concentration of around 20 nM. These results are shown on figure 13.

Signal obtained at about K_D concentration for each reagent

The assay was performed using the three reagents at a concentration near the K_D.

IL-1/3 was used at a final concentration of 2 nM sIL-lRI-Eur was used at a final concentration of 1.25 nM sIL-lRAcP-CY5 was used at a final concentration of 2r_M

The following table 6 shows the signal obtained, (cps means counts per second)

Table 6 :

HTRF Assay Summary

Table 7 below shows the preferred reagent concentrations for use in the HTRF ternary complex assay, on the basis of the developmental work described herein. Table 7:

Further optimizations were performed in-house in order to reduce the final assay volume from 200μl to 16μl and to transfer the assay from 96-well to 384-well format. Table 7 concentrations were conserved for the assay but volumes of each component were reduced from 50μl to 4μl in order to obtain 16μl as a final volume. The signal and stability of the complex were identical to original format (data not shown) .

Example 6: Best in-house optimization of the HTRF direct assay system

1) Materials and Methods Assay Buffer lOmM Hepes

0.2M KF

0.1% BSA pH to 7.4 with NaOH at 21°C

Required for all reagent and compound dilutions

Assay components

1. Soluble ILl-receptor-1- labelled with Europium Cryptate (sIL-lRI-Eur)

Batch 002 (labelled by Cis Bio) . Store at -80°C.

Stock concentration 200μg/ml. (4.762μM) Dilute 1:1190 to give 4nM (diluted 4 fold in assay to give a final concentration of InM)

2. Soluble IL-1 receptor accessory protein labelled with Cy5 (sIL-lRAcP-Cy5)

Batch 7 (Labelled by PGRD Cambridge) . Stock concentration 3.6μM

Dilute 1:90 to give 40nM (diluted 4 fold in assay to give a final concentration of lOnM)

3. ILlβ

Batch 1 (Produced at PGRD Cambridge) . Store at - 80°C.

Stock Concentration 5.0 μM Dilute 1:625 to give 8nM (diluted 4 fold in assay to give a final concentration of 2nM)

4. Test compounds/buffer

The assay was set forth as follow in order to use the biggest possible volumes : - 1.3 μl of compound in 10 % DMSO (250μM daughter plates)

- 6.7 μl of ILlβ, initial concentration = 4.8nM, final concentration 2nM - 4 μl of sIL-lRI-Eur, initial concentration =

2nM, final concentration InM

- 4 μl of sIL-lRAcP-Cy5, initial concentration = 20nM, final concentration lOnM

Microtitre Plates

All assays were performed in black plates.

Black Optiplate (96 well) , Packard Catalogue No 6005207

60 plates per box Assay volume 200μl

Black Optiplate (384 well) Packard Catalogue No 6005256

30 plates per box Assay volume 64μl

Black Proxiplate (384 well) , Packard Catalogue No 6006260

50 plates per box Assay volume 16μl

Blanks and system controls

Blanks/System Controls were incorporated into each plate, in duplicate as follows:

(All concentrations indicate final concentration in assay) 1. Buffer

2. sIL-lRAcP-Cy5 (lOnM)

3. sIL-lRI-Eur (l.OnM)

4. sIL-lRAcP-Cy5 (lOnM) + sIL-lRI-Eur (l.OnM) 5. sIL-lRAcP-Cy5 (lOnM) + sIL-lRI-Eur (l.OnM) + ILlβ (2nM)

Control 4 was used as the system blank for all calculations . Control 5 produces the maximum possible signal .

2) Protocol

Assays were set up as indicated in the Table below. Reagents and test compounds were all diluted to the appropriate concentration with assay buffer. The % DMSO in each well was <1% in all assays performed.

Reagents were added into the assay in order of appearance in the table .

Formation of the ternary complex was initiated by addition of IL-lβ (2nM for competition experiments) .

