WO1998040743A1 - Immune complexes and methods of detection and treatment thereof - Google Patents

Abstract

This invention relates to the use of SCR-containing proteins (SCR-Ps) which bind to Clq, for assays and methods of removal of Clq-containing immune complexes, and diagnoses and treatments of immune complex-associated disorders and complement activation-associated disorders, and isolated and modified SCR-Ps for use in these assays, and methods and devices for the excorporeal removal of Clq-containing immune complexes from body fluids.

Description

IMMUNE COMPLEXES AND METHODS OF DETECTION AND TREATMENT THEREOF

RELATED APPLICATION

This application is a continuation-in-part application of United States patent application Serial No. 08/814,264 filed March 10, 1997, the teachings of which are incorporated by reference in their entirety.

FUNDING

This invention was made with Government support under grant number NIH ROl 34028. The Government has certain rights in the invention.

BACKGROUND OF THE INVENTION

The introduction of foreign or noxious materials into an individual is often followed by an immune response. Specific antibodies produced in the course of this response bind to antigens, forming immune complexes.

In general, these complexes are phagocytosed and removed from circulation and tissues by a variety of normal mechanisms such as by fixed macrophages found in the liver, spleen and lymph nodes, and by circulating macrophages. However, at times these complexes are deposited in organ or tissue sites, including the glomerulus of the kidney and blood vessel walls, contributing to compromised immune system function and inflammatory pathology. Immune complexes can also be deposited m the lung, heart and joints causing both transient and permanent damage to those organs .

Once deposited m tissues, immune complexes activate a variety of potent mediators of inflammation, such as complement proteins, causing an influx of polymorphonuclear neutrophils and monocytes. These activated cells release toxic products of oxygen metabolism as well as various proteases and other enzymes, ultimately causing tissue damage. The clinical features of these diseases are quite diverse, ranging from mild cutaneous eruptions to severe organ involvement with pericarditis, glomerulonephritis, and vasculitis.

Current methods of measuring immune complexes lack sensitivity and specificity. For example, some methods do not detect immune complexes of all isotypes and sizes, and some methods are inaccurate Decause of interference oy non- complexed immunoglobulins.

It would be useful to identify and characterize methods for use m physiologically accurate diagnostic methods and treatment of immune complex-associated disorders and complement activation-associated disorders.

SUMMARY OF THE INVENTION

The present invention relates to the surprising and unexpected demonstration that the complement protein, Clq, binds to proteins containing snort consensus repeats . This invention also relates to methods of assaying and removing Clq-containing immune complexes, and detecting and treating immune complex-associated disorders and complement activation-associated disorders, by binding Clq to proteins which comprise amino acid sequences known as short consensus repeats (referred to herein as SCRs) . These proteins are also referred to herein as short consensus repeat proteins, or SCR-Ps.

The methods of the present invention can utilize one SCR-P or a combination of different SCR-Ps. The SCR-Ps of the present invention can be oligomerized and subsequently purified to obtain an SCR-P suitable for use in the present methods, using techniques well-known in the art. For example, the purification can be by column chromatography. The invention also relates to isolated and purified SCR-Ps and SCR-Ps modified to alter (e.g., increase) binding to Clq. For example, an SCR-P engineered to have one binding site for Clq and another binding site for another natural ligand of SCR-P, such as C3b, can be useful for a specific assay of relevant immune complexes. The SCR-P can be present in a complement protein, for example, complement receptors or complement inhibitors such as CR1 , the LHR-D region of CR1 , CR2 , CD46, CD55, factor H, Clr, Cis and C4bp. SCR-P can also comprise biologically active SCR-P and fragments, analogs, and derivatives thereof, wherein biological activity is defined as binding to Clq, C3b, iC3b and C4b or other proteins which bind to SCR-P.

The methods of the present invention include methods of detecting the presence of Clq-containing immune complexes m a biological sample by contacting the biological sample with an SCR-P that binds to Clq, and detecting SCR-P-bound Clq-containmg immune complexes. Molecules comprising Clq-containmg immune complexes bound to SCR-P are referred to herein as SCR-P-bound immune complexes, immuno-complexes , and products. The Clq- contammg immune complexes and the SCR-Ps can be from any animal, such as a mammal, including a human. In one embodiment, the SCR-P binds to both Clq and one or more additional components present m the Clq-containmg complex, such as complement proteins C3b, ιC3b, and C4b.

The biological sample can be any biological fluid or tissue including, e.g., plasma, serum, processed erythrocytes, eluant from intact erythrocytes , brain tissue, skin tissue, urine, lymphatic fluid, peritoneal fluid, oint fluid, cerebrospmal fluid, pleural fluid or a fluid eluted from blood cells. In one embodiment, the biological sample is tissue obtained by biopsy, e.g., an allograft or a xenograft tissue biopsy sample. In another embodiment, the sample is produced by elution of immune complexes from erythrocytes.

In one embodiment, the SCR-P (e.g., soluble CR1 or the LHR-D region of CR1) is immobilized on a solid support (e.g., a microtiter well or a bead) . The sample to be assayed is contacted with the immobilized SCR-P (e.g., added to microtiter wells containing SCR-P) under conditions suitable for Clq-containmg immune complexes present m the sample to bind to the immobilized SCR-P. The immobilized SCR-P-bound Clq-containmg immune complex is detected with a detectably labeled reagent which specifically binds to the Clq-containmg immune complex which is bound to SCR-P, for example, a labeled anti- lmmunoglobulm antibody, such as anti-IgG, anti-IgM or anti-IgE, or an immunoglobulm-reactive substance, such as protein A or protein G. Other detectably labeled reagents suitable for use m this assay are known to those of skill the art .

In another embodiment, a non-complement fixing fragment of an antibody, such as the Fab ₂ fragment that binds Clq-contammg immune complexes, is immobilized on a solid support. The sample is pretreated to enrich for immune complexes by fractionation m a way that will not precipitate free Clq. Fractionation methods are known to those of skill m the art. The antibody fragment can bind Clq, C3b, ιC3b or C4b . The sample to be assayed is contacted with the immobilized antibody fragment under conditions suitable for Clq-contammg immune complexes present m the sample to bind to the immobilized antibody fragments. The immobilized antibody fragment-bound Clq- contammg immune complex can be detected with detectably- labeled SCR-P that specifically binds to Clq. The SCR-P can be oligomeπzed. The SCR-P can comprise one or more binding sites for Clq, and additionally a binding site for another complement component, such as C3b or C4b. The SCR- P can be labeled by any detectable label known m the art, including a radiolabel, an indicator enzyme, biotm and fluorochrome . In another embodiment, Clq-containing immune complexes in a biological sample can be detected by reacting the sample with SCR-P which binds to Clq-containing immune complexes for a time and under conditions suitable to bind Clq-containing immune complexes, and then separating the SCR-P bound and unbound Clq-containing immune complexes, and quantifying the bound SCR-P-Clq-containing complexes with an indicator reagent . The indicator reagent can be an anti-SCR-P antibody. In another embodiment, the Clq-containing immune complexes are measured by surface plasmon resonance analysis. For example, the SCR-P (e.g., CR1) is immobilized on a sensor chip and the sample is applied across the chip under flow conditions, and specific binding of the Clq-containing immune complex to the SCR-P bound to the chip is detected by an increase in resonance units.

In the methods of the present invention, the Clq- containing immuno complex in a sample can be quantitated by binding an SCR-P to Clq-containing immune complexes in the sample, detecting Clq-containing immune complex bound to SCR-P, and comparing the amount of Clq-containing immune complex bound to the SCR-P in the sample to known amounts of Clq-containing immune complexes bound to SCR-P from known amounts of standard samples which form a standard curve. This value can be compared to a normal range of values to determine if abnormal quantities of Clq- containing immune complex are present . The normal range of values is typically obtained from a population which is not suspected of having an immune complex-associated disorder or a complement activation-associated disorder, e.g., from a healthy clinical population. The standard curve can be generated from patient sera or from normal sera into which known amounts of Clq-contammg immune complexes have been added.

In another embodiment, the assay can be used to screen for the presence of an immune complex-associated disorder or a complement activation-associated disorder, in a mammal. In this embodiment, a biological sample is obtained from the mammal and contacted with an SCR-P which binds Clq-contammg immune complexes, and the amount of bound Clq-contammg immune complexes is compared with known amounts of Clq-contammg immune complex which form a standard curve and with a normal range of values. If the amount of Clq-contammg immune complex m the sample is greater than a pre-determined value of the normal range, it is indicative of the presence of an immune complex- associated or complement activation-associated disorder m the mammal . The present invention also relates to a kit for detecting the presence of an immune complex-associated disorder m a mammal, or a complement activation-associated disorder, comprising an SCR-P which binds Clq-contammg immune complexes and a means for determining this binding. This invention also relates to methods of assessing the efficacy of a drug (or a treatment or therapy) for the treatment of an immune complex-associated disorder or a complement activation-associated disorder, by comparing the quantity of Clq-contammg immune complexes m a biological sample obtained from a mammal before and after administration of the drug.

The present invention also relates to in vivo and in vi tro treatments, prophylactics, and therapies involving binding Clq to SCR-P and removing Clq-containing immune complexes bound to SCR-P. For example, Clq-containing immune complexes in fluid can bind to SCR-P, and the fluid can be separated from the Clq-containing immune complexes bound to the SCR-P. The methods of treatment described herein can be extracorporeal , or ex vivo treatment. For example, the fluid can be removed from the mammal, and contacted with SCR-P to form SCR-P bound Clq-containing immune complexes. If required, the biological fluid can be pretreated to enrich for immune complexes. The SCR-P bound immune complexes can be contacted with an antibody specific for the Clq-containing immune complexes to produce antibody- bound SCR-P bound immune complexes which can be separated from the fluid. The antibody can be a non-complement fixing Fab'₂ fragment of antibody. The treated biological fluid can be returned to the mammal after the SCR-P bound Clq-containing immune complexes are removed. The SCR-P can bind to one or more additional components present in the Clq-containing complex, such as complement proteins C3b, iC3b, and C4b .

In another embodiment, Clq-containing immune complexes can be removed from a mammal by insolubilizing SCR-P which binds to Clq-containing immune complexes, immobilizing the insolubilized SCR-P on a solid surface, contacting the plasma portion of a whole blood sample from the mammal (a mammalian plasma sample) with the immobilized SCR-P, separating the bound Clq-contammg immune complexes from the plasma, thus producing immune complex-depleted plasma, and readmmistermg (returning) the immune complex-depleted plasma to the human.

This invention also relates to methods of inhibiting complement activation m biological sample by contacting the sample with a fragment of Clq comprising an SCR-P binding site or an SCR-P whicn binds bound Clq. Bound Clq includes Clq bound to Clr, Cis, a complement activator or an immune complex. In another embodiment, complement activation can be inhibited m a mammal by administering to the mammal a fragment of Clq that comprises an SCR binding site and an SCR-P that binds to bound Clq.

The invention also relates to isolated and purified SCR-P polypeptides, e.g., the LHR-D region of CR1 , and biologically active SCR-P fragments, analogs, or derivatives, that bind to Clq-contammg immune complexes. The SCR-P or SCR-P fragments, analogs or derivatives can be modified to optimize SCR-P binding to Clq or Clq-contammg immune complexes. The SCR-P polypeptides of the present invention can also bind to C3b, ιC3b and C4b, m addition to Clq. The present invention also relates to a fragment of Clq comprising an SCR-P binding site. Alternatively, the SCR-P binding site on Clq could be engineered into a therapeutic agent. Such agents could be used to inhibit Clq activation by inhibiting Clr and Cis from binding to Clq. The invention further relates to a device for removing Clq-contammg immune complexes from fluids, comprising SCR-P bound to a solid surface which binds to Clq- contammg immune complexes, and a means for encasing the solid surface so that the fluid may be contacted with the solid surface.

The invention further relates to methods of prognosis and prevention. The methods described herein can be useful to predict flare-ups associated with the presence of immune complexes, for example, m autoimmune disease. It also relates to the use of one or more SCR-Ps m a therapeutic composition or medicament.

As a result of the demonstration that the complement protein Clq binds to proteins containing short consensus repeats, physiologically accurate methods of detection and treatment of immune complex disorders are now available.

BRIEF DESCRIPTION OF THE DRAWINGS

Figures 1A-1D depict FACS analyses of CR1 expression on transfected cells. Figure IA: cells transfected with the control plasmid Hyg and reacted with control mAb.

Figure IB: cells transfected with control plasmid Hyg and reacted with anti-CRl mAb YZ1. Figure 1C: cells transfected with CR1 and reacted with control mAb. Figure ID: cells transfected with CR1 and reacted with anti-CRl mAb YZ1. Figure IE depicts adherence of K-CR1 cells, but not K-Hyg cells, to immobilized (C3b)₂.

Figure 2A depicts a FACS analysis of Clq^bl° binding to transfected K-Hyg and K-CR1 cells. Figure 2B depicts an analysis of ¹²⁵I-Clq binding to transfected K-Hyg and K-CRl cells incubated with increasing doses of ¹²⁵I-Clq diluted in 1/2 ionic strength buffer. At the highest input of ¹²⁵I-Clq a 50 -fold excess of unlabeled ligand inhibited specific binding by 82%(*) . The means of triplicate values ±SE are depicted.

Figure 3 depicts the effect of ionic strength on Clq^bl° binding to insolubilized rsCRl . Results represent the mean of duplicate values, while the bars depict the range of values.

Figure 4A depicts direct ¹²⁵I-Clq binding to plated rsCRl .

Figure 4B depicts competition of ¹²⁵I-Clq binding to CR1 by native Clq. Figure 4C depicts competition of ¹²⁵I-Clq binding to CR1 by C3b dimers .

Figures 5A and 5B depict a surface plasmon resonance analysis of C3b dimers binding to CR1.

Figures 6A and 6B depict a surface plasmon resonance analysis of Clq binding to coupled CR1.

Figure 7 is a schematic indicating the binding domains for Clq, C3b and C4b on the CR1 molecule.

Figures 8 A-W depict amino acid sequences of SCRs of CR1 (CD35) (SEQ ID NOS: respectively. Figure 9 depicts the upregulation of Clq^bl° binding by PMA buffer and FMLP .

Figure 10 depicts the upregulation of Clq^bl° binding on all PMN by PMA buffer and FMLP. Figure 11 depicts collagen tail inhibition of ^{i 5}I-Cιq binding to immobilized rsCRl .

Figure 12 depicts binding of ^l2SI-collagen tails to equimolar amounts of immobilized rsCRl and a construct containing LHR-D.

Figure 13 depicts total and specific binding of ¹²⁵I- collagen tails to normal human erythrocytes. * refers to inhibition of binding by about 50%.

Figure 14 depicts Clq-dependent binding of ¹²⁵I-heat aggregated IgG (HAG) to human erythrocytes.

Figure 15 depicts a bar graph of ¹²⁵I-MBL binding to microtiter wells. The white bar depicts ¹⁵I-MBL binding to control blocked wells, the shaded bar depicts ¹²⁵I-MBL binding to CR1 (CD25) -coated wells, and the black bar depicts ¹²⁵I-MBL-mannan binding to CRl-coated wells.

Figure 16 depicts a bar graph of reciprocal binding using ⁱ²⁵I-fluιd phase ligand. The two left bars depict the results using immobilized Clq. The two right bars depict the results using immobilized factor H and immobilized C4bp.

DETAILED DESCRIPTION OF THE INVENTION

The complement system consists of multiple proteins which interact m a sequential manner to generate effectors of the immune system. When the same components are activated m an unregulated manner, they become effectors of inflammation. There are three known patnways to activate the complement system: an alternative pathway, a collection (lectin) pathway, and activation of the first component of complement (Cl) , either directly or by antigen-antibody complexes.

The present invention relates to activation of Cl . Cl is a multimeπc complex consisting of a 462 kDA recognition subunit, Clq, and 2 moles of each of the enzymatic subunits Clr and Cis. The binding of Clq causes the sequential activation of Clr and Cis. Activation occurs when Clq avidly binds IgG and IgM m immune complexes and thereby "complements" acquired immunity. In addition, Clq can directly bind DNA, C-reactive protein complexes, serum amyloid P, isolated myelm, urate crystals and some gram negative and positive bacteria. By directly binding foreign or abnormally expressed host molecules and subsequently recruiting the remainder of the classical pathway, Clq functions as part of the innate immune system. Specific Clq binding leads to the autoactivation of Clr. Activated Clr then activates Cis, and activated Cis, turn, activates C4 and C2 , sequentially by proteolytic cleavages. Activated C2 cleaves C3 and C5 , and the remaining components C6, C7 , C8 and C9 are sequentially condensed onto a complex initiated by activated C5. This complex is designated C5b-9, or the terminal complement complex, or the membrane attack complex (MAC) .

When activated by either the acquired or innate immune system, the catalytic function of Cl is tightly regulated by the Cl inhibitor which covalently binds to the activated enzymatic subunits of Cl, Clr and Cis, and removes them from Clq. Clq, still bound to its immune complex or activating substance, is now free to interact with cells bearing receptors for the collagen region of Clq.

Receptors for, or proteins that bind aggregated or insolubilized Clq have been recognized on lymphocytes (Dickler et al . , J. Exp . Med . , 136:191-196 (1972)), monocytes, some polymorphonuclear leukocytes (PMN) , and B lymphocytes. (Tenner et al . , J. Exp . Med . , 126:1174-1179 (1981)). However, molecular descriptions of the Clq receptors or binding proteins have been varied and inconsistent, perhaps in part due to the apparent presence of different receptors on different cell types. As described herein, for the first time, it has been demonstrated that Clq binds to the short consensus repeat containing complement protein, Complement Receptor 1 (CRl) . Also as described herein, it is demonstrated that Clq binds to other SCR containing proteins such as factor H and C4bp. These results indicate that Clq binds to a conserved portion of these proteins, the SCR portion. It has also been demonstrated, as described herein, that CRl binds to a proteolytic fragment of Clq containing only the collagen "tail" portion of the molecule. This indicates that CRl binds to the portion of Clq which remains after Clq has been digested with pepsin, and not the globular portion of Clq which remains bound to the activator or immune complex. Thus, CRl can bind to Clq which is bound in Clq-containing complexes. As a result of these surprising demonstrations, methods are now available to specifically detect and quantify Clq-containing immune complexes, to diagnose Clq- containing immune complex and complement activator- associated disorders and to specifically remove Clq- contammg immune complexes from biological samples.

