JP2009165489A - Fusion protein composed of ligand bonded protein and stable plasma protein - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a new ligand bonded molecule and a receptor, and to provide a composition and a method for improving the circulating plasma half-life of the ligand bonded molecule. <P>SOLUTION: A polypeptide containing an immunoglobulin heavy chain dimer having no immunoglobulin light chain is provided. In the polypeptide, a variable domain of at least one immunoglobulin heavy chain is substituted with an amino acid sequence of a ligand bonding partner, which is a receptor, a carrier protein, a hormone, a growth factor, an enzyme or a nutrient substance and is not a lymphocyte-inducing receptor, a constituent element of an immunoglobulin gene super family, a homologous protein thereto nor a multi-subunit polypeptide encoded with an isolated gene, and the bonding partner is fused with an amino acid sequence of the immunoglobulin heavy chain constant domain at the C terminus, and holds bonding properties. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to novel ligand (ligand) binding molecules and receptors and compositions and methods for improving the circulating plasma half-life of ligand binding molecules. Specifically, the present invention relates to hybrid (hybrid) immunoglobulin molecules, methods for making and using these immunoglobulins, and nucleic acids encoding them.

  Immunoglobulins are molecules that contain polypeptide chains joined together by disulfide bonds and typically have two light chains and two heavy chains. One domain (region) (V) of each chain has a variable amino acid sequence that depends on the antibody specificity of the molecule. The other domain (C) has a fairly constant sequence common to the same class of molecules. These domains are numbered sequentially from the amino terminus.

  The immunoglobulin gene superfamily (superfamily) is composed of molecules with immunoglobulin-like domains. Members of this family include class I and class II major histocompatibility antigens, immunoglobulins, T cell receptors α, β, γ and δ chains, CD1, CD2, CD4, CD8, CD28, CD3 γ, δ. And ε chain, OX-2, Thy-1, intercellular or neuronal adhesion molecule (I-CAM or N-CAM), lymphocyte function associated antigen-3 (LFA-3), neurocytoplasmic protein (NCP-3) ), Poly-Ig receptor, myelin-binding glycoprotein (MAG), high affinity IgE receptor, major glycoprotein of peripheral myelin (Po), platelet-derived growth factor receptor, colony stimulating factor-1 receptor, macrophage Fc Receptors, Fc gamma receptors and carcinoembryonic antigens are included.

  It is known that one immunoglobulin variable domain (including the hypervariable region) can replace another immunoglobulin, and that one immunoglobulin variable domain can be replaced by a different species. EP0 173 494; EP0 125 023; Munro, Nature 312: (December 13, 1984); Neuberger et al., Nature 312 (December 13, 1984); Sharon et al., Nature 309: (May 24, 1984) Morrison et al., Proc. Natl. Acad. Sci. USA 81: 6851-6855 (1984); Morrison et al., Science 229: 1202-1207 (1985); Boulianne et al., Nature 312: 643-646 (December 1984) See 13th).

  Morrison et al., Science 229: 1202-1207 (1985) teaches the preparation of an immunoglobulin chimera having a variable region from another species fused to an immunoglobulin constant region from one species. Although this document suggests molecules having immunoglobulin sequences fused to non-immunoglobulin sequences (eg, enzyme sequences), this document only teaches immunoglobulin variable domains attached to non-immunoglobulin sequences. Morrison et al. EP 0 173 494 teaches a similar chimera. The authors use the term “receptor” and refer to “receptors such as immunoglobulins, enzymes, and membrane proteins” in the “background” section, but the “interesting receptors” “Body” includes “B cell and T cell receptors, more specifically immunoglobulins (eg, IgM, IgG, IgA, IgD and IgE, and various subtypes of individual groups)” (page 3). 10-13). The disclosure of this document is limited to immunoglobulin chimeras (see eg page 3, lines 21-30).

  It is also known that immunoglobulin variable-like domains from two components of the immunoglobulin gene superfamily (CD4 and its T cell receptor) can replace the variable domain of certain immunoglobulins (eg Capon et al., Nature 337). Traunecker et al., Nature 339: 68-70 (1989); Gascoigne et al., Proc. Nat. Acad. Sci. 84: 2936-2940 (1987); published European application EPO 0 325 224 (See A2 etc.)

  A number of proteinaceous substances have been found to function by binding specifically to target molecules. These target molecules are generally proteins, but are not necessarily proteins. Substances that bind to target molecules or ligands are referred to herein as ligand binding partners (ligand binding partners), including receptor and carrier proteins, as well as hormones, cell adhesion proteins, tissue specific adhesion factors Lectin-binding molecules, growth factors, enzymes, nutrients, etc.

  Lymphocytes are an example of cells targeted to specific tissues. Lymphocytes are mediators of normal tissue inflammation and pathological tissue damage such as occurs in rheumatoid arthritis and other autoimmune diseases. Vertebrates have evolved a mechanism that distributes lymphocytes with different antigen specificities to spatially distinct regions of the organism (Butcher, EC, Curr. Top. Micro. Immunol. 128: 85 ( Gallatin, WM et al., Cell 44: 673 (1986); Woodruff, JJ et al., Ann. Rev. Immunol. 5: 201 (1987); Duijvestijn, A. et al., Immunol. Today 10:23 (1989); Yednock, TA et al., Adv. Immunol. Printing (1989).

  This mechanism involves the continuous recirculation of lymphocytes between blood and lymphoid organs. The movement of lymphocytes between the blood with the highest mobility of lymphocytes and the lymphoid organs where lymphocytes meet isolated and processed antigens is associated with receptors on the lymphocyte surface and specialized capillaries. It begins with adhesive interactions with ligands on endothelial cells of venules (eg, high endothelial venules (HEV) and HEV-like vessels induced in chronic inflammatory synovium).

  Lymphocyte adhesion molecules are specifically called inductive (homing) receptors because they can localize or “home” these cells to specific secondary lymphoid organs.

  Candidates for lymphocyte-induced receptors have already been identified in mice, rats and humans (Gallatin, WM et al., Nature 303: 30 (1983); Rasmussen, RA et al., J. Immunol. 135: 19 (1985); Chin, YH et al., J. Immunol. 136: 2556 (1986); Jalkanen, S. et al., Eur. J. Immunol. 10: 1195 (1986)). The following references describe studies performed in this field using a monoclonal antibody called Mel14 against a murine lymphocyte surface protein: Gallatin, WM et al., Supra; Mountz, JD, etc. J. Immunol. 140: 2943 (1988); Lewinsohn, DM et al., J. Immunol. 138: 4313 (1987); Siegelman, M. et al., Science 231: 823 (1986); St. John, T. et al., Science. 231: 845 (1986).

  By immunoprecipitation experiments, this antibody recognizes ~ 90,000 dalton cell surface protein diffused on lymphocytes (Gallatin, WM et al., Supra) and ~ 100,000 dalton protein on neutrophils (Lewinsohn, DM et al., Supra) Has been shown to do.

  A partial sequence (13 residues) of a receptor to be a lymphocyte-derived receptor identified by radiolabeled amino acid sequencing of a glycoprotein defined by the Mel14 antibody has been disclosed by Siegelman et al. (Siegelman, M. et al., Science 231: 823 (1986)).

  Lectins are proteins with carbohydrate binding domains and are found in a variety of animals including humans, barnacles and fruit flies. The concept of lectin action in cell adhesion is exemplified by the interaction of certain viruses and bacteria with eukaryotic host cells (Paulson, JC, The Receptors, Volume 2, PMConn et al. (Academic Press, NY, 1985), p. 131; Sharon, N., FEBS Lett. 217: 145 (1987)). In eukaryotic cell-cell interactions, adhesive functions have been speculated for endogenous lectins in various systems (Grabel, L. et al., Cell 17: 477 (1979); Fenderson, B. et al., J. Exp. Med. 160: 1591 (1984); Kunemund, V., J. Cell Biol. 106: 213 (1988); Bischoff, R., J. Cell Biol. 102: 2273 (1986); Crocker, PR et al., J Exp. Med. 164: 1862 (1986)), which includes invertebrates (Glabe, CG et al., J. Cell. Biol. 94: 123 (1982); DeAngelis, P. et al., J. Biol. Chem. 262: 13946 (1987)) and vertebrate fertilization (Bleil, JD et al., Proc. Natl. Acad. Sci., USA 85: 6778 (1988); Lopez, LC et al., J. Cell. Biol. 101: 1501 (1985) )) Is included. The use of protein-sugar interactions as a means to achieve specific cell recognition appears to be well known.