Plates were sealed with TopSeal A (Packard) to prevent evaporation and incubated for a minimum of 3 hours, but generally overnight, prior to reading. The signal was stable for at least 24 hours.

Assay Composition See table 8 below. Table 8

Plate Reading

All validation studies experiments (using current reagents) were read using a Wallac Victor2 Plate Reader .

Early assay validation was performed using default settings on the Discovery Plate Reader. Note that a minimum assay volume of 60μl is required for optimal reading in standard 384 well plates. Excitation wavelenth 337nm. Emission wavelengths 620 and 665nm.

Example 7 HTRF assay Semi-direct System

A semi-direct assay format was designed utilizing biotinylated sIL-lRI and sIL-lRAcP directly labelled with CY5. Europium Cryptate coupled to Streptavidin was also used.

The interaction reaction was carried out in a one step procedure just by adding all components. All the reagents were diluted in HEPES buffer lOmM, pH 7.4, 0.1%BSA, 0.005% Tween 20 and 0.2M KF .

1) Protocol

The assay was set up as follows:

50μl IL-1/3 (or buffer in the negative test) 50μl Biotinylated-sIL-lRI and Streptavidin-Cryptate conjugate

50μl sIL-lRAcP-CY5 conjugate 50μl buffer

The reaction was incubated at room temperature for at least 15 minutes, the equilibrium being reached after about 1 hour (see figure 10) . Readings were taken on a Discovery instrument (Packard) under standards conditions for HTRF measurement (delay 50μs, gate 400μs) .

2) Results

The results were expressed as DeltaF :

R=ratio= ( F_665nm/F_S20 ) xlO⁴

DeltaF ( % ) = [ (Rpositive^~Rnegative) /RnegativeJ Xl00

Evaluation of the Assay Performances

The first experiment was performed in the following conditions:

IL-1/3 was used at a final concentration of 5nM. Biotin-sIL-lRI was used at a final concentration of 1.25 or 5nM. Streptavidin-Cryptate was used at 20ng/well. sIL-lRAcP-CY5 was used at a final concentration of 10 or 50nM.

EXAMPLE 8: Flashplate assay

The C-terminaly 6His tagged sIL-lRAcP is used to anchor the protein to a Ni-NTA coated flashplate (manufactured by NEN) . We know already from the BIAcore data that anchored in this orientation the sIL-lRAcP can interact with the IL-lβ/sIL-lRI binary complex to form a ternary complex. Thus this interaction is detected through the use of either tritium ( [³H] ) labelled IL-lβ or sIL-lRI. Light, which can be detected on an instrument such as a Packard TOPcount β-scintillation counter, is emitted from the flashplate upon binding of the radiolabelled binary complex to the sIL-lRAcP. Excitation of the scintillant embedded in the flashplate well occurs only when the radionuclide is in sufficient proximity to the well-surface i.e. when the radiolabelled binary complex has formed a ternary complex with the anchored sIL- lRAcP.

It is also possible to use the C-terminaly 6His tagged sIL-lRI in order to anchor the protein to a Ni-NTA coated flashplate (manufactured by NEN) . The ability of the IL-IRI to form firstly a binary complex with IL-lβ and subsequently a ternary with sIL-lRAcP has been demonstrated on the BIAcore with anchored sIL-lRI (not necessarily anchored via the C-terminus of the protein) . Thus in this format one may assay both for the formation of the binary and the ternary complex. When the binary complex formation is assayed [³H] labelled IL-lβ is used which results in scintillation upon interaction with sIL-lRI anchored to a flashplate. With the sIL-lRAcP labelled with [³H] the formation of the ternary complex is assayed.

Thus using an appropriate combination of the above assay formats one assays for molecules which may inhibit either the formation of the binary or ternary complex.

Alternatively the assay described above is run using either biotinylated sIL-lRI or biotinylated sIL-lRAcP. Thus one exploits the presence of these biotin moieties on the protein to anchor it to a streptavidin flashplate.