SHORT CONSENSUS REPEAT PROTEINS

The present invention relates to binding of Clq to proteins comprising short consensus repeats (SCRs) . SCRs (also known as complement control repeats) are 56-70 ammo acid sequences, typically, 60 ammo acids. Each SCR shares invariant or highly conserved ammo acid residues with other SCRs m the same protein or SCRs m other proteins. SCRs are found m a number of proteins, including all members of a family of proteins known as "regulators of complement activation" (RCA) . RCA proteins are all related genetically. They are encoded at a single chromosomal location (q32) on chromosome 1, known as the RCA gene cluster. The RCA protein family includes complement receptor type 1 (CRl or CD35) , complement receptor type 2 (CR2 or CD21) , C4b-bmdmg protein (C4bp) , membrane co- factor protein (MCP or CD46) , decay accelerating factor (DAF or CD55) and factor H. Other proteins m the complement proteins have SCR domains as well. Such proteins include Clr, Cis, C6, C7 and factor B. The cDNAs corresponding to these proteins have been sequenced. Hourcade, D., et al . , Advances m Immunol . , 45:381-416 (1989) , which is incorporated herein by reference, describes the relationship of these proteins based on alignment of deduced ammo acids from the cDNAs of these proteins. It is most important that Clr and Cis contain SCRs, because it may mean that they are binding to Clq by the same SCR-binding domain on Clq that CRl, Factor H and C4bp may utilize. The SCRs form the extracellular portions of those proteins which are membrane bound, and most of the structure of those proteins which are secreted. Four cystine residues form two loops within each SCR. They also generally comprise a tryptophan.

One member of the RCA family of particular interest is CRl. CRl (CD35) is a single chain, transmembrane glycoprotein that has been recognized as the major cellular receptor for C4b and C3b (Fearon, J. Exp . Med . , 152:20-30 (1980)). Although CRl is bivalent for C3b, it is monovalent for C4b. C4b is the major cleavage fragment of complement C4. It is frequently attached covalently through a thioester bond to complement activators, including immune complexes. C3b is the major cleavage fragment of complement C3. It is frequently attached covalently through a thioester bond to complement activators, including immune complexes. The extracellular domain of CRl is comprised of short consensus repeats, which are characteristic of C3/C4 binding proteins. CRl is comprised of 30 SCR motifs, see for example Figures 8A-8W for the amino acid sequences of some SCRs from CRl . Groups of seven SCRs are further organized into four long homologous repeats (LHRs) . For example, the LHR-D region of the CRl has been demonstrated herein to bind Clq.

Proteins that comprise the short consensus repeat sequences are referred to herein as short consensus repeat proteins or SCR-Ps. As used herein, SCR-Ps include lsolated and purified naturally occurring proteins comprising SCRs, including, but not limited to, RCA proteins. SCR-Ps include full-length proteins, and biologically active SCR-P fragments, derivatives, analogs, variants and mutations.

The term "biological activity" of an SCR-P, or a fragment, derivative, analog, variant or mutant SCR-P, is defined herein as the activity of the SCR-P to specifically bind Clq, a fragment of Clq, a Clq-contammg immune complex or a Clq-contammg immune complex with additional components present such as C3b, ιC3b or C4b . Such activity can be measured by the methods described herein, or by other methods known to those skilled m the art. Another biological activity of an SCR-P protein is the antigenic property of inducing a specific immunological response as determined using well-known laboratory techniques. For example, a biologically active SCR-P can induce an immunological response which produces antibodies specific for the SCR-P (anti-SCR-P antibodies) . In the instant case, antibodies that specifically react with or bind to, an SCR-P can be produced.

To be "functionally or biologically active", a SCR-P typically shares significant sequence (ammo acid or nucleic acid) identity (e.g., at least about 65%, preferably at least about 80% and most preferably at least about 95%) with the corresponding sequences of the endogenous SCR-P and possess one or more of the functions thereof . The SCR-Ps of the present invention are understood to specifically include an SCR-P protein having ammo acid sequences analogous to the sequence of the endogenous protein. Such proteins are defined herein as SCR-P analogs. An "analog" is defined herein to mean an ammo acid sequence with sufficient identity to ammo acid sequence of the endogenous SCR-P protein to possess the biological activity of the protein. For example, an analog of a polypeptide can be introduced with "silent" changes in the ammo acid sequence wherein one or more ammo acid residues differ from the ammo acids residues of the SCR-P, yet still possess Clq binding activity. Examples of such differences include additions, deletions, or substitutes of residues. Also encompassed by the present invention are proteins that exhibit lesser or greater biological activity of the SCR-P proteins of the standard invention.

The present invention also encompasses the production and use of biologically active fragments of the SCR-Ps described herein. Such fragments can include only a part of the full length ammo acid sequence of the SCR-P, yet possess biological activity. As used herein, the term "biologically active fragment" means a fragment that can exert a biological or physiologic effect of the full-length protein, or has a biological characteristic, e.g., antigenicity, of the full-length protein. Such activities and characteristics are described above. Such fragments can be produced by ammo and carboxyl terminal deletions as well as internal deletions. Also included are active fragments of the protein, for example, as obtained by enzymatic digestion. Such peptide fragments can be tested for biological activity.

One example of a biologically active fragment for use m the present invention would be a fragment of CRl which contains a Clq binding site. CRl comprises groups of SCRs organized into LHRs (long homologous repeats) . See, for example, U.S. Patent No. 5,472,939, the contents which are incorporated herein m their entity. One such fragment would comprise LHR-D of CRl. "Derivatives" and "variants" of SCR-Ps are SCR-Ps which have been modified. They include SCR-Ps which have been modified by alterations m the ammo acid sequence associated with the SCRs. They also include, but are not limited to, truncated and hybrid forms of SCR-Ps. "Truncated" forms are shorter versions of SCR-Ps, typically modified so as to remove the C-termmal regions which effect binding or secretion or sometimes modified further by one or more SCRs. "Hybrid" forms are SCR-P that are composed of portions of two or more SCR-Ps, i.e., SCRs of one SCR-P combined with SCRs of one or more other SCR-Ps. Variants can be produced using the methods discussed below. The SCR-P gene can be mutated m vi tro or in vivo using techniques known m the art, for example, site- specific mutagenesis and oligonucleotide mutagenesis. Manipulations of the SCR-P protein sequence can be made at the protein level as well. Any of numerous chemical modifications can be carried out by known techniques including, but not limited to, specific chemical cleavage by cyanogen bromide, trypsin, and papain. It can also be structurally modified or denatured, for example, by heat or by being immobilized on a solid surface.

Modifications of the SCR-Ps of the present invention can be based on the ammo acid sequence of corresponding SCRs of the proteins as sites for alteration. By

"corresponding SCR" is meant the most highly homologous SCR as judged by the ammo acid sequence of the protein. Exon structure can m some cases facilitate this assignment. For example, SCRs 1-2 of CRl correspond to SCRs 2-3 of the complement protein, decay accelerating factor (DAF) . In addition, SCRs 1-2 of complement proteins factor H, CRl, C4bp and membrane co- factor protein (MCP) are corresponding sequences among these proteins. As noted above, CRl is organized into a series of long homologous repeats (LHRs) containing seven SCRs. CR2 comprises a series of long homologous repeats of four SCRs m length. SCR 1-2 of CRl corresponds to SCR 3-4, SCR 7-8, SCR 11-12 and SCR 15-16 of CR2. Examples of such modifications are described, for example, m Atkinson et al . , U. S. Patent No. 5,545,619, the teachings of which are incorporated herein by reference m their entirety.

The ammo acid sequences of the SCR-Ps of the present invention can be altered to optimize SCR-P binding to Clq, or Clq-contammg immune complexes by methods known m the art by introducing appropriate nucleotide changes into native or variant DNA encoding the SCR-P, or by in vi tro synthesis of tne desired SCR-P.

Alterations can be created outside or withm the SCR-P domain or domains identified, such as those involved m the mteraction with a native ligand of SCR-P, for example Clq, C3b, ιC3b, or C4b. SCR-P variants mutated to enhance their binding direct association, such as binding, with a native ligand of SCR-P can be useful as inhibitors of native biological activities mediated by a native ligand of SCR-P, should the native ligand, when bound to Clq, mediate a deleterious effect. Engineering an SCR-P with one binding site for Clq and one for C3b would allow a very specific assay of relevant immune complexes. In addition, such variants, as well as native SCR-P and binding fragments thereof, will be useful m the diagnosis of pathological conditions associated with the over-expression or under- expression of those proteins or a native ligand of SCR-P, such as Clq. In general, mutations can be conservative or non- conservative ammo acid substitutions, ammo acid insertions or ammo acid deletions. The mutations can be at or near (within 5 or 10 ammo acids) the Clq binding sites . More preferably, DNA encoding an SCR-P ammo acid sequence variant is prepared by site-directed mutagenesis of DNA that encodes a variant or a nonvaπant version of SCR-P. Site-directed (site-specifIC) mutagenesis allows the production of SCR-P variants through the use of specific oligonucleotide sequences that encode the DNA sequence of the desired mutation, as well as a sufficient numoer of adjacent nucleotides, to provide a primer sequence of sufficient size and sequence complexity to form a stable duplex on both sides of the deletion junction bemg traversed. Typically, a primer of about 20 to 25 nucleotides length is preferred, with about 5 to 10 residues on both sides of the junction of the sequence bemg altered. In general, the techniques of site-specifIC mutagenesis are well known the art, as exemplified by publications such as Edelman et al . , DNA 2, 183(1983) . The site-specific mutagenesis technique typically employs a phage vector that exists m both a smgle-stranded and double- stranded form. Typical vectors useful m site- directed mutagenesis include vectors such as the M13 phage, for example, as disclosed by Messing et al . , Third Cleveland Symposium on Macromolecules and Recombinant DNA, A. Walton, ed., Elsevier, Amsterdam (1981). This and other phage vectors are commercially available and their use is well-known to those skilled the art. A versatile and efficient procedure for the construction of oligonucleotide directed site-specific mutations in DΝA fragments using M13 -derived vectors was published by Zoller, M.J. and Smith, M., Nucleic Acids Res . 10:6487-6500 (1982)). Also, plasmid vectors that contain a smgle-stranded phage origin of replication can be employed to obtain smgle-stranded DΝA. Veira et al . , Meth . Enzymol . 153:3 (1987). Alternatively, nucleotide substitutions can be introduced by synthesizing the appropriate DΝA fragment in vi tro, and amplifying it by PCR procedures known the art.

In general, site-specific mutagenesis herewith can be performed by first obtaining a smgle-stranded vector that includes within its sequence a DΝA sequence that encodes the relevant protein. An oligonucleotide primer bearing the desired mutated sequence is prepared, generally synthetically, for example, by the method of Crea et al . , Proc . Natl . Acad . Sci . USA 75, 5765 (1978) . This primer can then be annealed with the smgle-stranded protein sequence-containing vector, and subjected to DNA- polymeπzmg enzymes such as E. coli polymerase I Klenow fragment, to complete the synthesis of the mutation-bearing strand. Thus, a heteroduplex is formed wherein one strand encodes the original non-mutated sequence and the second strand bears the desired mutation. This heteroduplex vector can then be used to transform appropriate host cells such as JM101 cells, and clones can be selected that include recombinant vectors bearing the mutated sequence arrangement. Thereafter, the mutated region can be removed and placed m an appropriate expression vector for protein production.

The PCR technique can also be used m creating ammo acid sequence variants of an SCR-P. When small amounts of template DNA are used as starting material m a PCR, primers that differ slightly m sequence from the corresponding region m a template DNA can be used to generate relatively large quantities of a specific DNA fragment that differs from the template sequence only at the positions where the primers differ from the template. For introduction of a mutation into a plasmid DNA, one of the primers can be designed to overlap the position of the mutation and to contain the mutation; the sequence of the other primer is preferably identical to a stretch of sequence of the opposite strand of the plasmid, but this sequence can be located anywhere along the plasmid DNA. It is preferred, however, that the sequence of the second primer is located within 500 nucleotides from that of the first, such that in the end the entire amplified region of DNA bounded by the primers can be easily sequenced. PCR amplification using a primer pair like the one just described results in a population of DNA fragments that differ at the position of the mutation specified by the primer, and possibly at other positions, as template copying is somewhat error-prone.

If the ratio of template to product material is extremely low, the vast majority of product DNA fragments incorporate the desired mutation (s) . This product can be used to replace the corresponding region in the plasmid that served as PCR template using standard DNA technology. Mutations at separate positions can be introduced simultaneously by either using a mutant second primer or performing a second PCR with different mutant primers and ligating the two resulting PCR fragments simultaneously to the vector fragment in a three (or more) part ligation. Another method for preparing variants, cassette mutagenesis, is based on the technique described by Wells et al . Gene 34, 315 (1985) . The starting material can be the plasmid (or vector) comprising the SCR-P DNA to be mutated. The codon(s) within the SCR-P to be mutated are identified. There must be unique restriction endonuclease sites on each side of the identified mutation site(s) . If such restriction sites do not exist, they can be generated using the above-described oligonucleotide-mediated mutagenesis method to introduce them at appropriate locations m the SCR-P DNA. After the restriction sites have been introduced into the plasmid, the plasmid is cut at these sites to linearize it. A double stranded oligonucleotide encoding the sequence of the DNA between the restriction sites but containing the desired mutation (s) is synthesized using standard procedures. The two strands are synthesized separately and then hybridized together using standard techniques. This double- stranded oligonucleotide is referred to as the cassette. This cassette s designed to have 3 ' and 5 ' ends that are compatible with the ends of the linearized plasmid, such that it can be directly ligated to the plasmid. This plasmid now contains the mutated SCR-P DNA sequence, that can be expressed to produce SCR-P with altered binding activity.

The SCR-Ps of the present invention specifically bind to Clq, fragments of Clq, Clq-contammg immune complexes and Clq-contammg immune complexes that have additional complement proteins present m the complex. The immune complexes of the present invention are defined herein as antigen-antibody complexes formed by the binding of an antibody (e.g., IgG or IgM) to an antigen. The immune complexes of the present invention typically comprise the complement protein Clq as well as other complement proteins such as e.g., C3b, ιC3b and C4b. ιC3b is C3b which has undergone proteolytic cleavage by factor I serum protein.

The antigens which form immune complexes include components of infectious organisms, other molecules foreign to host organisms, tumor-associated molecules and, in many diseases, normal or modified host molecules.

The biologic activity of immune complexes has been studied in detail. The size of the immune complexes in the circulation is an important parameter of toxicity. In general, larger (>19 S) immune complexes cause more tissue damage than smaller complexes. The size is related to the concentration and molar ratio of antibody and antigen, as well as to the avidity of the antibody for the antigen. Net charge of antigen and antibody also appears to be important in determining the pathophysiologic effect of the complexes. It has been shown that positively charged immune complexes tend to deposit in renal glomeruli, while complexes containing similar antigens with neutral charge tend to penetrate glomeruli slowly.

There are a number of reasons why circulating immune complexes are often found in high levels in immunological disorders. The antigens and the immune system responses may be near maximal, and the immune complexes formed may overwhelm the capacity of the systems for their removal. Alternatively, some mechanism may have compromised the efficiency of the immune complex removal system, or the nature of the immune complex constituents, antigens, antibodies and possibly complement fragments, resulting in inefficient removal.

ASSAY METHODS

The SCR-Ps of the present invention can be used to detect the presence of Clq-containing activators and Clq- containing immune complexes. Any SCR-P described herein can be used in the assay methods described herein, including all biologically active SCR-P variants, derivatives, fragments, analogs, mutants, truncated, modified, recombinant and endogenous forms of SCR-P. For example, an SCR-P can be the LHR-D region of CRl. Specifically an RCA protein, or a portion of the protein (e.g., a fragment) can be used. RCA proteins can be selected from the group consisting of CRl, Clr, Cis, CR2 , CD46, CD55, factor H and C4bBP . The portions of protein used in the assays described herein comprise a short consensus repeat (SCR) . A particularly specific SCR-P would include one binding site for Clq and one binding site for C3b. Clq-containing immune complexes detected in the assays of the present invention can also contain one or more additional components, for example complement proteins, including but not limited to C3b, iC3b, and C4b.

"Activators" are agents which directly bind and activate Clq, such as DNA, C-reactive protein complexes, serum amyloid P, urate crystals and gram negative and positive bacteria. Clq-containing activators are activators which affix, bind, or are associated with Clq. The term "antibody" is meant to include polyclonal antibodies, monoclonal antibodies (mAbs), chimeric antibodies, anti-idiotypic (anti-ID) antibodies to antibodies that can be labeled in soluble or bound form, as well as fragments, regions or derivatives thereof. Anti- SCR-P antibodies of the present invention are capable of binding the SCR-P, thereby inhibiting the binding of SCR-P to one or more other proteins, including a native ligand of SCR-P, such as Clq. Anti-Clq antibodies, and antibodies to a native ligand of SCR-P, of the present invention are capable of binding to Clq, or to a native ligand of SCR-P, respectively, such that inhibition of binding of the protein to SCR-P occurs. MAbs may be obtained by methods known to those skilled in the art.

Samples to be assayed include any biological sample that contains immune complexes. The sample can comprise processed erythrocytes or eluant from intact erythrocytes . The samples can also be a tissue sample, for example, a biopsy tissue selected from the group consisting of allograft, xenograft, brain or skin tissue. Samples can also include, for example, a biological fluid, e.g., a blood sample (whole blood, serum or plasma) , pleural fluid, peritoneal fluid, lymphatic fluid, joint fluid, and cerebral spinal fluid. In one embodiment, the sample is a sample comprising processed erythrocytes. In another preferred embodiment, the sample is a sample eluted by high salt and/or low pH(=3.0) or high pH buffer ( = 9.0) from intact erythrocytes .

The SCR-Ps of the present invention can be used in an assay to detect the presence of, or to quantitate the amount of, Clq-containing immune complex in a biological sample. The assay can be a radioimmunoassay or enzymatic assay, either sandwich-type or competitive type. Such assays are well-known to those of skill in the art. (See, for example, Baseman et al . , U.S. Patent No. 5,158,870, the teachings of which are incorporated herein by reference.)

For example, a biological sample can be obtained from the mammal (referred to herein as a test sample) and contacted with an SCR-P which specifically binds Clq- containing immune complexes under conditions suitable for binding of Clq to SCR-P. The optimal conditions for SCR-P binding can be determined by one of skill in the art. Suitable conditions for producing an SCR-P-bound test sample are well-known and some exemplary conditions are described herein. The amount of SCR-P bound Clq-containing immune complexes in the test sample can be measured using standard techniques and the amount of SCR-P bound Clq- containing immune complex in the test sample can be compared to the amount of SCR-P bound immune complex in a standard curve. A standard curve can be determined using patient sera or exogenously-added immune complexes to normal sera, by techniques well known in the art. An amount of SCR-P bound Clq-containing immune complex in the test sample greater than the amount of SCR-P bound Clq- containing immune complex in the standard curve is an indication that the mammal has an immune complex-associated disorder.

In one embodiment of the present invention, SCR-P can be used as a capture reagent in a sandwich assay, with labeled anti-Ig antibody or labeled Ig-reactive substance used to detect and quantitate the presence of Clq- containing immune complexes in the sample. The concentration of SCR-P used to coat the microtiter wells can be empirically determined. Suitable coating buffer solutions are known to those of skill in the art. The coating solution containing the optional amount of SCR-P is incubated in the wells for a time and at a temperature sufficient for SCR-P to coat the wells. Such conditions are known to those of skill in the art. After incubations, the coating solution is removed, the wells rinsed with rising buffer and blocking buffer is added to the wells. Blocking buffer is incubated in the wells for a time sufficient to decrease non-specific binding in the wells. The blocking buffer is then removed from the well.