  The literature suggests that lectins are involved in the adhesive interaction between lymphocytes and their ligands (Rosen, SD et al., Science 228: 1005 (1985); Rosen, SD et al., J. Immunol. Printing) (1989); Stoolman, LM et al., J. Cell Biol. 96: 722 (1983); Stoolman, LM et al., J. Cell Biol. 99: 1535 (1984); Ydnock, TA et al., J. Cell Bio.104 : 725 (1987); Stoolman, LM et al., Blood 70: 1842 (1987); related methods by Brandley, BK et al., J. Cell Biol. 105: 991 (1987); Yednock, TA et al., In preparation; TA et al., J. Cell Biol. 104: 725 (1987)).

  The characteristics of surface glycoproteins that may be involved in human lymphocyte induction were studied using a series of monoclonal and polyclonal antibodies commonly referred to as Hermes. These antibodies recognized a ˜90,000 dalton surface glycoprotein found in many immune and non-immune cell types and thought to be related to the Mel14 antigen by antibody precleaning experiments (Jalkanen, S. et al., Ann. Rev. Med. 38: 467-476 (1987); Jalkanen, S. et al., Blood, 66 (3): 577-582 (1985); Jalkanen, S. et al., J. Cell Biol. 105: 983-990 ( 1987); Jalkanen, S. et al., Eur. J. Immunol. 18: 1195-1202 (1986)).

  Epidermal growth factor-like domains are found in a wide range of proteins including growth factors, cell surface receptors, developmental gene products, extracellular matrix proteins, blood clotting factors, plasminogen activators and complement (Doolittle RF et al., CSH Symp. 51: 447 (1986)).

  It has been characterized that lymphocyte cell surface glycoprotein (hereinafter referred to as “LHR”) mediates lymphocyte binding to the endothelium of lymphoid tissue. Full length cDNA clones and DNA encoding human and murine LHR (HuLHR and MLHR, respectively) have been identified and isolated, and this DNA is readily expressed by recombinant host cells. The nucleotide sequence and amino acid sequence of human LHR (HuLHR) are shown in FIGS. The nucleotide and amino acid sequences of murine LHR (MLHR) are shown in FIGS. LHRs with mutated amino acid sequences or unacceptable glycosylation, and other derivatives of LHR with improved properties, including increased specific activity and improved plasma half-life, and such mutations Also provided are methods that allow for the preparation of the body.

  Herein, the LHR has the following protein domains: signal sequence, carbohydrate binding domain, epidermal growth factor-like (egf) domain, at least one, preferably two complement binding domain repeats, a transmembrane binding domain (TMD), And a glycoprotein containing a charged intracellular or cytoplasmic domain. The LHR of the present invention contains at least one of these domains, but need not contain all.

  A successful method in the development of drugs for the treatment of many congenital abnormalities related to ligand-binding partner interactions has been the identification of antagonists (antagonists) that block the interaction or binding between the ligand and the binding partner. One method is to use an exogenous binding partner as a competitive antagonist of the natural binding partner. However, many ligand binding partners are cell membrane proteins that are anchored to the lipid bilayer of the cell. The presence of membrane components is typically undesirable from a manufacturing and purification standpoint. Furthermore, since these molecules are usually present only on the cell surface, it would be desirable to produce them in a more stable form in the circulation. Also, a truncated or soluble ligand binding partner with a truncated tip will have a relatively short in vivo plasma half-life, will not cross the placenta or other biological barriers, and will simply deliver no effector function. Isolating the ligand recognition site alone would be inappropriate for therapeutic purposes and would not be optimally effective as a therapeutic agent.

  Accordingly, the object of the present invention is to provide a ligand-binding partner fused to a moiety that functions to prolong the in vivo plasma half-life of a ligand-binding partner, such as an immunoglobulin domain or plasma protein, and to facilitate purification by protein A. Is to produce. A further object is to provide hybrid immunoglobulin molecules that combine the adhesion and target-selective properties of ligand binding partners with immunoglobulin effector functions such as complement binding and cell receptor binding. Yet another object is to use molecules with novel functionality as described above for therapeutic use or as diagnostic reagents for in vitro (in vitro) assays of ligand binding partners or their targets. It is to provide. Another objective is the multifunctionality that allows multiple ligands to bind and / or activate by associating multiple ligand binding partners, each of which can be the same or different. Is to provide molecules.

  Specifically, one purpose is to bind ligand binding partners such as toxins, cell surface partners, enzymes, nutrients, growth factors, hormones, or effector molecules such as constant domain-like portions of immunoglobulin gene superfamily components, The goal is to produce a molecule for directing to the cell that will hold the ligand for that ligand binding partner, and a molecule for use in facilitating purification of the ligand binding partner.

  Another object of the present invention is to use a ligand binding partner-immunoglobulin hybrid heteromultimer, particularly heterodimer and heterotetramer, for use in inducing a therapeutic moiety against a specific tissue and ligand. Is to provide. For example, to induce CD4-IgG into tissues infected with viruses such as human immunodeficiency virus (HIV), using a hybrid immunoglobulin composed of one LHR-IgG chain and one CD4-IgG chain Can do. Similarly, molecules with a ligand binding partner-plasma protein moiety combined with a toxin-plasma protein moiety are used to deliver the toxin to the tissue of interest.

  Another object is to provide a method for expressing these molecules in recombinant cell culture.

  According to the present invention, novel polypeptides, methods for producing them, and nucleic acids encoding them are provided. These polypeptides are useful as cell surface adhesion molecules and ligands and are useful in therapeutic or diagnostic compositions and methods.

  The object of the present invention is from a ligand binding partner fused to a stable plasma protein, particularly an immunoglobulin constant domain, which can increase the in vivo plasma half-life of the ligand binding partner when present as a fusion with the ligand binding partner. This is achieved by providing a novel polypeptide. Also provided are DNA encoding the polypeptide, cultures and methods for producing the polypeptide.

  Stable plasma proteins form multi-chain polypeptides that are assembled by joining the same or different polypeptide chains, usually by disulfide bonds and / or non-covalent bonds. In this case (e.g., immunoglobulins or lipoproteins), the fusion of the invention containing a ligand binding partner may also be produced and used as a multimer having essentially the same structure as the stable plasma protein precursor. I will. These multimers can be uniform with respect to the ligand binding partner they contain, or can contain multiple ligand binding partners. In addition, these multimers may contain one or more ligand binding partner moieties.

  In a preferred embodiment in which the stable plasma protein is an immunoglobulin chain, the ligand binding partner is replaced in at least one chain, and usually the variable region of the immunoglobulin or an appropriate fragment thereof is replaced with the ligand binding partner. However, it will be understood that the invention also encompasses fusions in which the same or different ligand binding partners are substituted in multiple immunoglobulin chains. When multiple ligand binding partners are different from each other, the assembled final multi-chain polypeptide can crosslink the ligand in a manner not possible with multifunctional antibodies with natural variable regions.

  Certain multi-chain fusions belonging to this class have the variable region of one immunoglobulin chain replaced by the ligand binding region of a primary receptor such as CD4, while the variable region of another immunoglobulin chain is Displaced by the binding functionality of LHR, both immunoglobulin chains are essentially bound to each other in the usual manner.

  The fusions of the invention can be further modified by conjugation to additional therapeutic moieties such as polypeptide toxins, diagnostic labels, or other functionalities through peptide bonds or bonds generated in vitro. .

  The fusion of the present invention is produced by transforming a host cell with a nucleic acid encoding the fusion, culturing the host cell, and recovering the fusion from the culture. Also provided are vectors and nucleic acids encoding the fusions of the invention, and therapeutic and diagnostic compositions containing them.

  The present invention relates to the LHR itself as one aspect thereof. The LHR of the present invention is a full-length mature LHR having the amino acid sequence described in FIGS. 1-3 and 4-6, and a naturally occurring allele, a covalent derivative produced by derivatization in vitro, or These are predetermined amino acid sequence variants or glycosylation variants.

  The novel compositions provided by the present invention are purified and formulated in a pharmaceutically acceptable carrier for administration to a patient in need of antiviral therapy, neuromodulation therapy, or immunomodulation therapy, and cell adhesion Used for modulation. The present invention is particularly useful for the treatment of patients with receptor-mediated abnormalities. Further, the composition provided by the present invention provides an intermediate for purification of a ligand binding partner from recombinant cell culture that absorbs the fusion by using an antibody or other substance that can bind to the stable plasma protein component. Or as a diagnostic assay for ligand binding partners where stable plasma proteins serve as indirect labels.