Biotinylation is also used as a method for [³H] labelling these molecules using commercially available [³H] -biotin. Labelling using [³H] -amino acids in the expression media is also a method for labelling these proteins with a high-specific activity.

Similarly SPA bead technology (Amersham) could be used as an alternative to the Flashplate method described above. Thus the flashplate would be replaced by a standard plastic plate (96 or 384-well) containing an appropriate quantity of SPA beads

Example 9 Orige Technology Assa (Igen)

sIL-lRAcP-6His is anchored to a Ni-NTA magnetic bead via the 6his tag and binary complex comprising ruthenium (II) tris- (bipyridine) labelled sIL-lR associated with IL-1/3 is then added. Light output from the label is then measured following electrochemical stimulation. Another possibility is to have sIL-lR-6His anchored to a Ni-NTA magnetic bead via the 6his tag and binary complex comprising ruthenium (II) tris- (bipyridine) labelled sIL-lRAcP associated with IL-1/3 or labelled IL-lβ associated with IL-lRAcP is then added. Light output from the label is then measured following electrochemical stimulation. A futher method for anchoring the appropriate protein to a magnetic bead would be to use streptavidin-coated beads and to biotinylate either sIL-lR or sIL-lRAcP as appropriate.

Table 9 shows the GenBank Accession numbers of nucleotide sequences which encode proteins involved in IL-1 bimolecular and/or trimolecular complex.

TABLE 9 , Table of Interleukin-1 relevant accession numbers

IL-1 Receptor type I Receptor Accessory protein Receptor type II Soluble AcP Soluble receptor type I

Human x02532 xl6896 AF029213 NM 004633 AF167343

Human X02531

Mouse NM_008361 NM_008362 NM 008364 NM 010555

Mouse NM 010554

Rat D00403 m95578 NM 012968

Rat M98820 NM 013123

Table 10 shows the GenBank Accession numbers of nucleotide sequences which encode proteins involved in IL-18 bimolecular and/or trimolecular complex.