A standard curve is prepared using known amounts of purified Clq, Clq fragment or Clq-containing immune complex, to cover a clinically relevant range of values. The buffer used to dilute Clq can be any suitable buffer. The sample to be tested for the presence of Clq-containing immune complex is also diluted so that the amount of Clq- containing immune complex contained in the test sample is detectable within the range of the standard curve . The optimal dilution of sample may be determined empirically using serial dilutions.

Unknown test sample and standards are added to the blocked wells and incubated for a time sufficient for the Clq-containing immune complex in the test sample and standard sample to bind to the SCR-P-coated wells. At. times, it may be desirable to use lower/higher temperatures for incubation, and the time of incubation can be increased or decreased respectively to obtain optimal reaction time. After suitable incubation, the reaction mixture of test or standard sample is removed from the wells, leaving only Clq or Clq-contammg immune complex, if present, specifically bound to the coated wells. The wells can be washed using a buffer such as described above. Bound sample is detected by adding a detectably labeled anti-Ig antibody or a detectably labeled Ig-reactive substance that binds to the immune complexes bound to the wells. The Ig- reactive substance can be, for example, protein A or protein G. Detectable labels are well-known to those of skill m the art and include, e.g., radioisotopes , biotm and fluorescem. The optimal amount of labeled antibody can be empirically determined. The labeled antibody is then incubated m the wells for a time sufficient for the labeled antibody to specifically bind to the SCR-P bound Clq-contammg immune complex bound to the wells. As described above, this time can vary depending on the temperature of incubation.

After this incubation step, one or more reagents can be added to aid m the detection (e.g., visualization) of the labeled Ig antibody or Ig-reactive substance. The optimal amounts of detection reagents added can be empirically determined, as well as the optimal conditions of incubation. The detection step can produce a color proportional to the amount of Clq-contammg immune complex m the sample. The optical density (O.D.) of this color product can be determined at a specific wavelength, and the O.D. of the unknown test sample is compared to the O.D. of the standards. Thus, a standard curve is prepared and the amount of Clq-contammg immune complex present the test sample is detected and/or quantitated by interpolation of the standard curve. The detection of the presence of Clq- contammg immune complex or of a threshold amount of Clq- containing immune complex (wherein a threshold amount is an amount greater than an amount of Clq-contammg immune complex present m a normal mammal) m the biological sample is indicative of immune complex-associated disorder or a compliment activation disorder, m the mammal. The determination of the amount of Clq-contammg immune complex in a biological sample can be a parameter by which to evaluate the severity of infection, or to evaluate the efficacy of a drug or a treatment protocol for immune complex-associated disorder. This invention also relates surface plasmon resonance assays described m the Example herein.

In another embodiment of the invention, the presence of the Clq-contammg immune complex can be detected by a competitive immunoassay procedure that comprises reacting SCR-P with a biological sample and with a competitive molecule that competes for binding to the SCR-P. In one embodiment, the detectably labeled competitive molecule is e.g., purified Clq, lmmunodetermmant-containing fragments (or biologically active fragments) of Clq, or Clq- containing immune complexes or other molecules that effectively compete with the Clq-contammg immune complexes for the SCR-P. As defined herein, biologically active Clq fragments are fragments that bind SCRs. The SCR-P or competitive molecule can be affixed with a detectable label, radiolabel , an indicator enzyme, biotin and fluorochrome or other labels known to those of skill m the art. In this embodiment, the amount of detectably labeled competitive molecule is inversely proportional to the amount of Clq-contammg immune complex m the sample. The assays of the present invention have advantages over other known assay techniques used to measure immune complexes. These techniques include: separation using polyethylene glycol; reducing the temperature of the solution containing the immune complexes (i.e., cold precipitation) ; binding of the immune complexes to complement protein Clq or to antibodies specific to the complement protein C3 and C3 degradation products; binding of the immune complexes to rheumatoid factors; binding to the bovine protein conglutmm; or binding of the immune complexes to platelets or to the Raj l lymphoblastoid cell line. There are disadvantages to each assay, including msensitivity, nonspeciflcity, inability to detect immune complexes of all sizes and of all immunoglobulin lsotypes and subisotypes, reliance on immune complexes containing complement protein, and interference by non-complexed immunoglobulins. Therefore, results gathered by any of the aforementioned assays may be considered to be of marginal value for diagnostic purposes, but may be used to support a diagnosis, to assess disease severity by correlating with amounts of immune complexes, or to monitor follow-up therapeutic treatment as suggested by Feldkamp m Clin . Chem . News, 3:5-6 (1987) . IMMUNE COMPLEX DISORDERS

The present invention also relates to the prevention, assay, diagnosis and treatment of immune complex-associated disorders. As used herein, "immune complex-associated disorder" refers to any disorder, disease, dyscrasia or condition which is related to immune complexes. It includes any disorder which involves abnormal, undesirable or inappropriate immune complex formation, deposition, degradation or removal, as a mediator and/or an indicator of the disorder, including, but not limited to, immune complex disorders (such as m autoimmune diseases, rheumatoid arthritis, systemic lupus erythematosus , proliferative nephritis, glomerulonephπtis, hemolytic anemia, and myasthenia gravis) ; diseases involving inappropriate or undesirable complement activation (such as hemodialysis disorders, hyperacute allograft and xenograft rejection, mterleukm-2 (IL-2) induced toxicity during IL- 2 therapy, hematologic malignancies such as AIDS) ; infections (such as lepromatous leprosy, AIDS, and sepsis); inflammation disorders (such as those present m autoimmune diseases, adult respiratory distress syndrome, Crohn ' s disease, thermal injury, burns, and frostbite); neurological disorders sucn as multiple sclerosis, Guillam Barre Syndrome, stroke, traumatic brain injury and Parkinson's disease); post-ischemic reperfusion conditions (such as myocardial infarct, balloon angioplasty, and post pump syndrome m cardiopulmonary bypass); autoantibody immune complex diseases such as ldiopathic thrombocytopenia purpura, systemic lupus erythematosus, myasthenia gravis, artnπtis, autoimmune hemolysis, glomerulonephritis, multiple sclerosis, Pemphigus vulgaris, cryoglobulmemia, and AIDS) , Epstein Barr virus associated diseases such as Sjogren's Syndrome, rheumatoid arthritis, Burkitt ' s lymphoma, Hodgkms αisease, virus (AIDS or EBV) associated B cell lymphoma, chronic fatigue syndrome, parasitic diseases such as Leishmania and lmmunosuppressed disease states (such a viral infection following allograft transplantation or AIDS) , undesirable primary antibody responses to immunotherapeutic agents such as xenogeneic monoclonal antibodies, used for immunosuppression (following allograft transplantation, for example) or for cancer therapy, vasculitis, synovitis, serositis, Raynaud's phenomenon, urticaria, poststreptococcal glomerulonephritis, vasculitis, polyarteritis (particularly when associated with hepatitis B antigenemia) , subacute bacterial endocarditis, viral infection, thrombotic conditions, chronic serum sickness, malignancies, dermatitis herpetiformis , and cystic fibrosis. Three of the causes of immune complex-associated disorder are chronic or persistent infections, autoimmunity, and extrinsic agents. Although many immune complexes are effectively and efficiently removed by phagocytic cells, small complexes m antigen-antibody equivalence can lead to immune complex diseases. Immune complex-associated disorders include, but are not limited to, immune complex-associated disorders caused by the following: CHRONIC INFECTIONS: Chronic infections that can result m immune complex-mediated disease include antigens from bacterial, parasitic, and viral microorganisms. Low-grade bacterial infections with certain streptococci can produce poststreptococcal glomerulonephritis or rheumatic fever. Staphylococcal infections can result m the subacute bacterial endocarditis or infected atπal ventricular shunts. Mycoplasma pneumoniae is the causative agent in primary atypical pneumonia, where the antibodies cross- react with erythrocytes to cause an acute hemolytic anemia as well as induce pulmonary infiltrates. Salmonella, Treponema pallidum, and Klebsiella have all been implicated m infection-associated glomerulonephritis. Plasmodium species, Toxoplasma gondii , filariasis, and schistosomiasis are also associated with immune complex glomerulonephritis. Chronic viral infections with hepatitis B, hepatitis C, measles, Epstem-Barr virus, retroviruses , and cytomegalovirus have also been associated with immune complex glomerulonephritis.

AUTOIMMUNITY: Immune complex-associated disease is a frequent complication of autoimmune disease. Tissue deposition m the blood vessel results m vasculitis or immune complex deposition m the glomeruli to precipitate glomerulonephritis. The variant of systemic lupus erythematosus (SLE) manifested by some lupus patients and most often implicated m renal disease is characterized by the presence of antibodies to double-stranded DNA. Some SLE patients have antibodies to SS-A, which have been associated with cutaneous manifestations of the disease, with congenital heart block in infants born to an asymptomatic mother, and with homozygous C2 and C4 complement component deficiencies. In immune complex-associated glomerulonephritis, antigen-antibody complexes in antibody excess are rapidly phagocytosed, but complexes in equivalence or slight antigen excess are deposited in the kidney. Deposition of immune complexes in the kidney and elsewhere initiates the cycle of complement activation, chemotaxis of PMNs, and release of enzymes from the granules of phagocytic cells that cause glomerular injury. Self-antigens such as those found in autoimmunity (e.g., DNA-anti-double standard DNA) are the primary mediators of glomerulonephritis. Exogenous or foreign antigens such as drugs or infectious agents also cause immune complex-mediated glomerulonephritis.

Rheumatoid arthritis is a systemic autoimmune disease frequently characterized by abnormal immune complexes. An antigenic stimulus is unknown but may result in the synthesis of an abnormal IgG by synovial lymphocytes. This abnormal IgG is recognized as foreign, which leads to increased production of rheumatoid factor (RF) . When RF reacts with the abnormal IgG within the joint space, complement is activated by the classical pathway. Chemotactic factors are generated, attracting PMNs and mononuclear phagocytes. Enzymes released into the joints by these phagocytic cells amplify the inflammatory response within the synovium. At later stages of RA, extra- articular manifestations occur, including vasculitis and pneumonitis .

EXTRINSIC AGENTS: Hypersensitivity pneumonitis is an immune complex disease of the lung induced by aerosol exposure to environmental antigens. Circulating IgG antibodies (and, much less frequently, IgM or IgA) are produced to those antigens. Upon repeated exposure, immune complexes are formed m the airways, resulting m an inflammatory process. The environmental antigens such as thermophilic actmomycetes (mold- infested hay, grains, musnroom compost, or contaminated water); Al ternaπa and other fungi, found m sawdust and tree bark; and various isocyanates .

The removal of immune complexes from the circulation may reduce many of the clinical problems associated with autoimmune and infectious diseases and cancer. For example, it is beneficial to monitor circulating immune complex levels m blood and to specifically bind and remove circulating immune complexes which may otherwise compromise immune system function or lead to acute or chronic inflammation.

COMPLEMENT ACTIVATION DISORDERS

Complement activation can account for substantial tissue damage m a wide variety of autoimmune/immune complex mediated syndromes such as systemic lupus erythematosus, rheumatoid arthritis, hemolytic anemias, myasthenia gravis and others involving inappropriate or undesirable complement activation, such as hemodialysis disorders, allograft and xenograft rejection, mterleukm-2 (IL-2) induced toxicity during IL-2 therapy, and hematologic malignancies such as AIDS. Inhibition of the complement system is likely to be a desirable therapeutic intervention m these cases. In some instances, specific inhibition of the classical pathway alone could be preferred since long-term inhibition of the alternative pathway could lead to grave side effects. Inhibition of complement activation could also be desirable cases that involve tissue damage brought about by vascular injury such as myocardial infarction, cerebral vascular accidents or acute shock lung syndrome. In these cases, the complement system may contribute to the destruction of partially damaged tissue as m reperfusion injury. Highly stringent inhibition of complement for relatively brief periods might be preferred m these instances and soluble SCR-Ps designed for higher potency could be especially useful. Complement inhibition can also be important m the prevention of xenograft rejection. It is possible that organs derived from animals transgenic for human SCR-Ps may be protected from complement -mediated hyperacute rejection by the expression of transgenic SCR-Ps on the cell surfaces of the xenograft. Animals transgenic for SCR-Ps designed for higher potency may provide more successful xenografts. Soluble SCR-Ps may also prove useful m protecting the transplant the recipient. SCR-Ps which exhibit the desired activity may have therapeutic uses in the inhibition of complement by their ability to act as a factor I cofactor, promoting the irreversible inactivation of complement components C3b or C4b (Fearon, D. T., Proc . Na tl . Acad . Sci . U. S . A . , 76:5867 (1979); lida, K. and Nussenzweig, V., J. Exp . Med. 153:1138 (1981) ) , and/or by the ability to inhibit the alternative or classical C3 or C5 convertases.

In a specific embodiment of the invention, an expression vector can be constructed to encode an SCR-P molecule which lacks the transmembrane region (e.g., by deleting the carboxyl-terminal to the arginine encoded by the most C-terminal SCR) , resulting in the production of a soluble SCR-P fragment. In one embodiment, such a fragment can have the ability to bind Clq, with or without C3b and/or C4b. Although soluble SCR-P may impede the clearance of immune complex disease, i.e. make it worse, it may be useful when given acutely or prophylactically to block complement activation at the Cl step as well as at the C3 and C5 steps. In such an embodiment, the SCR-P can be valuable in the treatment of disorders which involve undesirable or inappropriate complement activity (e.g., shock lung, tissue damage due to burn or ischemic heart conditions, autoimmune disorders, and inflammatory conditions) . Methods of the present invention can be used to inhibit complement activation can also reduce cellular adhesion and thrombus formation brought on by exposure of synthetic biomaterial to blood, for example, during the use of certain medical devices which require this exposure. (Ward, C . and P.G. Kalman, Medical Progress , 15:63-75 (1989) . Inhibition of complement activation could also be used to prevent or treat decompression sickness. (Ward, et al . , supra , 1989). This complement -mediated illness may be caused by activation of complement by air bubbles.

After C3b has covalently attached to particles and soluble immune complexes, the mactivation of C3b by proteolytic processing into ιC3b and C3dg has three biologic consequences: preventing excessive activation of the complement system via the amplification pathway, blocking C5 activation, and the formation of ligands that can engage receptors . The ιC3b fragment cannot bind factor B so that conversion to this state blocks additional complement activation v a the alternative pathway amplification loop. However, ιC3b can be bound by CRl and CR3 , the two complement receptors that mediate phagocytosis by myelomonocytic cells. Therefore, the primary biologic consequence of C3b to ιC3b conversion is cessation of complement activation without interference with CRl- and CR3-medιated clearance of the C3-coated complex. Assays of the present invention can be used, therefore, as a diagnostic for the detection of (e.g., an individual such as a human) , immune complex-associated disorders or complement activation-associated disorders m a mammal. The assay can also be used to screen a pharmacological agent or drug for its effectiveness at preventing, treating, or alleviating the symptoms of such disorders . REMOVAL OF IMMUNE COMPLEXES

The SCR-Ps, SCR-P fragments, analogs, and derivatives of this invention can also be used for excorporeal or ex vivo removal of immune complexes from fluids or tissue or cell extracts. In one embodiment, immune complexes are removed from blood and blood plasma by binding to SCR-Ps for the analysis of the antigen and antibody constituents of immune complexes. In another embodiment, SCR-Ps, SCR-P fragments, analogs and derivatives are advantageously used m extracorporeal devices, which are known m the art (see, for example, Seminars in Hematology, 26 (2 Suppl . 1 ) (1989) ) . Patient blood or other body fluid is exposed to the SCR-P resulting m partial or complete removal of circulating Clq (free or m immune complexes or bound to an activator) , following which the fluid is returned to the body. This lmmunoadsorption can be implemented m a continuous flow arrangement, with or without interposing a cell centrifugation step. See, for example, Terman, et al . , J. Immunol . 117:1971-1975 (1976). For example, SCR-P can be coated on a solid support surface which is encased online m an extracorporeal device through which whole blood or plasma can be circulated dynamically so that the immune complexes contained therein are bound and thereby removed from the plasma or blood. This process could also be used with conventional plasmapheresis or hemadialysis techniques. After removal of immune complexes, fluids can be returned to the patient (i.e., remfused) negating the need for fluid replacement therapy. The solid surface and encasing means of the device may be made of any biocompatible material. For instance, the solid surface may be a membranous surface, agarose-based beads or hollow fibers coated with SCR-P. The extracorporeal device may be a column packed with beads, a hollow fiber membrane encased m a cylinder like those used m renal dialysis, a microtiter plate containing wells, or any suitable surface, coated with SCR-P. The device may also include appropriate tubing for connecting it to a patient and a pump to aid the passage of the fluid through the device and back into the patient and to prevent air from entering the system. The device must be sterilized for therapeutic use, and sterilization may be accomplished m conventional ways such as purging with ethylene oxide or by irradiating the device.

The binding and removal of circulating immune complexes, or enhancing their clearance from the circulation, may result m a clinical improvement of individuals with circulating immune complexes and improve the effectiveness of other treatments for immune complex- associated disorders. For example, it may prevent lesions in organs and organ systems associated with immune complexes that occur m infectious diseases, arthritic and autoimmune diseases, as well as result m the clinical improvement of arthritic and autoimmune diseases.

METHODS OF TREATMENT

This invention also pertains to methods of treating a complement activation-associated disorder (and possibly an lmmune complex-associated disorder) m an individual comprising administering an SCR-P or a fragment comprising a homologous sequence of the SCR binding domain of Clq (e.g., an SCR antagonist with a special affinity for the SCRs of Clr and Cis) to the individual. SCR-P

"antagonists" are defined herein as molecules or proteins which decrease or inhibit (including competitive inhibition) one or more biological activities of Clq.

SCR-Ps, or SCR-P agonists (SCR-P agonists are defined herein as molecules or proteins which increase or activate one or more biological properties of SCR-P) and antagonists of the present invention can be administered either as individual therapeutic agents or a composition with other therapeutic agents. SCR-P agonists and antagonists, together, are considered modulators. They can be administered alone, but are generally administered in a composition with a physiologically compatible pharmaceutical carrier selected on the basis of the chosen route of administration and standard pharmaceutical practice.

In this context, drugs designed on the basis of the SCR-P protein sequence and intended for use m humans include small non-peptide molecules, peptides or proteins related to SCR-Ps or designed to modify the function of SCR-Ps, or DNA or RNA sequences encoding proteins or peptides related to an SCR-P or designed to modify the function of SCR-P.

The dosage administered (e.g., the effective amount) will, of course, vary depending upon known factors such as the pharmacodynamic characteristics of the particular agent, and its mode and route of administration; age, health, and weight of the recipient; nature and extent of symptoms, kind of concurrent treatment, frequency of treatment, and the effect desired.

It can be administered one to several times per day, depending on the mode of administration. Effective doses can be determined by those of skill the art. An effective dose of SCR-P is an amount sufficient to relieve the individual of the symptoms of an immune complex- associated disorder or a complement activation-associated disorder.

Methods of introduction of the agent at the site of treatment include, but are not limited to, mtradermal, intramuscular, mtraperitoneal, intravenous, subcutaneous, oral, mtranasal, gene therapy, cellular implantation or particle bombardment. Other suitable methods include or employ biodegradable devices and slow release polymeric devices . Because proteins are subject to being digested when administered orally, parenteral administration, e.g., intravenous, subcutaneous, or intramuscular, would ordinarily be used to optimize absorption.