DETAILED DESCRIPTION A ligand binding partner as defined herein is a protein known to function to specifically bind to a target ligand molecule and is generally secreted or membrane bound polypeptide in its native state. Exists as. Typical membrane-bound ligand binding partners include hydrophobic transmembrane regions or phospholipid anchors. Ligand binding partners include receptors, carrier proteins, hormones, cell adhesion proteins (proteins that direct or induce the attachment of other cells to one cell), lectin binding molecules, growth factors, enzymes, nutrients, etc. . CD antigens that are not members of the immunoglobulin gene superfamily or excluded in the above description are also suitable ligand binding partners (Knapp et al., Immunology Today 10 (8): 253-258 (1989)). Platelet growth factor receptor and insulin receptor can also be selected as ligand binding partners. Ligand binding partners include not only full-length natural types, but also amino acid sequence variants or other amino acid sequence variants that lack the tip that remains capable of binding to normal ligands.

  As used herein, the term “ligand binding partner” refers to polymorphic and non-polymorphic components of the immunoglobulin gene superfamily and proteins homologous thereto (eg, class I and class II major histocompatibility). Sex antigen, immunoglobulin, T cell receptor α, β, γ and δ chains, CD1, CD2, CD4, CD8, CD28, CD3 γ, δ and ε chains, OX-2, Thy-1, intercellular or neuronal Cell adhesion molecule (I-CAM or N-CAM), lymphocyte function associated antigen-3 (LFA-3), neurocytoplasmic protein (NCP-3), poly-Ig receptor, myelin-binding glycoprotein (MAG), high Affinity IgE receptor, peripheral glycomyelin major glycoprotein (Po), platelet derived growth factor receptor, colony stimulating factor-1 receptor, macrophage Fc receptor, Fc gamma receptor Body, such as carcinoembryonic antigen) to specifically exclude. The term homologous to a component of the immunoglobulin gene superfamily has the sequence of a component of the immunoglobulin gene superfamily for the purpose of this exclusion only, or is relative to a known component of this superfamily. Is meant to contain within it a sequence having an amino acid sequence homology equivalent to (or higher than) the amino acid sequence homology that an immunoglobulin variable or constant domain sequence has. Note, however, that this description does not exclude the manner in which the ligand binding partner fusion further aggregates with a component of the immunoglobulin gene superfamily or a fusion thereof to form a multimer.

  Also, from the term “ligand binding partner” is encoded by a discrete gene (a gene that does not encode a single-chain precursor polypeptide that leads to a multi-subunit polypeptide), and at least one subunit of the polypeptide is usually cell membrane Cellular receptors for extracellular matrix molecules (eg, integrins), exemplified by multiple subunit (chain) polypeptides inserted therein (PCT Publication WO 90/06953 (published June 28, 1990)) Are also specifically excluded. Note, however, that this description does not exclude the manner in which the ligand binding partner fusion further aggregates with a multi-subunit polypeptide or multi-subunit polypeptide fusion as defined in this paragraph to form a multimer. .

  A stable plasma protein is a protein that typically has about 30-2000 residues and exhibits increased circulation half-life (eg, about 20 hours or more) in its natural environment. Examples of suitable stable plasma proteins are immunoglobulins, albumin, lipoproteins, apolipoproteins and transferrin. Typically, the ligand binding partner is fused to the plasma protein at the N-terminus of the plasma protein or fragment thereof that can confer an increased half-life to the ligand binding partner. The ligand binding partner is generally fused to the plasma protein at its natural C-terminus. However, in some cases, a truncated form of the ligand binding partner (with the transmembrane and cytoplasmic regions deleted) is part of a stable protein that exhibits substantially hydrophobic hydropathic properties, typically about 20 residues. It fuses to the first site of the mature stable protein where a hydrophobic region having the above is observed. Such sites are present in transferrin and are also very universal in albumin and apolipoprotein and should not be difficult to identify. The remainder of the stable protein is then incorporated into the fusion as necessary to confer an increased plasma half-life on the ligand binding partner. Satisfactory results are achieved if the increase in plasma half-life of the ligand binding partner is about 100% or greater.

  In some preferred embodiments, the binding partner is LHR. LHR is defined as a polypeptide having qualitative biological activity similar to the LHR of FIGS. 1-3 or FIGS. 4-6, preferably the carbohydrate binding domain of the LHR of FIGS. 1-3 or FIGS. 4-6, epidermal growth factor Contains a domain, or a domain that is about 70% or more homologous to a carbohydrate binding domain.

  As used herein, the homology for LHR is aligned with the carbohydrate binding domain, epidermal growth in FIGS. 1-3 or FIGS. 4-6 after aligning the sequences and introducing gaps to achieve maximum homology rates if necessary. It is defined as the percentage of residues in the candidate sequence that are identical to the residues in the factor domain or complement binding domain.

  The range of LHR as a term used in this specification includes LHR having the amino acid sequence of HuLHR or MLHR described in FIGS. 1 to 3 or 4 to 6, and the same as LHR in FIGS. 1 to 3 or FIGS. Deglycosylated or non-glycosylated LHR derivatives, homologous amino acid sequence variants of the sequences of FIGS. 1-3 or FIGS. 4-6, and homologous variants and derivatives of LHR generated in vitro .

  The biological activity of the LHR or LHR fragment 1) immunologically cross-reacts with at least one epitope of LHR or 2) at least one adhesive, regulatory or effector function qualitatively equivalent to LHR It is defined as having.

  One example of qualitative biological activity of LHR is binding to ligands on specialized high endothelial cells of lymphoid tissue. Also, ligand binding often requires a divalent cation such as calcium.

  The term immunologically cross-reactive refers herein to the competitive qualitative biological activity of an LHR in which a candidate polypeptide has this activity with a polyclonal antiserum raised against a known active analog. It means that it can inhibit. Such antisera are prepared in a conventional manner by, for example, subcutaneously injecting a known active analog in complete Freund's adjuvant into a goat or rabbit, followed by intraperitoneal or subcutaneous injection with incomplete Freund. .

  Structurally, as shown in FIGS. 7-8, the LHR contains several domains, which are identified as follows (within ± 10 residues): signal sequence (residues 20-32) ), Followed by carbohydrate binding domain (identified as “lectin” domain in FIGS. 7-8) (residues 39-155), epidermal growth factor (egf) domain (residues 160-193), complement factor Binding domain (residues 197-317), transmembrane binding domain (TMD) (residues 333-355) and cytoplasmic domain (residues 356-372). The boundaries of the LHR extracellular domain are generally at the N-terminus of the transmembrane domain or within about 30 residues thereof and can be easily identified by reviewing the LHR sequence. Any or all of these domains can be used in the practice of the present invention.

  Figures 11-13 show various proteins with some homology to three of these domains. FIG. 11 shows a carbohydrate binding domain, FIG. 12 shows an epidermal growth factor domain, and FIG. 13 shows a somewhat homologous complement binding domain.

  Comparison of the amino acid sequences of HuLHR and MLHR listed in FIGS. 7-8 show that the overall sequence homology is high (˜83%). However, the degree of homology between the various domains found in HuLHR and MLHR varies. For example, the degree of sequence conservation between MLHR and HuLHR is about 83% in both the carbohydrate binding domain and the eggf domain, whereas the degree of conservation during the first complement binding iteration is reduced to 79%, There is only 63% of repeats, and the homology of the entire complement binding domain is about 71%. In addition, the two complement binding domain repeats of MLHR are identical, while differences are seen in HuLHR and also different from murine repeats. Interestingly, the degree of conservation between these two receptors in the transmembrane sequence and the surrounding region is nearly identical, perhaps with only one conservative hydrophobic substitution in the transmembrane anchor region.

  The surface glycoproteins described above that are recognized by a series of monoclonal and polyclonal antibodies commonly referred to as Hermes are excluded from the scope of the present invention.

  Immunoglobulins and certain variants thereof are known and many are produced in recombinant cell culture. For example, US Pat. No. 4,745,555; EP 256654; Faulkner et al., Nature 298: 286 (1982); EP120694; EP125503; Morrison, J. Immun. 123: 793 (1979); Kohler et al., PNASUSA 77: 2197 (1980) Raso et al., Cancer Res. 41: 2073 (1981); Morrison et al., Ann. Rev. Immunol. 2: 239 (1984); Morrison, Science 229: 1202 (1985); Morrison et al., PNASUSA 81: 6851 ( 1984); EP255694; EP266663; see WO88 / 03559. Reclassified immunoglobulin chains are also known. See, e.g., U.S. Pat. No. 4,444,878; WO88 / 03565; EP68763 and references cited therein.