TABLE 10

Table of Interleukin-18 relevant accession numbers

IL- 18 Receptor type I Receptor Accessory protein IL-18 binding protein

Human AF077611 NM 003855 NM_003853 AF215907

c c

Mouse NM 008360 NM 008365 NM 010553

REFERENCES

1. Dinarello, C. A. (1994) FASEB J. 8, 1314-1325

2. Clore, G. M. and Gronenborn, A. M. (1991) J. Mol . Biol. 221, 47-53

3. Labriola Tompkins, E. et al.(1991) Proc. Natl. Acad. Sci. U. S. A. 88, 11182-11186

4. Ju, G. et al(1991) Proc. Natl. Acad. Sci. U. S. A. 88, 2658-2662

5. Camacho, N. P. et al . (1993) Biochemistry 32, 8749-8757

6. Grutter, M. G. et al (1994) Protein Eng. 7, 663-671

7. Schreuder, H. et al (1995) FEBS Lett. 373, 39-40

8. Evans, R. J. et al (1995) J. Biol. Chem. 270, 11477- 11483

9. Vigers, G. P. et al (1994) J. Biol. Chem. 269, 12874- 12879

10. Yanofsky, S. D. et al (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 7381-7386

11. Wesche, H. et al (1996) FEBS Lett. 391, 104-108

12. Sarubbi, E. et al (1996) Anal. Biochem. 237, 70-75 13. Einstein, R. et al (1996) Cytokine. 8, 206-213

14. Wesche, H. et al (1997) J. Biol. Chem. 272, 7727- 7731

15. Schreuder, H. et al . (1997) Nature 386, 194-200

16. Reddy, S. A. (1997) J. Biol. Chem. 272, 29167-29173

17. Yoon, D. Y. and Dinarello, C. A. (1998) J. Immunol. 160, 3170-3179

18. Wesche, H. et al (1998) FEBS Lett. 429, 303-306

19. Malinowsky, D et al (1998) FEBS Lett. 429, 299-302

20. Volpe, F. et al.(1997) FEBS Lett. 419, 41-44

21. Dirk E. Smith et al . (2000), J. Biol. Chem. 2000 275, 1169-1175

22. Kolb, J. M., et al . (1996) Journal of biomolecular screening 1 (4) , 203-210

23. Vigers, G. P. . et al . (1997), Nature 386, 190-200

24. Sims, J. E. et al . (1988) Science, 241(4865), 585-9

25. SIMS, J. E. et al . (1993) Proc-Natl-Acad-Sci-U-S-A, 90 (13) , 6155-9

26. MCMAHAN, C. J. et al . (1991), EMBO-J. , 10(10), 2821- 32; 27. DINARELLO, C. A. (1996) , Blood, 87(6), 2095-2147;

28. KORHERR C. et al . (1997), Eur-J-Immunol . , 27(1), 262-7

29. D'ETTORRE, C. et al . (1997), Eur-Cytokine-Netw. , 8(2), 161-71

30. JONSSON, B. et al . (1991), Anal-Biochem. , 198(2), 268-77

31. HUANG, J. et al . (1997), Proc-Natl-Acad-Sci-U-S-A. , 94(24), 12829-32

32. HAMMOND, E. A. et al . (1999), Brain, 122 ( Pt 9), 1697-707

33. COLOTTA, F. et al . (1993), Science, 261(5120), 472-5

34. LIU, C. et al . (1996), J-Biol-Chem. , 271(34), 20965- 72

35. GREENFEDER S. A. et al . (1995), J-Biol-Chem. , 270(23) , 13757-65

36. LAYE, S. et al . (1998), FEBS-Lett., 429(3), 307-11.

37. SMITH et al (2000) J-Biol-Chem. 2000 Jan 14; 275(2) : 1169-75

38. JENSEN L. E.et al . (2000) The Journal of Immunology, 164, 5277-5286. 39. Rosenwasser LJ (1998) J Allerg Clin Immunol 102: 344-350.

40. Akira S (2000) Curr Opin Immunol 12: 59-63.

41. Cao R, Farnebo J, Kurimoto M, Cao Y (1999) FASEB J 13: 2195-2202.

42. Dinarello CA (1999b) J Allergy Clin Immunol 103: 11- 24.

43. Dinarello CA (1999a) Methods 19: 121-132.

44. Gracie JA, Forsey RJ, Chan WL, Gilmour A, Leung BP, Greer MR, Kennedy K, Carter R, Wei XQ, Xu D, Field M, Foulis A, Liew FY, Mclnnes IB (1999) J Clin Invest 104: 1393-1401.

45. Monteleone G, Trapasso F, Parrello T, Biancone L, Stella A, luliano R, Luzza F, Fusco A, Pallone F (1999) J Immunol 163: 143-147.

46. Reznikov, L. L. , Fantuzzi, G. , Tsang, M., Novick, D., Kim, S-H, Rubinstein, M. , and et al . in press. 2000

47. Reznikov LL, Kim SH, Westcott JY, Frishman J, Fantuzzi G, Novick D, Rubinstein M, Dinarello CA (2000b) Proc Natl Acad Sci U S A 97 : 2174-2179.

48. Neumann, D., Kollewe, C, Martin, MU. , and Boraschi, D. (2000) J. Immunology, 165(6), 3350-3357. 49. Nielsen, H. , Engelbrecht, J. , Brunak S., and von Heijne G. , (1997) Protein Engineering, 10, 1-6.

50. Hofmann, K. and Stoffel, W. (1993), Biol. Chem. Hoppe-Seyler 347,166

51. Lang D, Knop J, Wesche H, Raffetseder U, Kurrle R, Boraschi D, and Martin MU (1998) , J Immunol;

161 (12) :6871-7

52. Kumar et al . , (2000), JBC, 275: 10308-10314

53.Andrew Bowie and Luke A.J O'Neill The Interleukin-1 receptor/Toll-like receptor superfamily: signal generators for pro-inflammatory interleukins and microbial products Journal of Leukocyte Biology (2000) 67, 508-514.