For parenteral administration, particularly suitable are injectable, sterile solutions, preferably oily or aqueous solutions, as well as suspensions, emulsions, or implants, including suppositories. The SCR-P protein, or the Clq or SCR-P agonist or Clq antagonist, can be formulated as a solution, suspension, emulsion or lyopmlized powder m association with a pharmaceutically acceptable parenteral vehicle. Examples of such vehicles are water, saline, Ringer's solution, dextrose solution, and 5% human serum albumin. Liposomes and nonaqueous vehicles such as fixed oils can also be used. The vehicle or lyophilized powder can contain additives that maintain isotonicity (e.g., sodium chloride, mannitol) and chemical stability (e.g., buffers and preservatives). The formulation is sterilized by commonly used techniques. Suitable pharmaceutical carriers are described m the most recent edition of Remington's Pharmaceutical Sciences, A. Osol, a standard reference text m this field of art. Ampules are convenient unit dosages . Formulations for transdermal or transmucosal administration generally include penetrants such as fusidic acid or bile salts combination with detergents or surface-active agents. The formulation can then be manufactured as aerosols, suppositories, or patches.

Oral agents may be administered if formulated as to be protected from digestive enzymes. If administered orally, the SCR-P will be administered m a therapeutic composition which may also include an appropriate carrier (e.g., a physiologically compatible carrier) , a flavoring agent and a sweetener. Suitable pharmaceutical carriers include, but are not limited to water, salt solutions, alcohols, polyethylene glycols, gelatin, carbohydrates such as lactose, amylose or starch, magnesium stearate, talc, silicic acid, viscous paraffin, fatty acid esters, hydroxymethylcellulose, polyvinyl pyrolidone, etc. The pharmaceutical preparations can be sterilized and, if desired, mixed with auxiliary agents, e.g., lubricants, preservatives, stabilizers, wetting agents, emulsifiers, salts for influencing osmotic pressure, buffers, coloring, and/or aromatic substances and the like which do not deleteriously react with the active compounds. They can also be combined where desired with other active agents, e.g., enzyme inhibitors, to reduce metabolic degradation. CRl has never been identified as a ClqR before. This may be due to the low density of CRl expression on resting cells (Fearon et al . , Ann . Rev. Immunol . 1:243-250 (1983)). Further, the binding constants derived herein suggest that insolubilized Clq is a poor affinity ligand for CRl under normal binding conditions and the presence of detergent might further impair binding. Moreover, as described herein, Clq has at least two binding sites for CRl. It is likely that the multivalency of Clq binding to CRl resides in its six identical stems, because this is where Clr₂ Cls₂ binds and Clr₂ Cls₂ binding is known to interfere with ClqR binding. CRl, on the other hand, has one functional binding site for Clq. Thus, with equimolar concentrations of receptor and ligand, Clq (multivalent) binding would be much more robust to immobilized CRl, than would CRl (univalent) binding to immobilized Clq. This was shown BIAcore binding data (Example 6) . Thus, when Clq is immobilized as an affinity reagent for solubilized CRl in a membrane lysate, the concentration of CRl would be much below the Keq needed for binding to the Clq. In contrast, CR1 is expressed m clusters on erythrocyte and granulocyte plasma membranes and therefore would mimic binding conditions. Thus, the binding studies described herein, defining the multivalency of Clq and the functional univalency of CRl, provide an explanation as to why CRl was not previously recognized as the ClqR.

The relationship of CRl to the other ClqRs previously defined, namely calreticulm and the 100 kDa transmembrane protein, is not known. It is not likely that CRl is the only ClqR since Clq does bind and activate some cells, e.g. endothelial cells (Zhang et al . , Tissue Cell , 18:13-18 (1986); Lozada et al . , Proc . Natl . Acad . Sci . USA, 92:8378- 82 (1995) ) , which do not express detectable CRl (Shaw et al . , BP1.3 : Leucocyte Differentiation Antigen Database (1995)). Other SCR-Ps may also subserve Clq binding. There are important biological implications for Clq/CRl binding. CRl on erythrocytes is critical for the transport of immune complexes to the liver and spleen where they are cleared. Thus, CRl humans may be solely responsible for clearing immune complexes opsonized by complement, whether Clq and/or C3b/C4b. That Clq might participate m the clearance of immune complexes may explain several clinical observations. It is well recognized that deficiency of an early component of the classical pathway predisposes to autoimmune diseases, typically systemic lupus erythematosis, or SLE. The SLE associated with C2 and C4 deficiencies is frequently associated with serositis and cytopenias, but rarely causes serious organ damage. In contrast, Clq deficiency is very severe and often includes central nervous system disease and glomerulonephritis. Deficiency of C2 or C4 would result m defective opsonization by C4 and C3 , however immune complexes might still be effectively opsonized by Clq. In Clq deficiency, neither Clq, C3b nor C4b would effectively bind to immune complexes, and the complexes might then be deposited m vital organs such as the kidney, rather than the normal targeting to liver and spleen for clearance. That Clq has a role m innate immunity and protection from autoantibody disease is clearly illustrated by Clq- deficient humans: 31 of a total of 33 deficient patients suffer chronic infections and very severe autoimmune disease with a prominent photosensitive skin rash and often central nervous system disease and glomerulonephritis . The syndrome is more severe than the deficiency of C4 or C2 , indicating that the defect cannot be explained just by the inability to activate the remainder of the complement pathway. Thus, there is evidence that Clq as a ligand for the ClqR is involved m host defense and protection against autoimmunity.

Clq can directly opsonize pathogens, which facilitates their clearance by the host. Opsonization can potentially occur by two pathways innate immunity: first, Clq will bind to those antigen- ntibody complexes formed with naturally occurring IgM; second, Clq has the capacity to bind directly to many pathogenic bacteria, including group B streptococci and Salmonella species, and Clq can bind directly to the lipid A of soluble endotoxin. With respect to preventing autoimmunity, there are multiple mechanisms by which Clq and the ClqR might function. First, as described herein, Clq can mediate the binding of aggregated IgG to erythrocytes (Example 7) , and this process would function for the clearance of immune complexes, in the same manner that C3b opsonized immune complexes are cleared by erythrocytes. Second, Clq alone, and C-reactive protein (CRP) , which in turn binds Clq, complex with damaged cell membranes. Blebs on damaged keratinocytes contain antigens which are targets for autoantibody formation, and thus, deficiency of Clq would prevent ClqR-bearing cells from clearing the subcellular autoantigens and in that way predispose to autoantibody formation. Finally, there is evidence for Clq-dependent clearance of apoptotic cells in vivo .

The predisposition of Clq deficiency to lead to autoimmune disease has also received experimental confirmation: Clq knockout mice, like Clq deficient patients, develop a lethal crescentic glomerulonephritis. CRl also functions cooperatively with Ig Fc receptors in the phagocytosis of opsonized erythrocytes and microorganisms. In situations where Cl activation leads to the deposition of C4b and C3b, Clq would also be expected to be present. Finally, although Clq has many biologic effects, CRl does not apparently signal by itself. Thus, if CRl is important in signaling, it is likely that other molecules are recruited.

There are multiple inhibitor molecules to control the activation of the complement pathways and many of them act at the step of C3 , since the complement activation pathways converge and are amplified at this step. A common molecular motif of the C3 binding inhibitors is the short consensus repeat (SCR) . CRl usually contains 30 SCR. The SCR of CRl are organized into four long homologous repeats, which are designated as LHR-A, B, C, and D, where LHR-A is farthest from the plasma membrane. The binding site for C4b resides m LHR-A, while there are two binding sites for C3b, one m LHR-B and one m LHR-C. No known binding function had been described for LHR-D. Cl subcomponents Clr and Cis each contain SCR, and both Clr and Cis bind to Clq. It is reasonable to believe that the binding site on CRl for Clq would show the highest homology for the SCR of Clr and Cis. A computer generated alignment indicated that the LHR-D region of CRl was the most homologous to the SCR sequences of Clr and Cis, which makes it is most likely that Clq binds to LHR-D. Thus, within the CRl molecule there are binding domains for all the complement fragments involved m the clearance of immune complexes, namely Clq, C3b and C4b. (Fig. 7) As described herein, the ability of two other SCR containing proteins to bind to Clq: factor H, which is composed of 20 SCR, and C4 binding protein (C4bp) , which usually contains 56 SCR, was also tested. Both factor H and C4bp bound to msolubilized Clq. C3b dimers compete for ¹²⁵I-Clq binding, but not completely. This is consistent with the inhibition being secondary to steπc hmderance . This indicates that (C3b)₂ and Clq are binding to adjacent, but not identical, sites on CRl . Also, as demonstrated in Example 7, Cl will bind to erythrocytes, which are cells which lack integrins and Fc receptors, specifically, the collagen fragment of Clq will bind to erythrocytes. In another assay using radiolabeled heat aggregated IgG (HAG) it was demonstrated Clq-dependent binding of HAG to erythrocytes (Example 7) . To confirm the role of CRl in this binding, we found that anti-CRl antibody is able to block 80% of the binding of Clq-HAG to erythrocytes. Thus, we are now able to confirm that CRl expressed on normal cells, as well as CRl expressed on transfected cells, can bind Clq.

In Example 8, described herein, the binding of SPA and MBL to Clq receptor molecules was demonstrated. It has been claimed that Clq receptor molecules are also receptors for two other collectins, namely surfactant protein A (SPA) and mannose binding lectin (MBL) . Calreticulin, the p60 kDa intracellular Clq-binding protein also binds SPA and MBL, but calreticulin is not a membrane receptor. SPA-coated plates allowed monocytes to ingest IgG coated erythrocytes or IgM-C4b coated erythrocytes as efficiently as Clq coated plates. Although an mAb directed against the 126 kDa receptor blocked phagocytosis by both Clq and MBL, there was no evidence that Clq or MBL exhibited competitive binding, or that they bound directly to the 126 kDa membrane protein. Thus, there are no definitive data that SPA, MBL and Clq bind to the same membrane receptor.

Example 9 described binding assays of other SCR-Ps to Clq. PNH erythrocytes (>95% DAF deficient) bind Clq normally, and thus DAF must have a negligible role in Clq bmdmg. CD46 appears not to be involved, because CD46 is highly expressed on K562 cells, and the background Clq binding to these cells is low. As described m Example 9, using purified proteins, Clq bound well to both C4 binding protein (C4bp) and to factor H (Fig. 16) . Factor H bound less well than C4bp, possibly pecause factor H is monovalent, while C4bp is multivalent. The identification of one site on Clq that interacted with the serine protemases Clr and Cis, and the cellular receptor for Clq (CRl) is useful for the design of peptide inhibitors of Cl activation or Clq reactivity with cells.

Thus, as described herein, the identification of the ClqR clarifies role of Clq m innate immunity and sheds light on the particularly severe autoimmune disease seen m Clq-deficient mice and humans. Furthermore, knowing that Clq binds CRl and Clr₂Cls₂, both of which contain multiple SCR domains, suggests that other SCR-contammg proteins might bind Clq and act as ClqR or modify Clq's hemolytic activity. To define the SCR-Clq binding sites and explore the biological implications of this binding, a number of studies can be performed.

For example, as described m Example 10, the binding sites on the collagen domain of Clq for CRl and other relevant SCR-contammg proteins can be identified. When Clq is bound to its tetramer of catalytic subunits (Clr₂ Cls₂) , it cannot interact with the putative C IqR to stimulate cellular responses. The capacity of CRl to compete with Clr₂ Cls₂ for binding to Clq can be tested. If the results are positive, it would indicate that CRl can regulate the classical pathway at a new site m the complement cascade m addition to its known site at C3/C5 activation. Two methods that have been employed by others to define Clq domains can be used, namely: chemically modifying charged ammo acid residues on Clq, which are important m Clr₂Cls₂ binding, and developing a panel of anti-Clq monoclonal antibodies, which can be screened for their ability to block CRl binding. Finally, other membrane and serum proteins, which like CRl contain SCR, can be assessed as receptors for Clq and as inhibitors of Cl activity.

In addition, as described m Example 11, the binding site on CRl for Clq can be localized and it can be determined whether CRl is a receptor for other collectins. The extracellular domain of the common allotype of CRl is comprised exclusively of 30 homologous short consensus repeats (SCR) . The binding site for Clq on CRl can be localized by testing monoclonal anti-CRl antibodies for their ability to block Clq binding. In addition, deletional constructs of CRl can be genetically engineered and expressed K562 cells for evaluation of Clq binding. The exact binding sιte(s) will be defined by site direct mutagenesis. The other human plasma collectins, namely surfactant protein A (SPA) and mannose binding lectm (MBL) , which, like Clq, function m innate immunity as opsonms, bind to ClqR. Ligated SPA and MBL binding to CRl can be assessed, and the respective binding site of MBL on CRl can be mapped. Finally, identification of the binding sites on CRl for Clq and MBL potentially SPA and ligated, will make it possible to prepare a low molecular weight, polyvalent SCR-containing binding peptide that can be used to block the ligand binding- and complement activating-functions of Clq, and possibly SPA and MBL. As described in Example 12 , to determine the role of cellular CRl and/or other SCR-containing membrane proteins as Clq receptors, the primacy of CRl as the ClqR of circulating phagocytes and B cells in both Clq binding and Clq-mediated responses can be tested. If Clq binds to other membrane SCR-containing proteins, the focus can be broadened to include these ClqRs . After priming or CRl ligation with Clq, the presence of CRl-associated protein(s), in particular the 126 Mr transmembrane molecule involved with Clq function, can be tested. The role of CRl in mediating the biologic responses induced by Clq binding can, thus, be determined, and the functional involvement of the Clq-binding proteins can be put into context.

Having now generally described the invention, the same will be further understood by reference to certain specific examples which are included herein for purposes of illustration only and are not intended to be limiting in any way .

EXAMPLE 1: PREPARATION OF K562 CELLS WITH CELL SURFACE CRl

The plasmid pBSHyg, which directs the expression of hygromycin resistance, was prepared by ligation of the 2.0 kb Hind III-Nru I fragment from REP3 (Groger et al . , Gene, 81:285-294 (1989)) into the Hind III-Hinc II sites of bluescript KS- (Strategene, La Jolla, CA) . The plasmid paABCD directs the expression of the F allotype of human CRl (Klickstein et al . , J. Exp . Med . , 168:1699-1717 (1988)). Human erythroleukemia cell line K562 cells were electroporated (250V, 960μF) with 200 ng pBSHyg linearized with Xmn I with or without 20 μg of paABCD (which directs the expression of the common allotype of CRl) linearized with Sfi I. Transfectants were selected by culture in RPMI with 20% FCS supplemented with hygromycin at 200 μg/ml for 2 weeks, then transferred to RPMI with 10% FCS. K562 cells transfected with pBSHyg alone were termed K-Hyg and those transfected with paABCD and pBSHyg were termed K-CRl. The K-CRl cells were immunopanned (Wysocki et al . , Proc . Natl . Acad . Sci . USA, 75:2844-48 (1978)) on immobilized YZ-1 monoclonal anti-CRl antibody (YZ-1) (Changelian et al . , J^".

Immunol . , 134:1851-1858 (1985)) which binds to sites in the A, B, and C long homologous repeats near the amino terminus of CRl (Klickstein et al . , supra , 1988) to select a uniformly positive population of cells. The CRl expression on the control K-Hyg and K-CRl transfectants was assessed using anti-CRl mAb (YZ-1), aridin-FITC and FACS analysis. (Figures IA - ID) Once during the period of experimentation the K-CRl transfectants were reselected by panning using YZ-1 to enhance CRl expression.

CHARACTERIZATION OF TRANSFECTED CELLS: The K-CRl and the control K-Hyg transfectants were characterized for CRl expression using control mAb, then FITC-goat anti -mouse IgG (Figs. IA and IC) or anti-CRl mAb YZ-1, followed by FITC- goat anti-mouse IgG (Figs. IB and ID) . The cells were subsequently fixed and analyzed by FACS. There was no positive staining of the K-Hyg cells, compared with second antibody alone (MFC of FITC-second antibody alone = 9.25; MFC of anti-CRl plus FITC-second antibody = 8.69) (Figs. IA and IC) . In the same analysis, K-CRl demonstrated a broad histogram (a diffuse pattern) of positive staining (Fig. ID) compared with its control (Fig. IB). Because the K-CRl cells were selected by panning but not cloned, their heterogenous expression of CRl would be expected. The recombinant CRl was shown to be intact by immunoprecipitation and western blot analysis of NP-40 lysates of Hyg and K-CRl cells using rabbit anti-CRl (Yoon et al . , J^". Immunol . , 134:3332-38 (1985)) as a probe. The K-CRl cells have a unique 200k Mr band corresponding to intact CRl. The K-CRl cells bound to C3b immobilized on plastic, while K-Hyg cells did not (Fig. IE) . Adherent cells were counted.

EXAMPLE 2 : PREPARATION OF LABELED C1Q

BUFFERS/REAGENTS: PBS, 0.15 M NaCl, 0.05 M sodium/potassium phosphate, pH 7.4. 1,3 diaminopropane (Sigma Chemical, St. Louis, MO), dinonyl phthalate (Arcos, Ghent, Belgium) , dibutyl phthalate (Sigma) , were purchased as noted. Recombinant soluble human CRl (rsCRl) was provided by Drs. Una Ryan and Henry Marsh (T-Cell Sciences, Needham, MA) . COMPLEMENT Clq: Clq for biotmylation (as used m Fig. 2A) was isolated from human serum by a procedure using BioRex 70, as originally described m Tenner et al . , J^". Immunol . , 127:648-53 (1981b) and modified by Jack et al . , J. Immunol . , 153:262-9 (1994). Each batch of Clq^bl° was tested for functional activity (Tenner et al . , 1981b). Because of the propensity of the native Clq made by the BioRex method to aggregate, an alternative isolation method for Clq, utilizing fractional euglooulm precipitation and gel permeation chromatography, was devised to prepare Clq for radiolabelmg. In brief, 30 ml fresh serum, 5 mM EDTA was dialyzed against a euglobulm precipitation buffer (20 mM morpholme ethane sulfonic acid, 5 mM benzamidme, 0.5 mM EDTA, pH 6.5.) m tubing for 16 hours at 4°C. The precipitate was collected by centrifugation (8,000 x g, 10 mm), washed twice in 5 mM propanediamme, 0.05 mM EDTA, pH 8.8 (PDE buffer) (Liberti et al . , J. Immunol . Meth . 40:243- 245 (1981) ) , then dissolved m PDE buffer supplemented with 300 mM NaCl (PDE/NaCl) , 0.5 mM PMSF. The mixture was centrifuged at 10,000 x g for 5 mm to remove undissolved precipitate. The supernatant containing Clq was aliquoted (0.45 ml) to each of 3 microfuge tubes. One ml of cold water was added to each tube and after 10 mm at 4°C the resulting precipitates were collected by centrifugation (10,000 x g, 5 mm) . The precipitates each tube were dissolved in 75 μl PDE/NaCl buffer, and then 85 μl 2 x PBS were added. This material was recentπfuged (15,000 x g, 5 mm) and 150 μl was applied to a TSK G4000SW_XL column (Supelco, Beliefonte, PA) equilibrated m 2x PBS, 0.5 mM EDTA at a flow rate of 0.5 ml per mm. Clq was eluted as a peaK with a retention time of 18.6 mm (equivalent to a MW of 462 kDa) and the peak was collected by hand. Analysis of the Clq revealed the distinct a, b, and c chains by SDS PAGE with silver staining. The specific activity of the Clq was 400 hemolytic units (Z) per μg protein using the assay based on the BioRex drop through fraction of human serum (Tenner et al . , J. Immunol . 133:886-891, 1981b). Protein was assayed by the micro bicmchonmic acid (BCA) method (Pierce) . Since BSA and Clq have almost equivalent extinction coefficients (0.667 and 0.682) (Reid et al . , Biochem . J. , 130:749-763 (1972)), Clq concentrations were calculated from a BSA standard curve.