  Usually, a ligand binding partner is fused at the C-terminus to the N-terminus of the constant region of an immunoglobulin to replace that variable region, although N-terminal fusion of the binding partner is also desirable. These regions or sequences of the transmembrane region or ligand binding partner containing a lipid or phospholipid anchor recognition sequence are preferably inactivated or deleted prior to fusion.

  Typically, such a fusion leaves at least a functionally active hinge, CH2 and CH3 domain of the constant region of an immunoglobulin heavy chain. The fusion may occur at the C-terminus of the Fc portion of the constant domain, or immediately N-terminal to the corresponding region of the heavy chain CH1 or light chain. Usually this is accomplished by constructing the appropriate DNA sequence and expressing it in recombinant cell culture. Alternatively, however, the polypeptides of the invention can be synthesized according to known methods. The exact site at which the fusion occurs is not critical. The specific sites are well known and can be selected to optimize the biological activity, secretion or binding properties of the binding partner. The optimal site will be determined by routine experimentation.

  In some embodiments, the hybrid immunoglobulin associates as a monomer, or heteromultimer or homomultimer (particularly a dimer or tetramer). In general, these associated immunoglobulins will have a known unit structure represented by the following schematic diagram. The basic 4-chain structural unit is the form taken by IgG, IgD and IgE. In high molecular weight immunoglobulins, four chain units are repeated. IgM generally exists as a pentamer of basic 4-chain units linked together by disulfide bonds. IgA globulins, and possibly IgG globulins, can also be present in serum in multimeric form. In the case of multimers, each 4-chain unit can be the same or different.

In the schematic diagrams herein, “A” means at least a portion of a ligand binding partner that contains a ligand binding site capable of binding to its ligand. X represents an additional agent, which may be another functional ligand binding partner (identical or different from A), a multi-subunit (chain) polypeptide as defined above (eg integrin), or It is bound to a part of an immunoglobulin superfamily component (including natural or chimeric immunoglobulin variable regions) such as a variable region or variable region-like domain, or a toxin such as Pseudomonas exotoxin or ricin, or usually a constant domain. There may be no polypeptide therapeutic agent. V _L , V _H , C _L and C _H represent immunoglobulin light or heavy chain variable or constant domains. It will be understood that these schematics are merely examples of general associated immunoglobulin structures and do not encompass all possibilities. It will be appreciated that it may be desirable, for example, to have several different “A” or “X” in any of these constructs.

  It will be understood that these schematic diagrams are merely exemplary and that the multimeric chains are thought to be disulfide bonded in the same manner as natural immunoglobulins. According to the present invention, hybrid immunoglobulins are easily secreted from mammalian cells transformed with appropriate nucleic acids. In this secreted form, the binding partner epitope is present in a heavy chain dimer, light chain monomer or dimer, and the binding partner epitope is fused to one or more light or heavy chains. Present heavy and light chain heterotetramers (including heterotetramers in which up to four, or all four variable region analogs have been replaced). When a light-heavy chain non-binding partner variable-like domain is present, a heterofunctional antibody is provided.

  By using chains or building blocks of various structures, the monomers, heteromultimers and homomultimers and immunoglobulins of the present invention can be assembled. A specific example of these basic units is shown in the following schematic diagram, showing an equivalent display for omitting the following equations.

  Various examples of associated novel immunoglobulins produced by the present invention are schematically represented below. In addition to the symbols defined above, n represents an integer and Y represents a covalent bridging moiety.

(A) _AC L;
_(B) AC L _-AC L;
_{(C) AC H - [AC} H, AC L -AC H, AC L -V H C H, V L C L -AC H, V L C L -V H C H, XC H, XC L, XC L _{-XC H, XC L -V H C} H, XC H -V L C L, XC L -AC H , or _AC L -XC _{H,];
(D) AC L -AC H -} [AC H, AC L -AC H, AC L -V H C H, V L C L -AC H, V L C L -V H C H, XC H, XC L _{, XC L -XC H, XC L} -V H C H, XC H -V L C L, XC L -AC H , or _AC L -XC _{H,];
(E) AC L -V H C} H - [AC H, AC L -AC H, AC L -V H C H, V L C L -AC H, V L C L -V H C H, XC H, _{XC L, XC L -XC H,} XC L -V H C H, XC H -V L C L, XC L -AC H , or _AC L -XC _{H,];
(F) V L C L -AC} H - [AC H, AC L -AC H, AC L -V H C H, V L C L -AC H, V L C L -V H C H, XC H, _{XC L, XC L -XC H,} XC L -V H C H, XC H -V L C L, XC L -AC H , or _AC L -XC _H,];
(G) [A-Y] _n- [V _L C _L -V _H C _H ] ₂ ;
(H) XC _H or _{XC L - [AC H, AC} L -AC H, AC L -V H C H, V L C L -AC H, XC L -AC H , or _AC L _-XC H,];
_{(I) XC L -XC H -} [AC H, AC L -AC H, AC L -V H C H, V L C L -AC H, XC L -AC H , or _AC L -XC _{H,];
(J) XC L -V H C} H - [AC H, AC L -AC H, AC L -V H C H, V L C L -AC H, XC L -AC H , or _AC L -XC _H,] ;
_{(K) XC H -V L C} L - [AC H, AC L -AC H, AC L -V H C H, V L C L -AC H, XC L -AC H , or _AC L -XC _H,] ;
_{(L) XC L -AC H -} [AC H, AC L -AC H H, AC L -V H C H, V L C L -AC H, V L C L -V H C H, XC H, XC _{L, XC L -XC H, XC} L -V H C H, XC H -V L C L, XC L -AC H , or _AC L -XC _{H,];
(M) AC L -XC H -} [AC H, AC L -AC H, AC L -V H C H, V L C L -AC H, V L C L -V H C H, XC H, XC L _{, XC L -XC H, XC L} -V H C H, XC H -V L C L, XC L -AC H , or _AC L -XC _H,];
A, X, V, or C may be modified with a covalent bridging moiety (Y) to become (AY) _n , (XY) _n, or the like.

Ligand binding partner A may be, for example, a multi-chain molecule having chains arbitrarily designated A _α and A _β . These strands are placed as a unit at the site indicated above as the single strand “A”. Fusing one of the multiple chains (with the remaining chain in a normal manner, covalently or non-covalently linked to the fusion chain), to one of the immunoglobulin chains, or Where the ligand binding partner contains two chains, one chain is independently fused to an immunoglobulin light chain and the other chain is fused to an immunoglobulin heavy chain.

  Here, it is preferred to fuse only one strand of the ligand binding partner to the stable plasma protein. In this case, a peptide bond is fused between one of the binding partner chains and a stable plasma protein, and the other chain of the ligand binding partner is naturally bound, such as disulfide bonds or hydrophobic interactions. To bind to the fusion chain. The ligand binding partner chain selected for peptide fusion should be the chain containing the transmembrane domain, and the fusion takes place essentially from the N-terminal side of the transmembrane domain, or the transmembrane and cytoplasmic domains Would be done to replace Where there are multiple transmembrane domains, one is usually selected for fusion (or fusion after deletion) and the other is left unfused or deleted.

  The basic unit with the structure outlined below is an example of what is used to create monomers, heteromultimers and homomultimers (especially dimers and trimers) with multi-chain ligand binding partners. is there.

Various examples of novel associating antibodies having a two-chain ligand binding partner (“A _α and A _β ”) used in the above unit structure are schematically shown below.