54. Born et al . , (2000), JBC, 275: 29946-29954

55. Carrie et al . , (1999), Nature genetics, 23: 25-31

56. Sana et al . , (2000)

Claims

CLAIMS :

1) Assay method for determining the ability of a test compound to modulate the formation of a trimolecular complex including IL, a soluble IL-R and a soluble IL- RAcP, the method comprising:

2) Assay method for determining the ability of a test compound to disrupt or interfere with the formation of a trimolecular complex including IL, a soluble IL-R and a soluble IL-RAcP, the method comprising:

(a) providing a test compound;

(c) adding an IL polypeptide to the mixture obtained in step (b) ; and

(d) determining the amount of said trimolecular complex formed.

3) Assay method according to claim 1 for determining the ability of a test compound to enhance the formation of a trimolecular complex including IL, a soluble IL-R and a soluble IL-RAcP, the method further comprising:

-the addition of a soluble IL antagonist at step (a) ; and

-an additional step (c) in which the amount of said trimolecular complex formed at step (a) is compared with the amount of said trimolecular complex formed in the absence of test compound.

4) Assay method for determining the ability of a test compound to enhance the formation of a trimolecular complex including IL, a soluble IL-R and a soluble IL- RAcP, the method comprising: : (a) providing a test compound;

(c) adding an IL polypeptide to the mixture obtained at step (b) ;

(d) determining the amount of said trimolecular complex formed; and

5) Assay method according to claim 1 for determining the ability of a test compound to enhance the formation of a trimolecular complex including IL, a soluble IL-R and a soluble IL-RAcP, the method further comprising:

-the addition of an IL-R antagonist at step (a) ; and

-an additional step (c) in which the amount of said trimolecular complex formed at step (a) is compared with the amount of said trimolecular complex formed in the absence of test compound. 6) Assay method for determining the ability of a test compound to enhance the formation of a trimolecular complex including IL, a soluble IL-R and a soluble IL- RAcP, the method comprising: : (a) providing a test compound;

(c) adding a soluble IL-R polypeptide to the mixture obtained in step (b) ;

(d) determining the amount of said trimolecular complex formed; and

7) Assay method for determining the ability of a test compound to disrupt or interfere with the stability of a trimolecular complex including IL, a soluble IL-R and a soluble IL-RAcP, the method comprising:

(b) contacting a test compound with said trimolecular complex; and

8) Assay method according to any one of claim^'s 1 to 7 wherein the IL, sIL-R and sIL-RAcP are of mammalian origins .

9) Assay method according to any one of claims 1 to 8 wherein the IL, sIL-R and sIL-RAcP are from human, mouse or rat .

10) Assay method according to any one of claims 1 to 9 wherein the IL polypeptide is an IL amino acid sequence or a fragment, derivative, analog or active portion thereof which retains the biological activity of IL.

11) Assay method according to any one of claims 1 to 10 wherein the IL polypeptide is in its mature form and has a barrel shaped 3 -dimensional structure composed of 12 to 13 beta-strands and a cellular activity mediated by a ternary (i.e. trimolecular) complex at the cell surface.

12) Assay method according to any one of claims 1 to 11 wherein the IL polypeptide is an IL-1 or an IL-18 polypeptide.

13) Assay method according to claim 11 wherein the IL polypeptide has an amino acid sequence of SEQ ID NO: 1, or SEQ ID NO: 4, or SEQ ID NO: 5, or SEQ ID NO: 6, or SEQ ID NO: 7, or SEQ ID NO: 8, or SEQ ID NO : 9 , or SEQ ID NO: 10, or a fragment thereof.