BIOTINYLATED Clq: In a typical biotmylation reaction, 840 μg of Clq was reacted with 4.3 μg NHS-biotm (Pierce

Chemicals, Rockville, IL) m 1 ml of PBS for 30 mm at room temperature with intermittent agitation. The reaction was stopped by the addition of concentrated ethanolamme to give a final concentration of 0.1M. The reaction mixture containing biotinylated Clq (Clq^bl°) was subsequently dialyzed against 20 mM KCL, 10 mM Tris/HCl pH 7.4 and the protein assayed. The final bιotm:Clq ratio could not be determined using the reagents provided with the biotmylation kit, but it was necessary to biot ylate lightly m order to avoid aggregation. Each batch of Clq^bl° was tested for functional activity (Tenner et al . , supra) .

Biotmylatmg with NHS-biotm at 0.6 μg/ml and 30 μg/ml yielded two preparations of Clq^bl° with near normal CR1 binding, but with 23% and 90% losses m hemolytic activity, respectively.

RADIOIODINATED Clq: Glucose oxidase (Sigma) and lactoperoxidase (Sigma) were separately coupled to beads of cross linked bis-acrylamide/azlactone copolymer beads (3M Emphaze™, Pierce) according to the manufacturer's instructions. For both enzymes the coupling ratio was 1 mg of protein per hydrated equivalent of 24 mg of dried beads. The optimal ratio (1/4) of coupled glucose oxidase to coupled lactoperoxidase was determined m a preliminary experiment by combining the beads m different ratios and measuring the resultant enzymatic activity using 0.1% D- glucose tetramethylbenzidme solution (Kirkegaard & Perry, Gaithersburg, MD) as the substrates. The development of blue color was followed by eye.

At the time of radiolabelmg, glucose oxidase beads (3μl of a slurry) and lactoperoxidase beads (12 μl of a slurry) were washed into PBS. Three μl of sodιum-¹²Iodme (carrier free, 100 mCi/ml, New England Nuclear, Boston, MA) was added to the bead pellet, followed by 70 μl of Clq (400 μg/ml m 2xPBS) , 70 μl dHOH, and 10 μl glucose solution (100 μg/ml PBS) . The reaction proceeded for 20 mm at room temperature with intermittent shaking. The reaction supernatant was applied to a PD-10 gel filtration column (Pharmacia), which had been equilibrated m PBS, 0.1% gelatin. The radiolabeled Clq was pooled and characterized. 98% of CPM of the ¹²⁵I-Clq were precipitable with 10% TCA. ¹²⁵I-Clq was quantified by a sandwich ELISA usmg anti-Clq murine mAb (Quidel, San Diego, CA) as the capture antibody and goat ant1 -human Clq (IncStar, Stillwater, MD) as the indicator antibody. The reaction was developed with horseradish peroxidase conjugated rabbit anti-goat IgG (IncStar) and tetramethylbenzidme substrate. The color reaction was stopped by the addition of H₃P0₄ and the OD 450 was quantified using an ELISA plate reader (Molecular Devices, Menlo Park, CA) . Clq of a known protein concentration was used as a standard. Multiple lots of Clq were lodmated with specific activity ranging from 5- 7.5 x 10⁵ CPM per μg Clq as quantified by ELISA; and the functional activity was 400 hemolytic units (Z) per μg.

EXAMPLE 3: PREPARATION OF C3b DIMERS (C3b)₂

C3 purified from fresh human plasma by standard methods was treated with trypsin to produce C3b (Fearon, Ann . Rev. Immunol . , 1:243-50 (1983)). The trypsin was inactivated by addition of dnsopropylfluorophosphate and the C3b purified by chromatography on Sephacryl S300 m PBS. The fractions containing C3b were pooled, concentrated to 1.4 mg/ml (Centriprep, Amicon, Beverly, MA) and stored at 4°C for three weeks to allow formation of dimers via oxidation of the free sulfhydryl group (Arnaout et al . , supra, 1981). Dimeπc C3b was separated from monomeric C3b by gel filtration on Sepharose CL2B m PBS (Pharmacia LKP Biotechnology, Piscataway, NJ) and the expected Mr was confirmed by SDS-PAGE. Peak fractions were pooled, aliquoted and stored at -80°C.

EXAMPLE 4 : ANALYSIS OF Clq BINDING TO HUMAN PMN AND TO TRANSFECTED CELLS

FACS ANALYSIS: A FACStar (Becton Dickinson, San Jose, CA) was used and 10⁴ cells were analyzed for each variable. Transfected cells were assessed for CRl expression using the murine mAb YZ-1 and FITC-goat anti-mouse IgG (Tago- Biosource, Camarillo, CA) .

Clq BIO BINDING TO HUMAN PMN

Biotinylated Clq (Clq^bl°) bound to isolated human neutrophils (PMN) and this binding was upregulated by incubation of the cells in buffer at 37°C. The addition of either PMA or FMLP caused further upregulation of Clq^ binding (Fig 9) . The augmented Clq^bl° binding followed the same kinetics as the enhanced CRl expression described when PMN are incubated at 37°C, with the additional upregulation of CRl and CR3 expression, which occur when PMN are stimulated with PMA or FMLP. Clq^bl° binding was unimodal in the buffer-37°C-stimulated and FMLP-stimulated PMN (Fig 10) . Because it had been reported by others that FMLP could not upregulate Clq binding, and that Clq bound to only a subpopulation of human PMN, the experiments were repeated using different Clq^bl° preparations and PMN from different donors: unimodal and upregulatable Clq binding to PMN from normal donors was always noted. Furthermore, taxol inhibited the upregulation of Clq^D1° binding, just as it inhibited CRl and CR3 expression. In summary, the striking similarities m the control over Clq binding and CRl and CR3 expression suggested that the ClqR might be either CRl or CR3 , or at least be stored m the same vesicular compartment as CRl and some of CR3. Subsequent studies, described herein, showed that CRl, and not CR3 , were responsible for Clq binding.

Clq^b-° BINDING TO TRANSFECTED CELLS Transfected K-Hyg and K-CRl cells were reacted with increasing amounts of Clq^C10. Binding was performed m HBSS without Ca^+T and Mg^τ* (HBSS=) which was diluted 1/2 with glucose-BSA solution (5% glucose, 0.2% BSA). The cells were first reacted with avidm/biotm blocking reagents following the manufacturer's instructions (Vector Laboratories, Burlingame, CA) for 10 mm at room temperature to block non-specific avidm and biotm binding sites. Subsequently, samples of transfected K-Hyg and K- CR1 (5 x 10⁵) cells were aliquoted to tubes and buffer (HBSS⁼ which was diluted with an equal volume of 5% glucose containing 0.2% BSA) ± increasing concentrations of Clq⁰¹⁰ were added (final vol.=135μl) for an incubation of 30 mm at 37°C. After appropriate washes, FITC-avidm (Vector) diluted 1/250 was added for 25 mm at room temperature. After a wash the cells were fixed m 1% paraformaldehyde/PBS and analyzed by FACS.

A plot of the mean fluorescent channel (MFC) versus Clq^D1° input indicates specific binding of Clq^bl° (Fig. 2A) . MFC of the single dominant peak was determined based on analysis of 20,000 cells. This experiment was repeated two other times with similar net positive binding of Clq^bl° to K-CRl cells as compared with K-Hyg. The binding was saturable and half maximal binding was seen at a concentration of 35 μg/ml of Clq⁰¹⁰ concentration of 7.6 x 10⁸M.

¹²⁵I-Clq BINDING TO TRANSFECTED CELLS Transfected K-Hyg and K-CRl cells, each at 5 x 10^c//xl, were reacted with increasing doses of ¹²⁵I-Clq (specific activity: 1.1 x 10⁶/μg) . For each dose, ⁱ²⁵I-Clq binding to transfected cells was performed m a polypropylene microfuge tube m HBSS without Ca⁺⁺ and Mg⁺⁺ (HBSS=) which was diluted with an equal volume of glucose-gelatin solution. Thus, a 1/2 ionic strength buffer (1 vol. HBSS= : 1 vol 5% dextrose, 0.2% gelatin m water) was used. The total volume of 0.33 or 0.44 μl contained 5 x 10⁶ cells/ml and was incubated with increasing doses of ¹²⁵I-Clq (specific activity: 1.1 x 10⁶/μg) and the binding reaction proceeded at room temperature with regular agitation. After 45 mm, 3 aliquots (0.1 ml) from each reaction mixture were layered onto 250 μl of an oil mixture (85% dibutyl phthalate, 15% dinonyl phthalate) m microfuge tubes (0.4 ml polyethylene, #1404-1000, USA/Scientific Plastics, Ocala, FL) . The tubes were spun for 2 mm at 9,000 x g (Microfuge B, Beckman Instruments, Fullerton, CA) to pellet the cells and the tips containing the cell pellets, containing 5xl0⁵ cells, were cut off and counted in a gamma counter. The means of triplicate values + SE were determined. At the highest input of ¹²⁵I-Clq a 50-fold excess of unlabeled ligand inhibited specific binding by 82%. This experiment was repeated four times with similar results.

Although there was a dose dependent increase in ¹²⁵I- Clq binding with increasing ¹²⁵I-Clq input, it was not possible to reach saturation conditions (Fig. 2B) . In other experiments, the number of cells was decreased and the amount of radiolabeled ligand was increased, but it still was not possible to achieve saturation because the iodmated Clq aggregated at high concentrations. The large amount of Clq^bl° needed for was due to partial denaturation of the Clq by the biotmylation procedure. As indicated above, better results were obtained using radioiodmated ligand. A 50-fold excess of unlabeled Clq inhibited 82% of ¹²⁵I-Clq binding at its maximal input. In other experiments, Clq binding to K562 transfected to express CR3 was shown to be negligible, with or without activation of the CR3 by M⁺⁺ or mAb.

EXAMPLE 5: BINDING TO INSOLUBILIZED rsCRl

Recombinant, soluble CRl (rsCRl) was immobilized on microtiter wells to circumvent the problems of background Clq binding to other cell surface sites. rsCRl is a genetically engineered CRl molecule consisting of 30 ammo terminal SCRs that lacks the normal transmembrane domain and is, therefore, soluble. Preliminary titration of CRl binding to the plate indicated that 0.5 - 0.8 μg of rsCRl/well provided optimal Clq binding. Microtiter wells (Immulon™l Removawell™ strips, Dynatech Labs, Alexandria, VA) were exposed to 0.1 ml CRl (8 μg/ml) diluted m coating buffer (0.01M NaC0₃ , 0.04M NaHCO, for 2 hours at 37°C or overnight at 4°C. For the Clq^D1° binding studies, wells were blocked with SuperBlock™ (Pierce Chemical) . For the ¹²⁵I-Clq binding studies, blocking was done for 2 h at 37°C with 3% non-fat dried milk (BioRad, Hercules, CA) , 0.5% Tween-20 (Baker Chemical, Phillipsburg, NJ) m PBS. ¹²⁵I-Clq m binding buffer (0.67 x PBS, 0.05% Tween-20, mSι=6) was incubated m the CRl coated wells for 45 mm at room temperature. After the binding incubation, the wells were emptied by aspiration and then the plate was turned upside down and vigorously slapped 5 times against layers of filter paper. Subsequently, the wells were separated and individually counted m a gamma counter.

Clq binding to cells was facilitated by low ionic strength, but Clq also aggregated at low ionic strength, wnich can make binding assays technically difficult . To identify an ionic strength closest to normal for the plate assay that would permit reproducible measurement of Clq binding, the ability of Clq^D1° to bind under various ionic strength conditions was tested. rsCRl (5 mg/lml) was used to coat microtiter wells. After washing and blocking Clq^Di° (0.5 μg/well) diluted m the respective ionic strength buffer, was added for a 30 mm incubation at room temperature. The wells were washed twice with the respective buffer. This experiment was repeated with a similar negative effect of ionic strength of Clq^bl° binding. Adequate binding was measured in 0.1 M NaCl (Fig. 3) , thus, subsequent plate binding studies were done in a corresponding low ionic strength buffer (67% PBS, 33% dHOH, 0.05% Tween-20) . Results represent the mean duplicate values, while the bars depict the range of values. This experiment was repeated with a similar inverse relationship between ionic strength and Clq^bl° binding.

Binding to immobilized recombinant soluble CRl (rsCRl) was tested. The specific (direct) binding of ¹²⁵I-Clq to immobilized rsCRl is shown in Fig. 4A. Half maximal binding was observed at 0.54nM of ^:25I-Clq. The total binding observed in this assay was approximately 30 fold lower than in the cell binding assay (Fig. 2B) for two reasons: 1) the time required to separate bound from unbound ligand was longer in the plate assay; and 2) a higher ionic strength was used in the plate assay for the reasons detailed above. Binding of labeled Clq to immobilized rsCRl could be competed by native Clq (Fig. 4B) . A concentration of 0.4 μM unlabeled Clq inhibited ¹²⁵I-Clq by 50%. The proteolytic fragment of Clq containing only the collagen "tail" portion of the molecule, which was obtained by pepsin digestion of Clq, also competed with ¹²⁵I-Clq for binding to immobilized rsCRl, which indicated that CRl binds to the same part of the Clq molecule as does Clr and Cis. In two experiments

50% inhibition of ¹²⁵I-Clq was observed with about 10-15 nM of the collagen tails (Fig 11) . The C3b dimers competed for ¹²⁵I-Clq binding, but not completely (Fig. 4C) . The C3b dimer is the well described ligand for CRl (Fearon, D.T., J. Exp . Med . 152:20-30 (1980)). The partial inhibition afforded by (C3b)₂ binding suggests that Clq and (C3b) ₂ do not bind at the identical site on CRl, but perhaps they bind at adjacent sites that produce steric hindrance, LHR-D is adjacent to the (C3b)₂ binding sites (Fig 7), and until now it has had no ascribed binding function. When ^{1 5}l-Clq binding to molar equivalents of intact rsCRl versus a deletional mutant containing LHR-D alone were compared, LHR-D was more effective (Fig 12) . Thus, Clq binding utilizes a unique site on CRl, contained withm LHR-D.

To summarize, these experiments indicate that: 1) Clq can bind to CRl directly without the participation of any other membrane protein; 2) the binding is specific for Clq; and 3) the ability of (C3b) 2 binding to CRl to partially inhibit Clq binding is likely the result of steric hindrance .

EXAMPLE 6 : BIOSPECIFIC INTERACTION ANALYSIS

Surface plasmon resonance (SPR) analysis of binding of native Clq to immobilized sCRl was performed m real time using a BIAcore™ instrument (Pharmacia) . This technique measured the rsCRl (800 μg/ml m 10 mM citrate pH 4.8 buffer x 50 μl at a flow of 5 μl/mm) which was coupled to a dextran CM5 sensor chip (Pharmacia) using EDC and NHS as per the manufacturer's instructions (Johnsson et al . , 1991) . Binding studies were performed m PBS at 25°C using a flow rate of 5 μl/mm. Data was analyzed using BIAcore incorporated software (BIA Evaluation and BIA Simulation, Pharmacia) . Insolubilized C3b, or experimentally oligermerized C3b, is the natural ligand of CRl. The ability of soluble C3b dimer to bind to the coupled rsCRl was assessed (Table

1) .

REAL-TIME KINETIC MEASUREMENT OF C3b OR Clq BINDING TO rsCRl

To demonstrate that Clq could bind CRl under physiologic conditions and to avoid the deleterious effects of labeling procedures on the activity of Clq, surface plasmon resonance (SPR) analysis of binding was performed using a BIAcore instrument. This technique measures the association and dissociation of unlabeled ligand to an immobilized receptor, or visa versa, by changes the refractive index of an adjacent structure (Cullen et al . , Biosensors 3:211-25 (1987)). Recombinant sCRl was covalently coupled to a CM5 dextran chip and resulted m the net addition of 9052 SPR units to the chip. To confirm that rsCRl was functionally intact, the ability of unlabeled soluble (C3b)₂ to bind to the coupled rsCRl was assessed. Specific binding of (C3b)₂ to rsCRl was observed over a range of concentrations from 8.3 nM to 67 nM. The association phase of the interaction was modeled as simple bimolecular binding, because the (C3b)₂ ligand was employed at a concentration 20-60 fold below the Keq of monomeric C3b for CRl (Arnaout et al . , J. Immunol . 1581:1348-54 (1981) ) . Further, the sequential binding of two identical binding sites would not change the observed Ka (Gertler et al . , J. Biol . Chem . 271:24482-9 (1996)). Once bound, the two components of (C3b)₂ could dissociate either simultaneously or sequentially from the immobilized CRl. Only the second dissociation of a sequential dissociation would be observed as a decrease m plasmon resonance units. Thus, a curve fitting model assuming parallel dissociation of two distinct complexes was employed. The association and dissociation models closely fit the experimental data. Figure 5A illustrates the raw data of the ooserved association and dissociation of (C3b)₂ from which the constants were derived. Employing the parallel dissociation model, the more rapidly dissociating complex, presumably monomeric C3b binding to CRl, had an apparent Kd that ranged from 0.066 to 0.346 among the 5 concentrations analyzed. The more slowly dissociating complex, (C3b)₂-- CRl, had an apparent Kd that ranged from 1.78 x 10³ to

2.42 x 10³. Using the latter dissociation constants m a simple association model, the apparent Ka ranged from 5.5 x 10⁴ to 1.05 x 10⁵ (Table 1.) Keq (concentration of ligand at which half the receptor binding sites are occupied) for (C3b)₂ binding to CRl were calculated and ranged from 17.9 nM to 40.6 nM and the average calculated Keq was 27.6 nM. These values are not significantly different from the previous determination of 9.5 nM for a similarly prepared (C3b)₂. Thus, the immobilized rsCRl was m an apparently "native" state with respect to binding its known ligand.

An alternative analysis of the association kinetics, which does not assume a known Kd, yielded a Ka of 9.79 x 10⁴, an apparent Kd of 1.7 x 10³ and a calculated keq of 17.36 nM (Fig. 5B) . Using the average Kd determined from the dissociation kinetics, 2.078 x 10³, the calculated Keq for (C3b)₂ binding to CRl is 21.2 nM.