_{(N) A α A β C} L - [AC H, AC L -AC H, AC L -V H C H, V L C L -AC H, V L C L -V H C H, XC H, XC _{L, XC L -XC H, XC} L -V H C H, XC H -V L C L, XC L -AC H, AC L -XC H, A α A β C H, A α A β C L, _{A α C L -A β C H} , A β C L -A α C H, A α A β C L -V H C H, A α A β C H -V L C L, A α A β C L -XC _H or _{A α A β C H -XC L} ,];
_{(O) A α A β C} H - [AC H, AC L -AC H, AC L -V H C H, V L C L -AC H, V L C L -V H C H, XC H, XC _{L, XC L -XC H, XC} L -V H C H, XC H -V L C L, XC L -AC H, AC L -XC H, A α A β C H, A α A β C L, _{A α C L -A β C H} , A β C L -A α C H, A α A β C L -V H C H, A α A β C H -V L C L, A α A β C L -XC _H or _{A α A β C H -XC L} ,];
_{(P) A α C L -A} β C H - [AC H, AC L -AC H, AC L -V H C H, V L C L -AC H, V L C L -V H C H, XC _{H, XC L, XC L -XC} H, XC L -V H C H, XC H -V L C L, XC L -AC H, AC L -XC H, A α A β C H, A α A β _{C L, A α C L -A} β C H, A β C L -A α C H, A α A β C L -V H C H, A α A β C H -V L C L, A α A _β C _L -XC _H, or _{A α A β C H -XC L} ,];
_{(Q) A β C L -A} α C H - [AC H, AC L -AC H, AC L -V H C H, V L C L -AC H, V L C L -V H C H, XC _{H, XC L, XC L -XC} H, XC L -V H C H, XC H -V L C L, XC L -AC H, AC L -XC H, A α A β C H, A α A β _{C L, A α C L -A} β C H, A β C L -A α C H, A α A β C L -V H C H, A α A β C H -V L C L, A α A _β C _L -XC _H, or _{A α A β C H -XC L} ,];
_{(R) A α A β C} L -V H C H - [AC H, AC L -AC H, AC L -V H C H, V L C L -AC H, V L C L -V H C H _{, XC H, XC L, XC} L -XC H, XC L -V H C H, XC H -V L C L, XC L -AC H, AC L -XC H, A α A β C H, A α _{A β C L, A α C} L -A β C H, A β C L -A α C H, A α A β C L -V H C H, A α A β C H -V L C L, A _{α A β} C _L -XC _H, or _{A α A β C H -XC L} ,];
_{(S) A α A β C} H -V L C L - [AC H, AC L -AC H, AC L -V H C H, V L C L -AC H, V L C L -V H C H _{, XC H, XC L, XC} L -XC H, XC L -V H C H, XC H -V L C L, XC L -AC H, AC L -XC H, A α A β C H, A α _{A β C L, A α C} L -A β C H, A β C L -A α C H, A α A β C L -V H C H, A α A β C H -V L C L, A _{α A β} C _L -XC _H, or _{A α A β C H -XC L} ,];
_{(T) A α A β C} L -XC H - [AC H, AC L -AC H, AC L -V H C H, V L C L -AC H, V L C L -V H C H, XC _{H, XC L, XC L -XC} H, XC L -V H C H, XC H -V L C L, XC L -AC H, AC L -XC H, A α A β C H, A α A β _{C L, A α C L -A} β C H, A β C L -A α C H, A α A β C L -V H C H, A α A β C H -V L C L, A α A _β C _L -XC _H, or _{A α A β C H -XC L} ,];
_{(U) A α A β C} H -XC L - [AC H, AC L -AC H, AC L -V H C H, V L C L -AC H, V L C L -V H C H, XC _{H, XC L, XC L -XC} H, XC L -V H C H, XC H -V L C L, XC L -AC H, AC L -XC H, A α A β C H, A α A β _{C L, A α C L -A} β C H, A β C L -A α C H, A α A β C L -V H C H, A α A β C H -V L C L, A α A _β C _L -XC _H, or _{A α A β C H -XC L} ,];
The structures shown in the table above merely show important features, for example, the above structures do not show immunoglobulin binding (J) domains or other domains, nor do they show disulfide bonds. These are omitted for brevity. However, if such domains are required for binding activity, then such domains should be considered to be present in the normal position they occupy in the binding partner or immunoglobulin molecule.

If If immunoglobulin V _L V _H antibody combining site is shown above, or XC _L or XC _H is is shown in which X represents an immunoglobulin variable region, it can bind to a predetermined antigen It is preferable. Suitable immunoglobulin binding sites and fusion partners are obtained from IgG-1, -2, -3, or -4 subtypes, IgA, IgE, IgD, or IgM, with IgG-1 being preferred. The above schematic examples are representative examples of bivalent antibodies. More complex structures will be tricked by using immunoglobulin heavy chain sequences from other classes (eg, IgM).

A particularly preferred embodiment, the N- terminal portion of LHR containing binding sites for the lymphoid tissue endothelium, a fusion to C- terminal Fc portion of an antibody containing the effector functions immunoglobulin G _1. There are two preferred embodiments of this type. One is an embodiment in which the entire heavy chain constant region is fused to a part of the LHR, and the other is a sequence (heavy sequence starting in the hinge region immediately upstream of the papain cleavage site that chemically defines IgG Fc. Residue 216 (Kabat et al., “Sequences of Proteins of Immunological Interest”, 4th edition, 1987), or other immunoglobulin-like sites) is part of the LHR, with the first residue in the chain constant region as 114 It is the aspect which united. The latter embodiment is described in Example 4.

  These compositions of the present invention, particularly those in which the variable region of the immunoglobulin chain is replaced with a biologically active portion of a ligand binding partner, are believed to exhibit improved in vivo plasma half-life. These hybrid immunoglobulins are constructed in a manner similar to constructs in which immunoglobulin variable domains are replaced with immunoglobulin-like domains. For example, Capon et al., Nature 337: 525-531 (1989); Traunecker et al., Nature 339: 68-70 (1989); Gascoigne et al., Proc. Nat. Acad. Sci. 84: 2936-2940 (1987); See EPO 0 325 224 A2. The hybrid immunoglobulin of the present invention can be constructed in the same manner as a chimeric antibody in which a variable domain derived from one antibody is replaced with another variable domain. For example, EP 0 125 023; Munro, Nature 312: (December 13, 1984); Neuberger et al., Nature 312: (December 13, 1984); Sharon et al., Nature 309: (May 24, 1984); Morrison et al., Proc. Natl. Acad. Sci. USA 81: 6851-6855 (1984); Morrison et al., Science 229: 1202-1207 (1985); Boulianne et al., Nature 312: 643-646 (December 13, 1984) )checking. The DNA encoding the binding partner may be the 3 ′ end of the DNA encoding the binding partner domain of interest or its proximity, and the DNA encoding the N-terminus of the mature polypeptide or its vicinity (planning the use of different leaders). ), Or at the N-terminal coding region of the binding partner or in the vicinity thereof (when using natural signals), with a restriction enzyme. This DNA fragment can then be easily inserted into DNA encoding an immunoglobulin light chain or heavy chain constant region, and processed if necessary by deletion mutagenesis. If it is desired to utilize this variant for human in vivo therapy, it is preferably a human immunoglobulin.

  The LHR extracellular domain is generally fused at its C-terminus to an immunoglobulin constant region. The exact site at which the fusion takes place is not critical. In order to optimize the secretion and binding properties of the soluble LHR-Ig fusion, other sites near or within the extracellular region may be selected. The optimal site will be determined by routine experimentation. Fusion typically replaces the transmembrane and / or cytoplasmic domains.

  DNA encoding immunoglobulin light chain or heavy chain constant regions is known, is readily available from cDNA libraries, or can be synthesized. See, for example, Adams et al., Biochemistry 19: 2711-2719 (1980); Gough et al., Biochemistry 19: 2702-2710 (1980); Dolby et al., PNASUSA, 77: 6027-6031 (1980); Rice et al., PNASUSA 79: 7862 -7865 (1982); Falkner et al., Nature 298: 286-288 (1982); Morrison et al., Ann. Rev. Immunol. 2: 239-256 (1984).

  The DNA sequence encoding LHR is described herein. DNA sequences encoding other desirable binding partners known or readily available from cDNA libraries are also suitable for the practice of the invention.

  DNA encoding the fusion of the invention is introduced into a host cell for expression. If a multimer is desired, the host cell is transformed with DNA encoding each strand that makes up the multimer, with the host cell optimized so that the multimer strand can be assembled in the desired manner. Select If the host cell produces immunoglobulins prior to transfection, heteroantibodies can be produced simply by transfection with a binding partner fused to the light or heavy chain. Said immunoglobulin having one or more arms carrying a binding partner domain and one or more arms carrying associated variable regions, imposes dual specificity on the binding partner ligand and the antigen or therapeutic moiety. . Multiple co-transformed cells are used in conjunction with the above-described recombinant methods to produce multispecific polypeptides, such as the heterotetrameric immunoglobulins described above.

  In general, fusions are expressed in cells and are well secreted, but it has been found that there is always considerable diversity in the extent of secretion of various fusions from recombinant hosts.

  Also, methods for producing fully heteroantibodies from immunoglobulins having different specificities are known. These methods are employed for in vitro synthesis, or for the production of heterochimeric antibodies by simply substituting their binding partner-immunoglobulin chains if immunoglobulins or immunoglobulin hybrids have been used in the past.