14) Assay method according to any one of claims 1 to 13 wherein the soluble IL-R polypeptide is any soluble polypeptide fragment of a full-length IL-R which retains the extracellular binding activity of the full length protein. 15) Assay method according to any one of claims 1 to 14 wherein the soluble IL-R polypeptide comprises 3 Ig-like domains which are included in three structural domains identified as domain 1, domain 2 and domain 3 starting from the N-terminal end of the sequence, said domain 1 containing one Ig-like domain and two disulfide bonds; said domain 2 containing one Ig-like domain and two overlapping disulfide bonds; said domain 3 containing one Ig-like domain and one disulfide bond.

16) Assay method according to any one of claims 1 to 15 wherein the soluble IL-R polypeptide is an IL-IR or an IL-18R polypeptide.

17) Assay method according to claim 15 wherein the soluble IL-R polypeptide has an amino acid sequence of SEQ ID NO: 13, or SEQ ID NO: 14, or SEQ ID NO: 15, or SEQ ID NO: 16, or SEQ ID NO: 17, or SEQ ID NO: 18, or a fragment thereof .

18) Assay method according to any one of claims 1 to 17 wherein the soluble IL-RAcP polypeptide is any soluble polypeptide fragment of a full-length IL-RAcP which retains the extracellular binding activity of the full length protein.

19) Assay method according to any one of claims 1 to 18 wherein the soluble IL-RAcP polypeptide comprises 3 Ig- like domains which are included in three structural domains identified as domain 1, domain 2 and domain 3 starting from the N-terminal end of the sequence, said domain 1 containing one Ig-like domain and two disulfide ill

bonds; said domain 2 containing one Ig-like domain and two overlapping disulfide bonds; said domain 3 containing one Ig-like domain and one disulfide bond.

20) Assay method according to any one of claims 1 to 19 wherein the soluble IL-RAcP polypeptide is an IL-lRAcP or an IL-18RAcP polypeptide.

21) An assay method according to claim 19 wherein the soluble IL-RAcP polypeptide has an amino acid sequence of SEQ ID NO: 21, or SEQ ID NO: 22, or SEQ ID NO: 23, or SEQ ID NO: 24, or SEQ ID NO: 25, or SEQ ID NO: 26, or a fragment thereof .

22) A soluble tri-molecular complex comprising or consisting of the following elements :

(a) an IL polypeptide

(b) a soluble IL-R polypeptide which binds said IL polypeptide, and (c) a soluble IL-RAcP polypeptide which binds to the IL-R polypeptide/lL polypeptide binary complex.

23) A soluble complex according to claim 22 wherein the IL, sIL-R and sIL-RAcP are of mammalian origins.

24) A soluble complex according to claim 22 wherein the IL, sIL-R and sIL-RAcP are from human, mouse or rat.

25) A soluble complex according to claim 22 wherein at least one of the IL, sIL-R or sIL-RAcP is from human.

26) A soluble complex according to any one of claims 22 to 25 wherein the IL polypeptide has an amino acid sequence of SEQ ID NO: 1, or SEQ ID NO: 4, or SEQ ID NO: 5, or SEQ ID NO: 6, or SEQ ID NO: 7, or SEQ ID NO: 8, or SEQ ID NO: 9, or SEQ ID NO: 10, or a fragment thereof.

27) A soluble complex according to any one of claims 22 to 26 wherein the soluble IL-R polypeptide has an amino acid sequence of SEQ ID NO: 13, or SEQ ID NO: 14, or SEQ ID NO: 15, or SEQ ID NO: 16, or SEQ ID NO: 17, or SEQ ID NO: 18, or a fragment thereof.

28) A soluble complex according to any one of claims 22 to 27 wherein the soluble IL-RAcP polypeptide has an amino acid sequence of SEQ ID NO: 21, or SEQ ID NO: 22, or SEQ ID NO: 23, or SEQ ID NO: 24, or SEQ ID NO: 25, or SEQ ID NO: 26, or a fragment thereof.