Table 1. Kinetic Data for (C3b)₂ Binding to Immobilized CRl

In a similar analysis using unlabeled Clq m concentrations from 6.5 to 52.2 nM we were able to detect binding to CRl m normal ionic strength buffer (Fig. 6A) . No Clq binding was observed to a parallel channel m the chip that was not derivatized, or to a channel that was derivatized with Clq, data which demonstrated the specificity of Clq binding to rsCRl in this assay. Using the same models for dissociation kinetics, a good fit to the experimental data was obtained assuming parallel dissociation of two complexes. This might represent two combinations of the potential six binding sites for CRl on the hexameπc Clq molecule. Analysis of the dissociation data for the more rapidly dissociating complex yielded an apparent Kd that ranged from 0.019 to 0.033 among the five concentrations analyzed. The more slowly dissociating complex had an apparent Kd that ranged from 0.84 x 10^~3 to 2.09 x 10A Using the latter dissociation constants in a simple association model, the apparent Ka ranged from 2.86 x 10⁵ to 4.34 x 10⁵ (Table 2) . The Keq for the slowly dissociation Clq- -CRl complex were calculated and ranged from 2.27 nM to 5.17 nM, and the average calculated Keq was 3.9 nM. An alternative analysis of the association kinetics, which does not assume a known Kd, yielded a Ka of 4.34 x 10⁵, an apparent Kd of 1.95 x 10^~3 and a calculated Keq of 4.49 nM (Fig. 6B) . Using the average Kd determined from the dissociation kinetics, 1.49 x 10^~3, the calculated Keq for Clq binding to CRl is 3.43 nM.

Table 2. Kinetic Data for Clq Binding to Immobilized CRl

When Clq was immobilized to the chip of NHS/EDC chemistry, rsCRl binding to isotonic buffer was not detectable. However, when Clq^bl° was immobilized to a strepavidm chip it was possible to detect CRl binding m the concentration range of 435 nM to 3,480 nM. The very rapid association and dissociation of rsCRl to the immobilized Clq^bl° precluded accurate determination of the rate constants. The relatively high affinity of the multimeric Clq for CRl and the relatively poor binding of CRl to Clq suggests multiple binding sites on Clq for CRl, but a single binding site on CRl for Clq. An alternative explanation would be that the biotmylation of Clq has altered Clq binding to CRl.

The BIAcore™ software allows one to fit experimental curves assuming one or two binding sites, and then calculate the probabilities that either model fits the data on the basis of these iterations. Clq bound to the coupled rsCRl and rsCRl bound to coupled Clq.

EXAMPLE 7 : ERYTHROCYTE Clq RECEPTOR

It has never been shown that Clq can bind to erythrocytes. Erythrocytes are CR1⁺, but lack some of the confounding membrane proteins such as Fc receptors and integrins . ¹²⁵I -collagen tails of Clq did bind to E and the non-specific binding could be estimated from the linear portion of the curve, which allowed the calculation of specific binding (Fig 13) . At 5 nM a 10-fold excess of unlabeled ligand inhibited binding by about 50% (*) . This shows that the collagen tails of Clq can bind to erythrocytes (presumably via CRl) , and in addition it has important implications for the transport of immune complexes . In a different analysis, heat aggregated IgG (HAG) was radiolabeled and Clq-dependence of ¹⁵I-HAG binding to freshly isolated erythrocytes was tested. No ¹²⁵I-HAG binding occurred in the absence of Clq, and then there was a direct dose response of Clq input with increased ¹²⁵I-HAG binding to the erythrocytes (Fig 14) . The importance of this experiment is that it utilizes non-derivatized Clq to demonstrate the binding. In additional experiments, Fab ' ₂ fragments of polyclonal anti-CRl blocked up to 80% of Clq-¹²⁵I-HAG binding.

EXAMPLE 8: MANNOSE BINDING LECTIN (MBL), BUT NOT ¹²⁵I- SURFACTANT PROTEIN A (SPA) , BINDING TO CRl

rsCRl was immobilized in microtiter wells and the radioiodinated collectin was tested for binding. ¹²⁵I-MBL bound, and the binding was significantly augmented by the presence of its lectin ligand, mannan (Fig 15) . In contrast to MBL, ¹²⁵I-MBL is not only homologous to Clq, it, like Clq, binds to an SCR-containing serine esterase (MASP) . This experiment shows that MBL and Clq can bind to the same receptor protein. In similar experiments, ¹²⁵I-SPA alone did not bind. SPA binding can be assessed in the presence of the appropriate phospholipid ligand. These findings are consistent with the hypothesis that the opsonic property of MBL is mediated by CRl. Thus, CRl functions more broadly than previously appreciated as an opsonic receptor for the ligands of innate immunity.

EXAMPLE 9: BINDING OF FACTOR H AND C4 BINDING PROTEIN (C4bp) TO IMMOBILIZED Clq AND VICE VERSA

When Clq was immobilized (Left bars, Fig 16) factor H bound to Clq relatively less well than C4bp. This apparently is due to the fact that factor H, being a single chain, has monovalent binding to Clq, whereas C4bp, with multiple identical subunits, has multivalent binding. In the reciprocal experiment, when factor H and C4bp were immobilized and Clq was the fluid phase ligand, there was still better binding to C4bp, but the binding to factor H was significantly improved (right bars, Fig 16). Thus, factor H, like CRl, when it is immobilized can compensate for its functional monovalency in this binding assay. These results confirm that Clq can bind other proteins containing SCR domains, but the relative importance of this binding will depend on the concentration and valency of the protein.

EXAMPLE 10: BINDING SITE(S) ON THE COLLAGEN DOMAIN OF Clq FOR CRl AND OTHER RELEVANT SCR-CONTAINING PROTEINS

The above examples document that Clq can directly bind to CRl. Further testing can demonstrate the diverse Clq-mediated responses of cells can be related to CRl and/or to structurally homologous SCR-contammg proteins. Examples of Clq-mediated responses include: 1) for phagocytes--superoxide production and enhanced phagocytosis; 2) for B cells- -enhanced Ig production; 3) for fibroblasts (murine) --activation of Ca⁺⁺-actιvated K+ channels and chemotaxis; and 4) for platelets- -expression of procoagulant activity.

The identification of the CRl binding site on Clq can afford analysis of the mechanism of binding. Clr and Cis contain SCR domains. If, the binding of Clr₂Cls₂ to Clq blocks the ability of CRl to bind to Clq, there will be evidence for a common SCR binding site on Clq. The implications of a common SCR binding site on Clq are: 1) an SCR protein might normally be complexed with free Clq m plasma and thereby regulate Clq's ability to participate m Cl activation; and 2) other SCR membrane proteins might subserve the role of CRl for cells which do not express significant amounts of CRl.

1. ASSESS THE ABILITY TO CRl AND Clr₂Cls₂ TO COMPETE FOR BINDING TO Clq

Two direct approaches can be taken to test for the ability of CRl and Clr₂Cls₂ to compete for binding to Clq.

The first method utilizes unlabeled Clq and measures the ability of the Clq--Clr₂Cls₂ complex to bind to immobilized CRl. Binding can be assessed by the detection of the active

Clr₂Cls₂ subunit using a colorimetπc substrate.

Microtiter wells containing immobilized IgG (or, if necessary, heat aggregated IgG, or immune complexes) can be used as the positive control for Clq--Clr₂Cls₂ binding.

The second binding assay is based on the ability of ¹²⁵I-Clq when reacted with various molar ratios of zymogen Clr₂Cls₂ to bind to either immobilized rsCRl, or to bind to CRl transfected cells. (Zymogen Clr₂Cls₂ has about 200-fold higher affinity for Clq than does active Clr₂Cls₂) . Cells transfected with the plasmid minus the CRl construct can serve as the control cells to define non-specific binding.

2. SPECIFICALLY MODIFY THE CRl BINDING SITE ON Clq AND MEASURE THE EFFECTS ON BINDING

The most direct approach would be to use site directed mutagenesis, but it is likely the binding site on Clq has contributions from at least two of the three chains of Clq. In addition, one can use chemical modification techniques, which have been used successfully to define the positively charged residues on the collagen domain of Clq tnat are critical for Clr₂Cls₂ binding. Similar procedures can be used to modify lysine residues with pyridoxal 5 ' -phosphate and acetic or citraconic anhydride. Argmme residues can be modified with cyclohexanedione . Confirmation of the modified ammo acid residues can be obtained directly by sequencing and functionally by testing Clr₂Cls₂ binding, which should be inhibited. If the neutralization of basic residues on Clq does inhibit CRl binding, acidic residues can e modified on CRl using 2- (N-morpholmo) ethanesulfonic acid with EDC and glycine ester, a method that has modified Clr such that Clr₂Cls₂ cannot bind to Clq. If these methods do not block the Clq-CRl interaction, the strategy can be reversed and the basic residues on CRl and the acidic residues on Clq can be modified. The profound inhibition by salt of the Clq-CRl interaction, like the Clq-Clr₂Cls₂ interaction, suggests that ionic bonds are crucial.

3. USE OF MONOCLONAL ANTIBODIES TO Clq FOR THE MAPPING OF THE CRl BINDING SITE

Monoclonal antibodies (mAbs) directed against Clq are particularly valuable reagents not only for mapping the CRl and Clr₂Cls₂ binding sιte(s), but also for the affinity purification of Clq fragments. This technique has been successfully used to map the binding site on the A-cham of Clq for CRP, serum amyloid P, and lipid A of endotoxin. One anti-Clq mAb has been described to block the interaction of Clq with Clr₂Cls₂. This mAb and its parent hybridoma clone are no longer existence. mAbs can be made by a standard protocol as described herein. For example, mice can be immunized twice (three weeks apart) mtraperitoneally with purified human Clq, the first time using complete Freund ' s adjuvant and the second time using incomplete Freund' s adjuvant. Subsequently, serum titers of anti-Clq antibody from individual mice can be checked m a standard sandwich ELISA for Clq. Those mice with the highest titer can be re-boosted with more Clq given intravenously. Three days later the mice can be sacrificed and their spleen cells used for fusion, as per standard protocol. All hybridomas can oe recloned until they are stable. Hybπdoma cells can be screened for anti-Clq production by a standard Clq sandwich ELISA. Positive wells can be identified for their binding to the collagen or globular domain. Some globular domain specific antibodies can be cloned; they could be useful reagents for blocking Clq's reactivity with immune complexes. Antibodies specific for the collagen domain can also be used. Collagen domain-specific mAbs will be further screened for their ability to bind to ¹²⁵I-Clq and prevent ¹²⁵I-Clq from subsequently binding to immobilized CRl. Anti-Clq collagen domain specific mAbs can be screened for their ability to block Clr₂Cls₂ binding by a functional assay. Clq can be immobilized m microtiter wells, the mAbs added, then isolated Clr₂Cls₂ added, and finally a colorimetπc assay of Clr₂Cls₂ activity can be employed. Diminished generation of a colcrimetπc product can imply less Clq-Clr₂Cls₂ binding. Alternatively, a functional assay for C4 cleavage can be used. A family of mAbs of differing specificities for Clq can be useful not only for mapping binding sites, but also to affinity purify the collagen fragment of Clq and smaller peptides.

4. DETERMINE IF OTHER SCR-CONTAINING PROTEINS BIND TO Clq

To assess membrane SCR-contammg proteins, K562 cells can be transfected to express the relevant protein. As shown herein, K562 cells have only a low background level of Clq binding. Because K562 cells, which are known high expressors of CD46, do not bind Clq well, it may be preferable to study DAF. DAF has 4 SCRs and it is ciustered m the membrane and therefore could potentially provide a multivalent binding site. Preliminary studies nave shown that Clq binding to PNH erythrocytes (98% DAF-deficient) is m the normal range. For a more complete evaluation of the role of DAF, the GPI -anchor deficient K562 cells previously characterized can be used. If the results of this experiment indicate GPl-anchor dependent Clq binding, recombinant soluble DAF can be prepared to perform direct binding studies. Potential binding sites can be mapped as described for CRl (See Example 11) using described reagents. Positive results can be followed up with BIAcore studies.

In terms of other SCR-contammg membrane proteins that might subserve the function of an ClqR, a good candidate is complement receptor 2 (CR2, CD21) , which is comprised of 15-16 SCRs. Aggregated Clq has been shown to bind and modify B cell function, and the Raj I B cell line, which lacks CRl, can still bind Clq. CR2 can form complexes with CRl which makes it possible that CR2 and CRl, together m a complex, each contribute a binding site for Clq. Binding to K562 cells transfected with CD21 alone and co-transfected with CD21 and CRl can be assessed for Clq binding. The mapping of Clq binding to CR2 can be studied using domain deletions of CR2. There is an array of commercially available antι-CR2 mAbs, which bind to all regions of CR2 except SCR 5-8. There are also mAbs directed against SCRs 5-8. In addition, a high titer rabbit antι-CR2 can be used m blocking studies. Positive results can be followed up with BIAcore studies using recombinant, soluble CR2.

Serum proteins, such as factor H, B2 -glycoprotein- 1 , and C4 binding protein (C4bp) can be studied. As described m Example 9, ¹²⁵I-Clq collagen domains bound to immobilized purified factor H (20 SCRs) (Fig 16) . Factor H may resemble CRl m terms of having only one low affinity binding site for Clq, and thus it is likely that when the experiment is done m the reverse that factor H will exhibit poor binding to immobilized Clq. Finally, factor H can potentially bind to the cell surface through its heparm-heparan sulfate binding domain, and m this capacity it could potentially participate m Clq binding. Additionally, factor H-like protein 1, which has four SCRs, contains an RGD sequence that could be used m cell binding.

C4bp is comprised of six identical subunits (eight SCRs/subunit) and one homologous 13 subunit (three SCRs) . Each of the subunits can bind a C4b; while the 13 subunit, through its first SCR, binds one molecule of vitamin K dependent protein S of the clotting system. In preliminary experiments, ^j-²⁵I-Clq collagen domains bound well to immobilized C4bp, and because of multivalency, ¹²⁵I-C4bp bound well to immobilized Clq (Fig 16) . Additional binding studies of C4bp to CRl can be performed. C4bp can be purified by standard methods, and m addition the three other known ligands of C4bp, namely C4b, serum amyloid P (SAP) and vitamin K dependent protein S, can be purified. Because all three ligands bind to independent sites on C4bp, the apility of each to block the interaction of C4bp with CRl can be assessed. Finally, we will test the relative affinities of factor H, B2-glycoprotem-l , C4bp, CRl, and Clr₂Cls₂ for immobilized Clq.

5. DETERMINE IF SCR-CONTAINING PROTEINS REGULATE Cl ACTIVATION

Aggregated Clq appears to bind better to cells. A 30-fold enhancement of HAG binding to erythrocytes occurred when an equivalent amount of aged Clq, contrast to freshly isolated Clq, was used. Indeed, the need to use low xonic strength puffers m order to demonstrate Clq binding to cells, may reflect the need to aggregate the Clq. Thus, it may be that Clq must aggregate by binding to an activator, before it can interact with the CRl on cells. Alternatively, the binding of SCR-contammg plasma proteins to Clq, which was observed m vi tro, may have a role m regulating Cl activity m plasma.

There is an emerging model of Cl m which some of the catalytic tetramer, Clr₂Cls₂, circulates with non-covalently bound Cl inhibitor (Cl INH) , leaving an equimolar fraction of Clq free. At issue is what prevents this free Clq from interacting with CRl. According to this model, when Clq binds an activator, the Clr₂Cls₂ -Cl INH comp ex binds Clq, then the Cl INH dissociates and allows the Clr₂Cls₂ to assume a figure of "8" conformation around the stems of Clq. The Clr subunits then autoactivate and subsequently activate the Cis subunits. Cl INH binds activated Clr and Cis, this time covalently, and thereby permanently inactivates them. Two molecules of Cl INH are consumed each time a Cl molecule is permanently inhibited. The importance of Cl INH is underscored by the disease associated with a functional deficiency of Cl INH, hereditary angioedema . However, it is notable that the symptoms of angioedema are episodic, which suggests that there are alternative mechanisms to dampen "spontaneous" Cl activation.

Clq plasma may be complexed with an SCR-contammg protein, which inhibits the free Clq from interacting with Clr₂Cls₂. Then once Clq binds to an activator, Clq's affinity for Clr₂Cls₂ would become greater than its affinity for the SCR-contammg inhibitor. Thus, Clr₂Cls₂ would replace the SCR-contammg inhibitor on bound Clq and thereby assemble macromolecular Cl . In this model the

SCR-contammg inhibitor would need to be multivalent, and although there are several potential candidates m serum, C4 binding protein (C4bp) with its seven SCR-contammg branches, six of which are identical, is a logical candidate molecule to investigate first. This hypothesis would provide an explanation for the observed high efficiency of CRl and C4bp m inhibiting the classical pathway.

To determine if Clq is complexed with another protein, fresh heparmized plasma can be obtained from normal donors. (Chelation of Ca" leads to the immediate activation of Cl) . Clq can be immunoprecipitated, the precipitate can be radioidmated, and the complex resolved by SDS-PAGE. Unlabeled Clq lmmunoprecipitates can be scrutmized for the presence of C4bp antigen using ELISA and immunoblotting techniques. In addition, C4bp can be immunoprecipitated and the presence of Clq can be detected by the same techniques. For immunoprecipitation F(ab')₂ fragments of antibody can be used to avoid activating Clq. If an SCR-contammg protein, such as C4bp, is normally complexed with Clq m plasma, functional Cl hemolytic assays can be used with purified proteins to determine if the SCR-protem can regulate Cl activity.

EXAMPLE 11: LOCALIZATION OF THE BINDING SITE ON CRl FOR Clq

The most common allotype of CRl m humans is comprised of 30 SCRs m the extracellular domain m a linear array. The ammo-terminal 28 SCRs are organized m four long homologous repeats (LHR) of seven SCRs each. The strong sequence homology between LHRs of CRl results m conserved restriction sites between LHRs. These sites were used to generate a panel of deletion mutants of cDNA encoding CRl which were used to identify the location of C3b and C4b binding sites. This panel of deletion mutants was also used to localize the epitopes for mAb binding. Because of the substantial sequence identity between LHRs, most mAb to CRl bind more than once to each molecule. For example, the mAb YZ-1 has three epitopes on the common allotype of CRl. The dual approach of testing binding of Clq to the deletion mutants of CRl, together with mAb mapping and blocking studies, can enable localization of the Clq binding site (s) to a defined number of SCRs. Site directed mutagenesis of the SCRs involved in Clq binding can further define the residues important for binding. Because there is no structural homology between C3b/C4b and Clq, and (C3b)₂ only partially inhibited Clq binding to CRl, it is unlikely that the point mutations previously created in the C3b/C4b binding sites of CRl will be helpful, and a new panel can be prepared.

1. DEFINE THE DOMAIN OF CRl THAT IS INVOLVED IN BINDING Clq CRl binding domains can first be defined by a) expressing domain deletion mutants of CRl on transfected cells; and b) by using mAb anti-CRl to block Clq binding.

After the general region of Clq binding is determined, the binding site on CRl can be more closely defined by c) adding and deleting specific SCR; and then d) using site directed mutagenesis to define the relevant amino acid residues for binding.