  As an alternative to heterofunctional antibody production, a host cell (eg, a transfected myeloma) that produces a binding partner-immunoglobulin fusion can be used to secrete an antibody with a desired associated specificity for an antigen or B Fuse with cells. By recovering the heterobifunctional antibody from the culture medium of such a hybridoma, the heterobifunctional antibody can be produced somewhat more conveniently than the conventional in vitro reclassification method (EP 68763).

  The invention also encompasses amino acid sequence variants of LHR or other binding partners. Amino acid sequence variants of the binding partner are prepared with various intents, which include increased binding partner affinity for the ligand, binding partner stability, purification and preparation assistance, plasma half-life modification, therapeutic utility And the occurrence of, or reduction in the severity of, side effects when the binding partner is used therapeutically. The following discussion describes amino acid sequence variants of LHR that are typical examples of variants that can be selected as other ligand binding partners.

  Amino acid sequence variants of the ligand binding partner can be classified into one or more of the following three types: insertion variants, substitution variants, or deletion variants. These variants are usually produced by obtaining the DNA encoding the variant by site-directed mutagenesis into nucleotides in the DNA encoding the ligand binding partner and then expressing the DNA in recombinant cell culture. . However, fragments of up to about 100-150 amino acid residues can be conveniently produced by in vitro synthesis. The following discussion is partly related to LHR, but fits and gives equal effect to any ligand binding partner to the extent applicable to the structure or function of the ligand binding partner.

  Amino acid sequence variants of LHR are predetermined variants that are not found in nature or that are not found in natural alleles. LHR variants typically exhibit the same nature of biological activity (eg, ligand binding activity) as naturally occurring HuLHR or MLHR analogs. However, as long as at least one LHR epitope remains active even for LHR variants and derivatives that cannot bind its ligand, (a) as a reagent for use in a diagnostic assay for antibodies to LHR or LHR, (b) When insolubilized by a known method, as a reagent for purifying an anti-LHR antibody from antiserum or hybridoma culture supernatant, and (c) as an immunogen for extracting an antibody against LHR, or construction of an immunoassay kit Useful as an element (labeled as a competitive reagent for natural LHR, or unlabeled as a standard for LHR assays).

  The site for introducing an amino acid sequence mutation is predetermined, but the mutation itself need not be predetermined. For example, to optimize the effect of mutations at a site, random (random) or saturation mutation methods (inserting all possible 20 residues) are performed at the target codon and the expressed LHR variant is desirable Screen for optimal combination of activities. Such screening belongs to conventional methods in the field.

  Amino acid insertions will usually be on the order of about 1-10 amino acid residues. Substitutions are typically introduced for one residue. Deletions (deletions) will range from about 1 to 30 residues. Deletions or insertions are preferably made in adjacent pairs, ie deletion of 2 residues or insertion of 2 residues. It will be appreciated from the discussion below that the final construct is reached by introducing or combining substitutions, deletions, insertions and any combination thereof.

  An inserted amino acid sequence variant of an LHR is one in which one or more of the amino acid residues that are heterologous to the LHR are introduced at a predetermined site in the target LHR and a previously existing residue is moved.

  Usually, the insertion variant is a fusion of a heterologous protein or polypeptide to the amino or carboxyl terminus of LHR. Such a variant is referred to as a fusion of the LHR with a polypeptide containing a sequence other than that normally found at the position where the insertion was made in the LHR. This includes several groups of fusions.

  The novel polypeptides of the present invention are useful for diagnosis or purification of ligand binding partners by immunoaffinity techniques (which are known per se). Alternatively, in the purification of the binding partner, the novel polypeptide can be used to adsorb the fusion from an impure mixture and then elute the fusion and optionally cleave the binding partner from the fusion, eg, by enzymatic cleavage. Collect by.

  Desirable fusions of a binding partner (which may or may not be immunologically active) include a fusion of a mature binding partner sequence and a signal sequence heterologous to that binding partner.

  In the case of LHR and if desired for other selected binding proteins, signal sequence fusions are used to more quickly direct secretion of LHR. LHR is secreted when the heterologous signal replaces the natural LHR signal and the resulting fusion is recognized, i.e., cleaved upon processing by the host cell. Signals are selected based on the intended host cell, and these signals can include bacterial, yeast, mammalian and viral sequences. Natural LHR signals or herpes gD glycoprotein signals are suitable for use in mammalian expression systems.

  Substitutional variants are those in which at least one residue of the sequence of FIGS. 1-3 or FIGS. 4-6 has been removed and a different residue inserted at that position. When it is desired to finely adjust the characteristics of LHR, such substitution is generally performed according to Table 1 below.

  Novel amino acid sequences are included within the scope of the present invention, along with active center analogs (amino acids or other methods).

  Intrinsic changes in function or immunological identity can be achieved by selecting substitutions that are less conservative than those in Table 1, ie: (a) the structure of the polypeptide backbone within the substitution region (eg, sheet or helical conformation). This occurs by selecting residues that have a significantly different effect on maintaining the conformation), (b) molecular charge or hydrophobicity at the target site, or (c) bulkiness of the side chain. The substitutions that are generally considered to have the greatest change in the properties of LHR would be the following substitutions: (a) Hydrophilic side chains (eg, serine or threonine residues) and ()) hydrophobic residues (eg, : Leucine, isoleucine, phenylalanine, valine or alanine residues) (with); (b) replace (with) cysteine or proline with (with) other residues; (c) have positively charged side chains Substituting (for) negatively charged residues (eg glutamic acid or aspartic acid residues) with residues (eg lysine, arginine or histidine residues); or (d) remaining bulky side chains Substituents (eg, phenylalanine) are substituted with (for) residues that do not have a side chain (eg, glycine).

  Some of the deletions, insertions, and substitutions will not drastically change the nature of the LHR molecule. However, if it is difficult to predict the exact effect before making a substitution, deletion or insertion, for example when modifying the LHR carbohydrate binding domain or immune epitope, it may be necessary to evaluate the effect by routine screening methods. The merchant will want. For example, typically, the variant is site-directed mutagenesis into the LHR-encoding nucleic acid, expression of the variant nucleic acid in recombinant cell culture, and optionally adsorbing the variant, for example (by at least one residual immune epitope By purification from its cell culture by immunoaffinity adsorption on a polyclonal anti-LHR column. The cell lysate or purified LHR variant activity is then screened with a screening assay suitable for the desired characteristics. For example, changes in the immunological characteristics of LHR, such as affinity for certain antibodies such as Mel-14, are measured by a competitive immunoassay. As more is learned about the function of LHR in vivo, other assays will be useful for such screening. Changes in protein properties such as redox stability, thermal stability, hydrophobicity, proteolytic degradation sensitivity, tendency to associate with carriers, or tendency to associate with multimers are well known to those skilled in the art. Test with.

  Substitutional variants of LHR also include variants in which one or more of the LHR domains defined above are replaced by functionally homologous domains of other proteins by conventional methods. Where the variant is a fragment of a particular domain of LHR, it is preferred (but not essential) to be at least ˜70% homologous to the corresponding LHR domain as defined herein. As a source of such replaceable domains, one skilled in the art can use FIGS. For example, Nikuba electin, whose sequence is shown in FIG. 11, can be modified to make it at least ~ 70% homologous to the carbohydrate binding domain of LHR, replacing that domain. Similarly, coagulation factor X, whose sequence is shown in FIG. 12, can be modified to make it at least ~ 70% homologous to the LHR egf-domain, thereby replacing that domain. It may be desirable to make similar substitutions for the signal sequence, complement binding domain, transmembrane domain and cytoplasmic domain.

  Another type of LHR variant is a deletion variant. Deletion (deletion) is characterized by the removal of one or more amino acid residues from the LHR sequence. Typically, the transmembrane and cytoplasmic domain of the LHR, or only the cytoplasmic domain, is deleted. However, deletions from the LHR C-terminus to other suitable sites on the N-terminal side of the transmembrane region, such that the biological activity or immune cross-reactivity of the LHR is preserved, are appropriate. Protein digest fragments obtained so far in the process of elucidating the amino acid sequence of LHR and protein fragments having a homology of less than 70% to any of the LHR domains identified above are within the range of deletion variants. Excluded.

  Immunoglobulin fusions can be made using LHR fragments such as complement binding domains, carbohydrate domains, and epidermal growth factor domains. Complement binding domain fusions are useful for diagnosis and treatment of complement-mediated diseases and oligomerization of the fusion with other components on the LHR or lymphocyte surface.

  Deletion of cysteine or other labile residues may also be desirable, for example, in increasing the oxidative stability of LHR. Deletion or substitution of a potential proteolytic site (eg ArgArg) is achieved by deleting one of its basic residues or replacing it with a glutamine or histidine residue.

  In some embodiments, the LHR is composed of a carbohydrate binding domain without a complement binding domain and / or an eggf domain. This embodiment may or may not contain one or both of the transmembrane and cytoplasmic regions.

  Preferred types of substitution or deletion mutations are those involving the transmembrane region of LHR. The transmembrane region of the LHR subunit is a highly hydrophobic or lipophilic domain with an appropriate size to span the lipid bilayer of the cell membrane. These are thought to fix LHR in the cell membrane and allow homo- or heteromultimeric complex formation with LHR.

  Inactivation of the transmembrane domain of LHR and other binding partners in which the transmembrane domain is present, typically by deletion or substitution of transmembrane domain hydroxylated residues, reduces its cellular or membrane lipophilicity and Improved water solubility will facilitate recovery and formulation. Removing the transmembrane and cytoplasmic domains can potentially be recognized by the body's exposure to natively foreign cells that might be recognized as foreign by the body, or by insertion of a potentially immunogenic heterologous polypeptide. Introduction of immunogenic epitopes is avoided. Inactivation of the membrane-binding function can be achieved by deleting a sufficient number of residues at this site, or replacing them with a heterologous residue that achieves the same result. Achieved.

  A fundamental advantage of transmembrane inactivated LHR is that it can be secreted into the culture medium of the recombinant host. This variant is soluble in bodily fluids such as blood and does not have a lipid affinity for cell membranes that can be easily sensed, making it much easier to recover from recombinant cell culture.

  As a general proposition, it would be preferable that all variants have no functional transmembrane domain and no functional cytoplasmic sequence.

  For example, the transmembrane domain is a random or pre-determined amino acid sequence that exhibits fully hydrophilic hydropathic properties (eg, about 5-50 serine, threonine, lysine, arginine, glutamine, aspartic acid, etc.). Sequence). Similar to the deleted (truncated) LHR, these variants are secreted into the culture medium of the recombinant host.

  Examples of HuLHR amino acid sequence variants are listed in the following table. The residue following the residue number indicates the substituted amino acid or the inserted amino acid.

  These variants are preferably conservative substitutions. It will be appreciated that some variants will show a decrease or loss of biological activity. These variants are useful as a standard for LHR immunoassays as long as they retain at least one LHR immune epitope.

  Glycosylation variants are included within the scope of HuLHR. These include variants that are completely lacking in glycosylation (non-glycosylated), variants that have at least one less glycosylation site than the native (deglycosylated) and variants that have altered glycosylation. Deglycosylated and non-glycosylated amino acid sequence variants, deglycosylated and non-glycosylated LHR with native unmodified LHR amino acid sequences, and other glycosylation variants are included. For example, substitution or deletion mutation methods are used to remove the N- or O-linked glycosylation site of LHR (e.g., delete asparagine residues or with another basic residue such as lysine or histidine). Replace). Alternatively, in order to inhibit glycosylation by removing the glycosylation recognition site, the asparagine residue is left intact, but the adjacent residues that form the glycosylation site are replaced or deleted.

  Alternatively, non-glycosylated LHR having the amino acid sequence of native LHR is produced in recombinant prokaryotic cells because prokaryotes cannot introduce glycosylation into the polypeptide.

  Glycosylation variants are produced by selecting an appropriate host or by in vitro methods. For example, yeast introduces glycosylation that is significantly different from mammalian systems. Similarly, mammalian cells of different species (eg, hamsters, mice, insects, pigs, cows or sheep) or different tissue origins (eg, lung, liver, lymphoid, mesenchymal, epidermis), eg, mannose Routine screening for the ability to introduce variant glycosylation characterized by increased levels or ratio changes in mannose, fucose, sialic acid, and other sugars typically found in mammalian glycoproteins. In vitro processing of LHR is typically accomplished by enzymatic hydrolysis (eg, neuraminidase digestion).

Covalent modifications of the LHR molecule are included in the scope. Such modifications can be achieved by reacting the target amino acid residues of the recovered protein with organic derivatization reagents that can react with selected side chains or terminal residues, or functioning in selected recombinant host cells. Introduced by utilizing the mechanism of modification. The resulting covalent derivatives are useful in studies to identify residues important for the preparation of anti-LHR antibodies for biological activity, LHR immunoassay, or immunoaffinity-purification of recombinant LHR. For example, if the biological activity of the protein is completely inactivated after reaction with ninhydrin, it is suggested that at least one arginine or lysine residue is essential for its activity and then contains a modified amino acid residue By isolating peptide fragments, individual residues that have been modified under selected conditions are identified. Such modifications are routine in the art and are performed without undue experimentation.

  Derivatization with bifunctional reagents is a water-insoluble support matrix or surface for producing intermolecular assemblies of hybrid immunoglobulins and polypeptides, and for use in assaying or affinity purification of the hybrid immunoglobulins for their ligands. Useful for cross-linking. In addition, intrachain crosslinking studies will also provide direct information on conformational structures. Commonly used crosslinking reagents include 1,1-bis (diazoacetyl) -2-phenylethane, glutaraldehyde, N-hydroxysuccinimide esters (eg, esters with 4-azidosalicylic acid), disuccinimidyl esters Homobifunctional imidoesters including (example: 3,3′-dithiobis (succinimidyl-propionate)) and bifunctional maleimides such as bis-N-maleimide-1,8-octane are included. A derivatizing reagent such as methyl-3-[(p-azido-phenyl) dithio] propioimidate can provide a photoactivatable intermediate that can form a bridge in the presence of light. Alternatively, system reactions described in U.S. Pat. Nos. 3,959,080, 3,969,287, 3,691,016, 4,195,128, 4,247,642, 4,229,537, 4,065,635, and 4,3430,440. An active water-insoluble matrix, such as a sex substrate and cyanogen bromide activated carbohydrate, is used for protein immobilization and crosslinking.

  Certain post-translational derivatizations are the result of the action of recombinant host cells on the expressed polypeptide. Glutamine and asparagine residues are often post-translationally deamidated to the corresponding glutamic and aspartic acid residues. Alternatively, these residues are deamidated under mild acidic conditions. Any form of these residues is included within the scope of the present invention.

  Other post-translational modifications include hydroxylation of proline and lysine, phosphorylation of the hydroxyl group of serine or threonine residues, methylation of α-amino groups of lysine, arginine and histidine side chains (TECreighton, Proteins: Structure and Molecular Properties, WHFreeman & Co., San Francisco, pp. 79-86 (1983)), acetylation of N-terminal amines, and (in some examples) amidation of C-terminal carboxyls.

  Other derivatives consist of the novel polypeptides of the present invention covalently linked to non-proteinaceous polymers (polymers). Usually, non-proteinaceous polymers are hydrophilic synthetic polymers, ie polymers that are not found in nature. However, polymers that exist in nature and are produced by recombinant or in vitro methods are useful, as are polymers isolated from nature. Hydrophilic polyvinyl polymers such as polyvinyl alcohol and polyvinyl pyrrolidone are within the scope of the present invention. Particularly useful are polyalkylene ethers such as polyethylene glycol, polypropylene glycol, polyoxyethylene esters, methoxypolyethylene glycol; polyoxyethylene, polyoxypropylene, and block copolymers of polyoxyethylene and polyoxypropylene ( Polyoxyalkylenes such as Pluronics; polymethacrylates; carbomers; sugar monomers D-mannose, D- and L-galactose, fucose, fructose, D-xylose, L-arabinose, D -Branched or unbranched poly consisting of glucuronic acid, sialic acid, D-galacturonic acid, D-manuronic acid (eg polymanuronic acid or alginic acid), D-glucosamine, D-galactosamine, D-glucose and neuraminic acid Sugar (lactose , Including amylopectin, starch, hydroxyethyl starch, amylose, dextran sulfate, dextran, dextrin, glycogen and other homopolysaccharides and heteropolysaccharides, or polysaccharide subunits of acid mucopolysaccharides (eg hyaluronic acid); polysorbitol And sugar alcohol polymers such as polymannitol; heparin or heparon.