DOMAIN DELETION MUTANTS OF CRl:

Domain deletion mutants of CRl can be expressed on the surface of transfected cells. This method has worked for the expression of full length CRl in the K562 cell line. Because of the homology between LHR, and the fact that specific functional binding domains are contained within LHR, deletional constructs can be made that are missing LHR-D. Constructs of LHRs AD, BD, CD, and D that were previously used to define the C4b and C3b binding sites have also been used to define a Clq binding site on LHR-D. As an alternative, recombinant soluble protein can be prepared corresponding to each deletion mutant and the protein can be immobilized on plastic or a chip for BIAcore analysis. Purification can be from CHO transfectants in serum-free media or via a specific C-terminal tag, e.g. his tag, myc tag or a GST-fusion protein. Clq is very sticky and pilot Clq binding studies on irrelevant proteins with the various tags can be prepared to make sure the tags themselves do not significantly raise nonspecific binding.

MAPPING THE BINDING SITE WITH mABS :

Anti-CRl mAb can also be used to map the binding site. Twenty three different mAbs directed against CRl are available from the combined lists of the American Type Culture Collection (ATCC) and Linscott ' s Directory of Immunological and Biological Reagents, 9th ed. Most of the anti-CRl mAbs are available commercially. For those 20 mAbs without a defined binding site, the binding site can be defined using deletional mutations of CRl. Alternative methods for epitope mapping include the use of recombinant soluble peptides for ELISAs and the use of BIAcore technology. Examples of mAbs that can be used are: 1) YZ-I binds LHRs A, B and C, but does not block C4b or C3b binding unless second antibody is used; 2) 3D9 does block C3b binding; 3) 543, which does not block Cab or C3b binding; and 4) an mAb mapped to LHR-D. Once epitope maps have been constructed, transfected K562 expressing CRl with a specific mAb, can be treated, and ¹²⁵I-Clq or collagen ta i binding can be assessed m standard cell binding assays .

2. DETERMINATION OF THE SPECIFIC BINDING SITE

Once a binding region is identified by the techniques apove, a soluble protein containing only this sequence and proteins larger by one SCR on either or both sides can be prepared. These proteins can be immobilized to perform binding studies to demonstrate that the proposed region is not only necessary, but also sufficient, for binding Clq. The BIAcore technology can be employed to determine the apparent binding and dislocation constants for the candidate binding sites and these can be prepared with the constants that have been determined for full length CRl. In studies of C3b binding, while three SCRs of CRl (8-10 or 15-17) are sufficient to generate a C3b binding site, four SCRs are necessary to create a binding site with the same apparent affinity as the parent molecule.

Once the Clq binding site has been localized to a limited number of SCRs by the methods outlined above, site-directed mutagenesis can be used for the production of mutants. This method has been used to produce mutations m the cytoplasmic domain of CRl. Residues that can be targeted include: residues suggested to be involved by the chemical modification studies outlined m Example 10, residues that are conserved between CRl and Clr/Cls out that are not part of the framework consensus of the SCR; and those that may oe identified by alignment with other SCR-contammg proteins that are found to bind to Clσ Example 10. Alternatively, if the binding site were localized to one or two SCRs, a set of overlapping oligonucleotides can be synthesized where sequences encoding 4-5 contiguous ammo acids would be changed to codons encoding alanme, a strategy termed alanme scanning mutagenesis. None of the invariant framework residues of the SCR need be changed. An advantage of this strategy is that all non- framework residues would be examined. Site directed mutagenesis has already been used successfully to determine the residues important for C4b and C3b binding. The panel of mutant proteins can be assayed individually for Clq binding, and residues m each cluster identified by the alanme scanning technique can be evaluated individually . If a relatively short linear stretch of ammo acids are identified as a major component of the Clq binding site, this might be a starting point for the development of peptide or small molecule inhibitors of Cl, which would be useful to investigate the role of acute classical pathway activation m disease states. The chronic use of an inhibitor that blocked Cl activation would probably be harmful because it would inhibit C3b and C4b deposition as well as block clearance by ClqR. There are clinically important examples of acute Cl activation, however, which would benefit from Cl inhibition, such as the reversal of heparm anticoagulation with protamme. 3. DETERMINE IF THE OTHER HUMAN COLLECTINS, NAMELY SURFACTANT PROTEIN A AND MANNOSE BINDING PROTEIN, ARE ABLE TO BIND TO CRl

Surfactant Protein A (SPA) and mannose binding lectin (MBL) together with Clq form a subgroup of the collectm family of opsonms . All three proteins have globular heads and collagen-like triple helix tails, and have six identical subunits to form a tulip- like structure. These proteins function as important elements of innate immunity by binding directly to a variety of surfaces, microorganisms, etc., and target them for phagocytosis. MBL also has two associate serine proteases with strong homology to Clr and Cis, termed MBL-associated serine proteases (MASP-1, MASP-2) , and has been shown to be able to activate complement. ClqR may also bind SPA and MBL, although the receptor molecule has not been clearly identified. CRl, or another SCR-contammg membrane protein, may be the general collectm receptor.

In preliminary studies, ¹²⁵I-SPA did not bind to immobilized CRl, while ^12ΞI-MBL did bind. Since MBL binding was greatly enhanced by the presence of mannan (Fig 15) , the SPA binding can be repeated also m the presence of its ligand. Reciprocal competition experiments between the proteins can be performed to see if one collectm will compete for binding to CRl with the others. If specific binding is demonstrated, the binding studies can be repeated with some of the CRl mutants prepared herein, to determine by an alternate method whether the binding sites for these proteins overlap with that for Clq. This second strategy can be important because the collectins are large, approximately 500 kD, and may interfere with each other sterically, while not having overlapping or identical binding sites. If specific binding is observed, then BIAcore assay of binding kinetics can be performed for each collectm to calculate an apparent equilibrium dislocation constant for each ligand binding to CRl. If different binding sites are observed, then the binding site for each new collectm can De determined as previously discussed for Clq. It is likely that they bind to the same site, because the conditions and circumstances for activation of each of these three collectins are predominantly nonoverlapping; SPA is found m the airways, MBL and Clq are both serum proteins, but MBL is primarily activated by mannan-bearmg surfaces, and Clq by antigen-antibody complexes.

EXAMPLE 12 : THE ROLE OF CELLULAR CRl AND/OR OTHER SCR- COΝTAIΝIΝG MEMBRANE PROTEINS AS Clq RECEPTORS

CRl may not be the only ClqR: the existence of a second ClqR is presumed because there are cells lacking CRl, such as the Raj l B cell line, that can bind Clq. Although this Example is focused on the role of CRl, if positive information on other Clq-bmdmg membrane proteins is obtained (see Example 10) , the focus of the experiments can De broadened. Human phagocytic cells (PMN and monocytes) and B cells can be studied. Routine cell binding assays can be used, and if necessary, the collagen fragment of Clq can be used to minimize non-specific binding.

Although there is only one form of CRl expressed on different cell types, its function is quite different. Erythrocyte CRl is clustered, which allows multivalent binding of Clq, C3b, or C4b opsonized immune complexes. The clusters are separated in such a way that the adherent immune complexes do not entice phagocytic cells to see the erythrocytes as a target for ingestion. CRl is not attached to the erythrocyte cytoskeleton, nor can its expression be upregulated on this cell. Finally, ligation of erythrocyte CRl does not lead to its phosphorylation. PMN, on the other hand, have a large intracellular pool of CRl in secretory vesicles, and a wide variety of stimuli (physical stresses, exogenous substances, and endogenous mediators) cause the immediate upregulation of CRl expression on the membrane, where it is also expressed in clusters. CRl is phosphorylated in phagocytic cells and it becomes associated with FcyR and the cytoskeleton. Thus, the same molecule has very different potential for signaling in different cell types.

1. ROLE OF CRl AS THE Clq BINDING PROTEIN ON PMN, MONOCYTES, B CELLS AND MUTAGENIZED THP-I AND RAJI CELL LINES ANTI-CR1 BLOCKING STUDIES ON PHAGOCYTES:

The first step is to determine if CRl is the principal and/or sole ClqR on PMN and monocytes. The large intracellular pool of CRl on phagocytes, which is easily upregulatable to the cell surface, can easily confound binding studies. Taxol inhibits this process and taxol treated cells can be used for these studies. Affinity purified rabbit anti-CRl F(ab')₂ fragments can be used to block binding sites on CRl. The F(ab')₂ must be carefully absorbed with protein A/G to remove all intact IgG that might independently bind Clq. In addition, binding studies can utilize the collagen fragment of Clq, which lack the Ig-b ding globular domains. Monoclonal antibodies against CRl that are specific for the Clq binding site are particularly useful (Example 11) . If it is impossible to obtain/make mAbs that block Clq binding, mAbs that bind to neighboring sites on CRl can be augmented with F(ab')₂ fragments of second antibody. This technique was used successfully to block C3b binding to CRl. Anti-HLA framework mAb (w6/32) can be used as a control and its input can be adjusted by flow cytometry to correspond to a saturating amount of anti-CRl mAb.

For these studies one can confirm that the affinity purified rabbit anti-CRl blocks 100% of Clq binding to CRl. This can be done m Clq binding assays using CRl transfected K562 cells as targets. In addition, one can confirm that the antibody can block 100% of Clq binding to rsCRl immobilized on plastic. If the available rabbit anti-CRl does not fully block binding, a new polyclonal antisera can be made and affinity absorbed with the Clq binding site domain of CRl. Fragments of rabbit anti-human B-2 -microglobulm can be used as a control and its input can be adjusted by flow cytometry to correspond to a saturating amount of polyclonal anti-CRl Ab .

If the antibody blocking studies indicate that CRl is responsible for the specific Clq binding on PMN and monocytes/macrophages, one can proceed with the remaining projects. If, on the other hand, CRl does not account for all the specific Clq binding, the role of other membrane SCR-contammg proteins can be investigated, as identified m Example 10. If DAF is identified as binding Clq, two approaches can be used to eliminate/block DAF as a receptor. First, the cells can be enzymatically treated with PIPLC, which will remove GPI -anchored membrane molecules, including DAF; and second, DAF can be blocked with affinity purified anti-DAF F(ab')₂ fragments. Reagents such as Trifluoperazme can be used during binding studies to block the upregulation of DAF from intracellular stores .

KNOCKOUT CRl ON THE MONOCYTE CELL LINE THP-1:

An alternative approach to the identification of non-CRl ClqR can be to knockout CRl expression on THP-I cells. A human cell line can be used rather than cells derived from a CR1/CR2 knockout mouse, because all the Clq-bmdmg and functional studies have been done with human cells. THP-I monocytic cells were chosen because they are one of the few cell lines with even moderate expression of CRl. First, a THP-1 cell clone with uniform high expression of CRl can be obtained. The clone can then be subjected to chemical mutagenesis with ethylmethanesulfonate (EMS) and CRl negative clones will be selected by panning using mAb an 1-CRl. A positive clone that has been carried through the mutagenesis and most of the selective procedure can be selected for use as a control. These paired cell clones will only differ m CRl expression- -to be confirmed by transfectmg CRl and restoring the mutant cell to wild type. The paired cell clones can allow the assessment of the contribution of CRl to the total Clq binding capacity of the cell.

ROLE OF CALRETICULIN IN BINDING Clq AT THE CELL SURFACE:

While not a ClqR, as originally claimed, calreticulin is a ubiquitous, sticky, intracellular chaperone protein, which can bind Clq. It is found predominately m the ER, but it can localize m the secretory granules of some cells and has been reported on the surface of PMN. Calreticulin expression on gated populations of buffy coat cells, which have not been subjected to isolation procedures, can be compared with calreticulin expression on isolated populations of monocytes and PMN. Double staining with propidium iodide and rabbit anti-calreticulm (commercially available) plus FITC-second antibody can be used to gate out dead cells. It is possible that steps during the routine preparation of cells, such as the hypotonic lysis step used to eliminate erythrocytes, induces sufficient degranulation/damage for calreticulin to be expressed on the cell surface. If this is so, cell isolation techniques can be modified to avoid tnis outcome. The presence of calreticulm on even a small subset of the population of interest would confound the Clq binding experiments.

ClqR BLOCKING STUDIES ON B CELLS:

B cells express both CRl and CR2 , which has m its most common form 15 SCR. The ammo terminal two SCRs of CR2 contain the C3dg binding and EBV binding sites. CR2 can associate/bind withm the B cell membrane with CRl and CD19; and has been reported to bind to CD23 on the same or adjacent cells. Activation and modulation of the B cell response is a complex process, and two objectives can be focused upon: first, determining what SCR-contammg molecules bind Clq on B cells: and second, m a later section, determining if the Clq-mediated effects on B cells can be accounted for by the Clq-bmdmg receptors that are identified.

Routine ¹²⁵I-Clq binding studies using freshly isolated B cells from peripheral blood and lymphoid organs can be performed. B cells will be purified by Percoll gradients and assessed for purity by CD19 antigen, and the lack of T cell and monocyte markers using FACS. Functional purity of the B cell population will be confirmed by lack of a mitogenic response to phytohemagglutmm. The separate populations of B cells from the Percoll gradient can be used: low density cells = activated cells; middle density = pre-activated cells; high density = resting cells. F(ab')₂ fragments of anti-CRl mAb that block Clq binding can be used (Example 11) to determine if there is residual Clq binding to other ClqRs . If CR2 is identified as binding Clq (Example 10), F(ab')₂ fragments of polyclonal antι-CR2 to block it. The purpose of the antibody blocking studies is to define the conditions which will allow the assay the individual contribution of a particular ClqR m the functional studies.

Clq BINDING STUDIES ON RAJI CELLS:

Although the Raj l B cell line, which expresses neither CRl nor the 126 kDa protein, reportedly does bind Clq and has been used as a source of Clq-bmdmg proteins, this is a transformed cell line and its Clq binding molecules may not be relevant to normal B cells. It is of interest that the Raji cell line expresses high levels of both surface IgM and CR2. Thus, if Clq binds well to Raj l cells, but the collagen tails do not, it would be compatible with Clq binding the surface IgM. However, if the collagen tails bind well, it would be consistent that CR2 was a ClqR. If the results of Example 10 do not support CD21 as a ClqR, the cells can be mutagenized, as per the THP-1 cell line, selecting for surface IgM negative cells and repeating the Clq binding assays. If Ra l cell Clq binding is not explained by either surface IgM or CD21, determination of the Clq binding molecule (s) remains an important goal.

2. ROLE OF CRl IN MEDIATING Clq-INDUCED BIOLOGICAL RESPONSES CRl is likely the major ClqR of phagocytes. If the results described herein suggest an additional molecule, the focus of this section can be broadened. The assays used by others can be repeated as evidence of Clq-mediated responses, and in addition the studies can be extended to better define the role of Clq in innate immunity.

PHAGOCYTOSIS STUDIES: In certain previous assays, Clq is immobilized to plastic and the monocytes, which are in suspension with the target IgG coated erythrocytes, are added later. This is a complicated assay and it is difficult to conceptualize in terms of the zipper model of phagocytosis. However, one explanation is that some of the aggregated Clq transfers from the plastic to the antibody complexes on the erythrocyte (E) (which would settle before the monocytes) , thereby converting the target from an EA to EAClq, and thus recruiting CRl as well as the Fc receptor of the monocytes. Phagocytosis by the Fc receptor is known to be greatly enhanced by recruiting CRl. Furthermore, Clq is known to be able to transfer between surfaces and/or immune complexes. Although CRl is already thought to participate in the plated Clq-mediated phagocytosis assay, it may be difficult to completely block CRl function in this experimental setup when the Clq is immobilized beneath the monocytes. Sheep erythrocyte targets opsonized with Clq and other ligands as noted below can be used. The potential role of the 126 kDa protein can be tested using blocking antibodies in our phagocytic assays. In addition, reagents that block Clq-CRl binding (CRl- clone of THP-i cells, anti-CRl mAbs, or SCR-containing peptides) can be used in an assay of phagocytosis. The results of these two approaches should indicate whether the 126 kDa membrane molecule functions cooperatively with CRl to mediate phagocytosis .

For PMN it is controversial whether CRl ligation alone can cause a Ca⁺⁺ flux, with most investigators believing that ligation is also required. While CRl ligation with natural soluble ligand does not induce endocytosis, ligation does induce CRl to physically associate with the cytoskeleton. It has been difficult to isolate the effects of CRl ligation, because CRl also has cofactor activity for factor I cleavage of C3b to ιC3b. ιC3b is the natural ligand for CR3 , which is an efficient mediator of phagocytosis. It is likely that many C3b opsonized particles are passed on from CRl to CR3 by this mechanism. Because Clq cannot be further processed, it will provide unique data as a natural and specific ligand for CRl. In addition, CRl has been shown to physically associate with FcyR (FcyR type not defined) on PMN. Although Clq can directly opsonize certain supstances, there are also many circumstances where the Clq will be recruited by bound IgG. In the latter situation, the cooperative interaction of CRl and FcyR would be very important. The ability of Clq opsonized particles, with and without the addition of IgG, to be ingested by PMN, monocytes, and macrophages can be assessed. Fibronectin and larnmm can enhance

C3b-medιated phagocytosis and both of these molecules can bind to Clq. One can investigate whether the enhancement of phagocytosis can be attributed to bound Clq. These proposeα studies will define the role of CRl m mediating Clq enhanced phagocytosis, a potentially important arm of innate immunity.

SUPEROXIDE STUDIES:

PMN layered on to Clq-coated plates will be stimulated to produce superoxide, and this production can be partially inhibited by antι-CR3 mAb, but not by antibody to the 126 kDa protein. Because ligation of CRl alone does not induce superoxide production, superoxide production may represent the coordinated ligation of CRl and CR3 , such as has Deen reported for FcγIII and CR3. This hypothesis can be tested by ligatmg CRl and CR3 (with mAb or ιC3b) separately, and then together. It may also be necessary to further crosslink with anti-mouse Ig. F(ab')2 fragments can be used exclusively to avoid engaging FcyR. An alternative explanation for Clq-mediatmg superoxide production, is that CRl ligation may lower the threshold for a second unappreciated stimulus, such as endotoxin or adherence to plastic, contaminating IgG, each of which is known to stimulate superoxide production. A continuously recording plate reader (Molecular Devices) can be used.

Clq-MEDIATED B CELL RESPONSES:

In these experiments one can determine if the published Clq-mediated effects on B cells can be accounted for by the Clq-bmdmg receptors that are identified. In one study of anti CRl ligation with polyclonal F(ab')2 fragments, it was claimed that there was enhanced Ig production with submaximally pokeweed mitogen-stimulated lymphocytes. It is difficult to interpret these experiments because there were T cells present, and some T cells are CR1⁺ . In addition, the B cells were undoubtedly contaminated with some monocytes. In a second study, aggregated Clq bound to both large and small B lymphocytes (>95% surface Ig+, <5% CD3+, <5% esterase +) derived from tonsils or peripheral blood. These B cells could be stimulated by exogenous Cl, alone, to produce more Ig; and the Clq acted synergistically with Staphylococcus aureus Cowen-1 for enhanced Ig production. In a third study, the production of IL-I by B cell lines (Raji, Daudi and Wil₂WT) was inhibited by exogenous Clq 1781.

The published functional B cell assays can be repeated and the methods can be modified so as to isolate the variables and make the results more interpretable . CRl can be ligated using either Clq or F(ab')2 fragments of anti-CRl mAb. Although ligation of CRl on B cells fails to effect a Ca^+T flux, it may modulate another signal . The use of pokeweed mitogen, which is primarily a T cell stimulant, and Staphylococcus aureus, which is both a T cell and a B cell stimulant should be avoided. Instead, membrane Ig can be cross-linked using F(ab')₂ fragments of murine anti-IgG or IgM and goat anti -mouse Ig as an auxiliary signal. The effects of added Clq will be assessed on purified B cells by 1) proliferation assays; 2) ELISA assays for Ig production, and 3) specific immunoassays for the quantification of relevant cytokines (11-2, 11-6). EBV-transformed B cell lines should be avoided because of the confounding influence of virus in these cells. The results of these preliminary studies will indicate if Clq can signal to B cells. If purified B cells do not respond to Clq, it is possible that the previously described Clq-mediated responses were mediated through contaminating CR1⁺ T cells.

3. PARTICIPATION OF ACCESSORY MOLECULES IN CR1-MEDIATED RESPONSES

CRl is apparently the major ClqR of phagocytes, but if the results of the experiments described herein suggest additional molecule (s), the focus of this section can be broadened. There are two types of signaling in which CRl participates. Signaling to CRl includes those pathways used by different agonists that "prime" PMN and by so doing functionally increase CRl expression and lead to more efficient phagocytosis by CRl with or without the participation of FcyR. An intriguing finding is that platelet activating factor (PAF) , FMLP and phorool ester induces the phosphorylation of CRl phagocytic cells (PMN, monocytes, and eosmophils) , but phorbol ester fails to phosphorylate CRl m non-phagocytic cells (B cells and erythrocytes) . Since the cytoplasmic tail of CRl from B cells and PMN is identical, the difference phosphorylation must be accounted for between the phagocytic and non-phagocytic cells by a cell type-specific difference m CRl regulation or CRl-associate proteins.

Signaling from CRl involves how CRl ligation m phagocytes induces phagocytosis and connections with the cytoskeleton and FcyR. In these experiments one can concentrate on using Clq as the ligand and identifying CRl associate molecules.

NEAREST NEIGHBOR STUDIES :

Because CRl has a short cytoplasmic tail and may not be able to signal, there is reason to believe it may have an associate signaling molecule. Molecules can be determined which physically associate with Clq-ligated CRl, before and after priming the cells with fMLP, using several standard biochemical techniques. First, cells can be surface labeled, solubilized with 1% digitonm, immunoprecipitated with antI -CRl and the lmmunoprecipitates can be analyzed by SDS-PAGE. This method has been used successfully by others to lmmunoprecipitate CRl-associate proteins from B cells. Second, anti-CRl can be used to lmmunoprecipitate membrane protein after using membrane impermeant cross-linkers. There are a wide variety of cross-linking reagents available, differing m arm length and water solubility, but SASD (sulfosuccmιmιdyl-2 - {p-azιdo-salιcylamιdo} ethyl -1 , 3 ' -dithiopropionate, Pierce) is particularly useful because it can donate a radiolabel from one protein to another. Third, once an idea of the molecular weight of CRl associate protein (s) is obtained, antl -CRl lmmunoprecipitates can be blotted, and the blots can be probed with appropriate antisera. Candidate associating molecules would be CR3 , specific FcyRs , CD47, members of the transmembrane four superfamily, and the 126 kDa membrane molecule described above. The 126 kDa protein is particularly intriguing because it is known to participate m Clq-facilitated phagocytosis, and it contains a C-type lectin domain, five EGF-like domains, a transmembrane domain, and a cytoplasmic tail . The cytoplasmic tail has a consensus repeat that is recognized by tyrosme kinases, and thus, it has the potential to signal .

If CRl associates with other molecule (s) m phagocytic cells, the domains m the cytoplasmic tail of CRl that are required for this association can be determined. The available deletional mutations of the cytoplasmic tail of CRl can be transferred into an appropriate host cell, and immunoprecipitation experiments can be performed and functional assays to define the critical domains.

EXAMPLE 13 : PREPARATION OF MONOCLONAL ANTIBODIES To raise monoclonal antibodies to novel complement receptor proteins, about six BALB/c female mice of 2-3 months age can be immunized mtraperitoneally, boosted with a ta l vein injection. The immunized mice can be tested once for an immune response by ELISA using serum obtained by a retro-orbital bleed. All animals can be sacrificed, but only the animals making antibodies can be used as a source of spleen cells. If it is difficult to obtain an immune response, or to obtain positive clones with BALB/c mice, RBF/DnJ can be substituted. Two additional adult mice of the appropriate strain can receive pristane (0.1-0.2 cc) mtraperitoneally for the induction of ascites, followed by injection of monoclonal antibody producing hybridoma cells. The animals can be separately housed and quarantined, and monitored daily for the size of the ascites and any morbidity which would require treatment or sacrifice of the animal. Each animal can be cannulated mtraperitoneally two times for the collection of ascites fluid. After the second cannulation the animal can be sacrificed.

EQUIVALENTS

Those skilled in the art will know, or be able to ascertain, using no more than routine experimentation, many equivalents to the specific embodiments of the invention described herein. These and all other equivalents are intended to be encompassed by the following claims.

Claims

CLAIMSWhat is claimed is:

1. A method of detecting the presence of Clq-contammg immune complexes m a biological sample comprising contacting the biological sample with an SCR-P that binds to Clq under conditions appropriate for binding of SCR-P with Clq, whereby producing SCR-P-bound Clq- contammg immune complexes are formed, and detecting SCR-P-bound Clq-contammg immune complexes, wherein detection of bound Clq-contammg immune complexes is indicative of the presence of Clq-contammg immune complexes m the sample, optionally wherein the SCR-P is affixed to a chip and detection of bound Clq is by surface plasmon resonance analysis.

2. A method of quantitatmg Clq-contammg immune complexes m a biological sample comprising the steps of: a) contacting the sample with an SCR-P that binds to Clq-contammg immune complexes under conditions suitable for the SCR-P to bind to Clq-contammg immune complexes present m the sample; b) detecting Clq-contammg immune complexes bound to SCR-P; and c) comparing the amount of Clq-contammg immune complexes bound to SCR-P detected m the sample to known amounts of SCR-P bound Clq-containing lmmune complexes which form a standard curve, whereby tne amount of Clq-contammg immune complexes m the sample are quantitated.

3. The method of Claim 1 or Claim 2 wherein: a) the SCR-P binds to at least one additional component present the Clq-contammg immune complex, said component selected from the group consisting of: C3b, ╬╣C3b, and C4b; and/or b) the SCR-P is an SCR-P present m a complement protein selected from the group consisting of:

CRl, the LHR-D region of CRl, CR2 , CD46, CD55, factor H, C4bp, Clr, Cis or the SCR-P is a biologically active SCR-P fragment, analog or derivative thereof, wherein biological activity is defined as specific binding to Clq-contammg complexes; and/or c) the biological sample is selected from the group consisting of: plasma, processed erythrocytes, eluant from intact erythrocytes, brain tissue, skm tissue, urine, serum, lymphatic fluid, peritoneal fluid, joint fluid, cerebrospmal fluid, pleural fluid; a fluid eluted from blood cells and a tissue biopsy sample (e.g. allograft tissue or xenograft tissue) .

4. An assay for detecting the presence of Clq-containmg immune complexes m a biological sample, comprising the steps of : a) providing the sample; b) reacting the sample with SCR-P which has been immobilized on a solid support, under conditions suitable for binding of Clq-contammg immune complexes m the sample to the SCR-P, thereby producing an immune complex product; and c) detecting the immune complex product with a labeled reagent that specifically binds to Clq- contammg immune complexes, optionally wherein the labeled reagent is selected from the group consisting of anti- immunoglobulin antibodies and immunoglobulm-reactive substances .

5. An assay for detecting the presence of Clq-contammg immune complexes a biological sample, wherein: (I) the sample is pretreated and the assay comprises the steps of : a) providing the sample; b) reacting the sample with a non-complement fixing Fab ₂ fragment of an antibody that binds to Clq-contammg immune complexes, said fragment being immobilized on a solid support, under conditions suitable for binding of Clq-contammg immune complexes m the sample to the fragment, thereby producing an immune complex product; and c) detecting the immune complex product with an SCR-P that specifically binds to Clq- contammg immune complexes; or (II) the assay comprises the steps of: a) providing the sample; b) reacting the sample with an SCR-P that binds to Clq-contammg immune complexes for sufficient time and under conditions suitable for binding of Clq-contammg immune complexes m the sample to the SCR-P, thereby producing a mixture comprising SCR-P bound Clq-contammg immune complexes, and unbound SCR-P; c) separating the SCR-P bound Clq-contammg immune complexes from the unbound SCR-P; and d) quantifying the SCR-P bound immune complexes by contacting the SCR-P bound immune complexes with an indicator reagent.

6. The assay of Claim 5 wherein the SCR-P is: a) selected from the group consisting of oligomerized SCR-P; SCR-P comprising at least two binding sites for Clq; SCR-P comprising a binding site for Clq and a binding site for C3b; and SCR-

P comprising a binding site for Clq and a binding

b) is affixed with a label, optionally wherein the label is selected from the group consisting of a radiolabel, an indicator enzyme, biotm and a fluorochrome .

7. The assay of Claim 5 (II) wherein: a) the indicator reagent is a labeled anti-SCR-P antibody; and/or b) the SCR-P is immobilized on a solid surface ; and/or c) the SCR-P is is affixed with a label, optionally wherein the label is selected from the group consisting of a radiolabel, an indicator enzyme, biotm and a fluorocnrome .

8. A method of screening for the presence of an immune complex-associated disorder or a complement activation-associated disorder a mammal comprising the steps of : a) obtaining a biological sample from the mammal, thereby producing a test sample; b) contacting the test sample with an SCR-P wnich binds Clq-contammg immune complexes under conditions suitable for binding of Clq-contammg immune complexes m the test sample to the SCR-P, thereby producing an immune complex product; c) measuring the amount of immune complex product the test sample; and d) comparing the amount of immune complex product present the test sample with a normal range of values of immune complex product, wherein an amount of immune complex product greater than the normal range of values is indicative of an immune complex-associated disorder or a complement activation-associated disorder m the mammal.

9. A method of assessing the efficacy of a drug for the treatment of an immune complex-associated disorder or a complement activation-associated disorder in a mammal, comprising the steps of: a) Obtaining a first biological sample from the mammal, thereby producing a first biological sample ; b) contacting the first biological sample with an

SCR-P which binds Clq-contammg immune complexes under conditions suitable for binding of Clq- contammg immune complexes the first biological sample to the SCR-P, thereby producing SCR-P bound Clq-contammg immune complexes; c) quantitatmg the SCR-P bound Clq-contammg immune complexes m the first biological sample, thereby obtaining a pre-admmistration concentration of SCR-P bound Clq-contammg immune complexes m the mammal; d) administering the drug to the mammal; e) obtaining a second biological sample from the mammal, thereby producing a second biological sample ; f) contacting the second biological sample with an

SCR-P which binds Clq-contammg immune complexes under conditions suitable for binding of Clq- contammg immune complexes m the second biological sample to the SCR-P, thereby producing SCR-P bound Clq-contammg immune complexes; g) quantitatmg the SCR-P bound Clq-contammg immune complexes in the second biological sample, thereby obtaining a post -administration concentration of SCR-P bound Clq-contammg immune complexes; and h) comparing the pre-admmistration concentration of SCR-P bound Clq-contammg immune complexes to the post-admmistration concentration, wherein a post-administration concentration that is less than the preadmmistration concentration is indicative of the efficacy of the drug for the treatment of the immune complex-associated disorder or the complement activation-associated disorder.

10. A method of removing Clq-contammg immune complexes m a biological fluid sample comprising the steps of: a) contacting a biological fluid sample with an SCR- P which binds to Clq-contammg immune complexes under conditions wherein the Clq-contammg immune complexes bind to the SCR-P, thereby forming SCR-P bound Clq-contammg immune complexes m the sample,- and b) separating the SCR-P bound Clq-contammg immune complexes from the sample, optionally wherein ) the biological fluid sample is a fluid obtained from a mammal m need of removal of immune complexes, optionally wherein

11) the biological fluid is returned to the mammal after the Clq-contammg immune complexes are removed.

11. The method of Claim 10 wherein the SCR-P: a) binds to at least one additional component present m the Clq-contammg complexes, wherein said component is selected from the group consisting of C3b, ╬╣C3b, and C4b; and/or b) is selected from the group consisting of CRl, the LHR-D region of CRl, CR2 , CD46, CD55, factor H and C4bp, Clr, Cis or the SCR-P is a biologically active SCR-P fragment, analog or derivative thereof, wherein biological activity is defined as specific binding to Clq-contammg complexes.

12. The method of Claim 10 and 11 wherein the fluid is selected from the group consisting of plasma, lymphatic fluid, peritoneal fluid, joint fluid, pleural fluid and cerebrospmal fluid.

13. A method of removing Clq-contammg immune complexes m a biological fluid sample comprising the steps of: a) contacting the fluid sample with the SCR-P which binds to Clq-contammg immune complexes present in the sample, thereby producing SCR-P bound Clq- containing immune complexes; b) contacting the SCR-P bound Clq-containing immune complexes with antibody that specifically binds to the SCR-P or Clq-containing immune complexes, thereby producing antibody-bound SCR-P bound Clq- containing immune complexes; and c) separating the antibody-bound SCR-P bound Clq- containing immune complexes from the fluid sample, thereby removing the Clq-containing immune complexes from the fluid, optionally wherein the antibody is a non-complement fixing Fab ' ₂ fragment of antibody.

14. A method of removing Clq-containing immune complexes from a mammalian plasma sample, comprising the steps of: a) obtaining a whole blood sample comprising a plasma portion from the mammal; b) separating the plasma portion from the whole blood sample; c) contacting the plasma portion with SCR-P immobilized on a solid support for a time and under conditions sufficient for specific binding of the Clq-containing immune complexes to the immobilized SCR-P, thereby removing the Clq- containing immune complexes from the plasma, thereby producing plasma which is free of Clq- containing immune complexes; and d) returning the plasma which is free of Clq- contammg immune complexes to the mammal, optionally wherein the Clq-contammg immune complexes contain an additional component selected from the group consisting of C3b, ╬╣C3b, and C4b.

15. A device for removing Clq-contammg immune complexes from biological fluids comprising: a) SCR-P which binds to Clq-contammg immune complexes, said SCR-P immobilized to a solid surface; and b) a means for encasing the solid surface so that the fluid may be contacted with the solid surface.

16. A method of inhibiting complement activation m a biological sample comprising contacting the sample with a polypeptide selected from the group consisting of a fragment of Clq that comprises an SCR-P binding site and an SCR-P that binds to bound Clq.

17. A method of inhibiting complement activation m a mammal comprising administering to the mammal a polypeptide selected from the group consisting of a fragment of Clq that comprises an SCR-P binding site and an SCR-P that binds to bound Clq.

18. A method of binding Clq-contammg immune complexes m a biological sample comprising contacting the biological sample with an SCR-P that binds to Clq under conditions appropriate for binding of SCR-P with Clq, whereby SCR-P bound Clq-containing immune complexes are formed.

19. An isolated and purified polypeptide comprising SCR-P or a biologically active fragment, analog, or derivative thereof, wherein the biologically active fragment, analog, or derivative specifically binds to Clq-containing immune complexes, optionally wherein the polypeptide binds to Clq-containing immune complexes comprising at least one additional component, said component selected from the group consisting of C3b, iC3b, and C4b .

20. A polypeptide consisting of a fragment of Clq comprising an SCR-P binding site.

21. A kit for detecting the presence of Clq-containing immune complexes in a biological sample comprising an SCR-P (e.g. as defined in any one of the preceding claims) which binds Clq-containing immune complexes and means for determining binding of said SCR-P to the Clq-containing immune complexes in the biological sample .

22. A kit for removing Clq-containing immune complexes comprising the components of: a) a device as in claim 15; and b) buffer reagents necessary for removing Clq- containing immune complexes from biological fluids .

23. An SCR-P as defined m any one of the preceding claims which binds to Clq-containing immune complexes, for use in therapy or prophylaxis (e.g. in humans) .

24. The SCR-P of claim 23 wherein the therapy or prophylaxis comprises removing Clq-containing immune complexes from a biological sample (e.g. according to a method as defined in any one of claims 10-14) .

25. The SCR-P of claim 24 wherein the method comprises removing the complexes from a fluid sample (the fluid e.g. being as defined m claim 12) obtained from a mammal (e.g. from a mammal m need of removal of immune complexes), the method comprising the steps of: a) contacting a biological fluid sample with an SCR- P which binds to Clq- containing immune complexes under conditions wherein the Clq-containing immune complexes bind to the SCR-P, thereby forming SCR-P bound Clq-containing immune complexes in the sample; and b) separating the SCR-P bound Clq-containing immune complexes from the sample; and c) returning the biological fluid to the mammal after the Clq-containing immune complexes are removed .

26. A polypeptide selected from the group consisting of a fragment of Clq that comprises an SCR-P binding site and an SCR-P that binds to bound Clq, for use m therapy or prophylaxis (for example m human therapy or prophylaxis, e.g. m a method of inhibiting complement activation) .

27. A pharmaceutical composition comprising: a) an isolated and purified polypeptide comprising SCR-P or a biologically active fragment, analog, or derivative thereof, wherein the biologically active fragment, analog, or derivative specifically binds to Clq-contammg immune complexes (optionally wherein the polypeptide binds to Clq-contammg immune complexes containing at least one additional component, said component selected from the group consisting of C3b, ╬╣C3b, and C4b) ; or b) a polypeptide consisting of a fragment of Clq comprising an SCR-P binding site.

28. Use of: a) an SCR-P as defined m any one of the preceding claims; or b) a polypeptide selected from the group consisting of a fragment of Clq that comprises an SCR-P binding site and an SCR-P that binds to bound Clq; or c) an isolated and purified polypeptide comprising SCR-P or a biologically active fragment, analog, or derivative thereof, wherein the biologically active fragment, analog, or derivative specifically binds to Clq-contammg immune complexes (optionally wherein the polypeptide binds to Clq-contammg immune complexes containing at least one additional component, said component selected from the group consisting of C3b, ╬╣C3b, and C4b) , for the manufacture of a medicament for use m therapy or prophylaxis (e.g. m a method as defined m any one of claims 23-26) .

29. A process for the manufacture of a medicinal composition characterized m the use, as an essential constituent of said composition, of: a) an SCR-P as defined any one of the preceding claims; or b) a polypeptide selected from the group consisting of a fragment of Clq that comprises an SCR-P binding site and an SCR-P that binds to bound Clq; or c) an isolated and purified polypeptide comprising SCR-P or a biologically active fragment, analog, or derivative thereof, wherein the biologically active fragment, analog, or derivative specifically binds to Clq-contammg immune complexes (optionally wherein the polypeptide binds to Clq-containing immune complexes containing at least one additional component, said component selected from the group consisting of C3b, iC3b, and C4b) .