  If the polysaccharide is a natural glycosylation or a glycosylation that occurs concomitant with recombinant expression, the substitution site is a natural N- or N- or O-linked site introduced into the molecule. The position may be different from the O-linked glycosylation site. Mixtures of such polymers may be used and the polymer may be uniform. The polymer prior to crosslinking is preferably water soluble (although not necessary) and the final complex must be water soluble. In addition, the polymer must not be highly immunogenic in its complex form and should not be of a viscosity that is not compatible with such methods if intended for administration by intravenous infusion or injection.

  It is preferred that the polymer contains only one reactive group. This helps to avoid cross-linking of protein molecules. However, it is within the scope of the present invention to optimize reaction conditions to reduce cross-linking or to purify the reaction product through gel filtration or chromatographic sieving to recover essentially homogeneous derivatives. included.

  The molecular weight of the polymer is desirably in the range of about 100-500000, preferably about 1000-20000. The molecular weight chosen will depend on the nature of the polymer and the degree of substitution. In general, the higher the hydrophilicity of the polymer and the greater the degree of substitution, the lower the molecular weight that can be used. The optimal molecular weight will be determined by routine experimentation.

  Generally, the polymer is covalently attached to the hybrid immunoglobulin of the invention through a multifunctional cross-linking reagent that reacts with one or more amino acids or sugar residues of the hybrid and the polymer. However, direct crosslinking of the polymer by reacting the derivatized polymer with the hybrid, or vice versa, is also within the scope of the present invention.

  Covalent cross-linking sites on hybrid immunoglobulins include N-terminal amino groups, ε-amino groups present in lysine residues, and other amino groups, imino groups, carboxyl groups, sulfhydryl groups, hydroxyl groups or other hydrophilic groups. Sex groups are included. The polymer may be covalently bonded directly to the hybrid without the use of a polyfunctional (usually bifunctional) cross-linking reagent. Covalent bonds to amino groups include cyanuric chloride, carbonyldiimidazole, aldehyde reactive groups (PEG alkoxide + diethyl acetal of bromoacetaldehyde; PEG + DMSO and acetic anhydride; or phenoxide of PEG chloride + 4-hydroxybenzaldehyde, succinimidyl active esters, This is achieved by known chemistry based on activated dithiocarbonate PEG, 2,4,5-trichlorophenyl trichloroformate or p-nitrophenyl chloroformate activated PEG). The carboxyl group is derivatized by coupling PEG-amine with carbodiimide.

  Chemicals in a manner similar to that described in Bayer et al., Methods in Enzymology, 62: 310 (1979) or Heitzmann et al., PNAS, 71: 3537-341 (1984) for labeling oligosaccharides with biotin or avidin. By oxidation with (eg mesoperiodate) or an enzyme (eg glucose or galactose oxidase) (both producing aldehyde derivatives of each carbohydrate) and then reacting with hydrazide or amino derivatized polymer The polymer is attached to the oligosaccharide group. In addition, other chemical or enzymatic methods that have been used so far to bind oligosaccharides and polymers are also suitable. Substituted oligosaccharides are particularly advantageous because they generally have fewer substitutions for derivatization than amino acid sites, and thus the oligosaccharide product will be more uniform. Optionally, modifications may be made to the oligosaccharide substituents prior to polymer derivatization by enzymatic digestion to remove the sugar (eg, neuraminidase digestion).

  The polymer will carry groups that react directly with the amino acid side chains or N- or C-terminus of the polypeptides of the invention, or groups that react with multifunctional crosslinking reagents. In general, polymers having such reactive groups are known for the production of immobilized proteins. In order to use such chemistry in the present invention, water-soluble polymers should be used that have been derivatized in the same manner as insoluble polymers previously used for protein immobilization. Cyanogen bromide activation is a particularly useful method in crosslinking polysaccharides.

  The term “water soluble” with respect to the starting polymer means that the polymer used for the conjugate or its reactive intermediate is sufficiently water soluble that it can participate in the derivatization reaction.

  The term “water-soluble” with respect to a polymer complex means that the complex is soluble in physiological fluids such as blood.

  The degree of substitution by such polymers depends on whether the entire protein is used, fragments thereof, whether the protein is a fusion with a heterologous protein, the number of reactive sites on the protein, the molecular weight of the polymer , Hydrophilicity and other properties will vary depending on the particular protein derivatization site selected. Any heterologous sequence can be substituted with an essentially unlimited number of polymer molecules as long as the desired activity is not significantly adversely affected, but generally the complex contains 1-10 polymer molecules. The optimal degree of cross-linking can be readily determined by a series of experiments that determine the ability of the complex to function in the desired manner after changing the degree of substitution by varying time, temperature and other reaction conditions.

  Polymers such as PEG are cross-linked in a variety of (per se) known ways to covalently modify proteins with non-proteinaceous polymers such as PEG. However, some of these methods are not preferred for the purposes of the present invention. Cyanuric chloride chemistry causes many side reactions, including protein cross-linking. Furthermore, the inactivation of proteins containing sulfhydryl groups is particularly susceptible to hesitation. Carbonyldiimidazole chemistry (Beauchamp et al., Anal. Biochem. 131: 25-33 (1983)) requires high pH (> 8.5) and this condition can inactivate proteins. Furthermore, since the “activated PEG” intermediate can react with water, a large molar excess of “activated PEG” relative to the protein is required. The high concentration of PEG required for carbonyldiimidazole chemistry also causes problems in purification because it adversely affects both gel filtration chromatography and hydrophobic interaction chromatography. In addition, high concentrations of “activated PEG” can precipitate proteins, and the problem itself has been addressed in the past (Davis, US Pat. No. 4,179,337). On the other hand, aldehyde chemistry (Royer, US Pat. No. 4002531) is more effective because it requires only a 1 to 2 hour incubation with a 40-fold molar excess of PEG. However, the manganese dioxide proposed by Royer for the preparation of PEG aldehydes is “because PEG has a very high tendency to form complexes with metallic oxidants” (Harris et al., J. Polym. Sci., Polym. Chem. Ed.22: 341-352 (1984)) There are many problems. This problem can be avoided by using moffatt oxidation using DMSO and acetic anhydride. Furthermore, the sodium borohydride proposed by Royer must be used at high pH and has a significant tendency to reduce disulfide bonds. In contrast, sodium cyanoborohydride, which acts at neutral pH and has little tendency to reduce disulfide bonds, is preferred.

  The complex of the invention is separated from unreacted starting material by gel filtration. The heterogeneous species of the complex are purified from each other in the same way.

  The polymer may be water-insoluble as a hydrophilic gel or as a surgical tube in the form of a catheter or drainage conduit.

  DNA encoding LHR and other ligand binding partners can be synthesized in vitro or easily obtained from a lymphocyte cDNA library. Methods of synthesizing DNA encoding LHR with or without automated equipment are generally known to those skilled in the art, especially in light of the teachings herein. Examples of the current state of the art in terms of polynucleotide synthesis include Maniatis et al., Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory (1984) and Horvath et al., An Automated DNA Synthesizer Employing Deoxynu cleoside 3'-Phosphoramidites, Methods in Enzymology 154: Reference can be made to 313-326 (1987).

Alternatively, since the entire DNA sequences for the preferred embodiments are known, ie HuLHR (FIGS. 1-3) and MLHR (FIGS. 4-6), DNA encoding LHR is obtained from a murine or non-human source. In order to contain homologous sequences in a cDNA library obtained from a particular animal using labeled DNA encoding HuLHR or MLHR or fragments thereof (usually greater than about 20 bp, usually about 50 bp) To detect a clone, it is only necessary to perform a hybridization screen and then identify the full-length clone by analyzing the clone by restriction enzyme analysis and nucleic acid sequencing. If full-length clones are not present in the library, appropriate fragments are recovered from the various clones and assembled by ligating these fragments at a common restriction site. DNA encoding the LHR of other animal species can be obtained by probing a library obtained from that species with human or murine sequences or by synthesizing the gene in vitro. Other binding partner DNAs of known sequence can be obtained using similar routine hybridization methods.

  The present invention provides nucleic acid sequences that hybridize under stringent conditions with the fragments of FIGS. 1-3 or FIGS. 4-6 that are greater than about 10 bp, preferably 20-50 bp, and even greater than 100 bp. Also included within the scope of the invention are nucleic acid sequences that hybridize under stringent conditions to LHR fragments other than the signal or transmembrane domain or cytoplasmic domain.

  Further hybridize under stringent conditions with LHR cDNA or LHR genomic genes (including introns and adjacent gene or 5 ′ or 3 ′ flanking regions spanning about 5000 bp, whichever is longer) Nucleic acid probes that can be included are also within the scope of the present invention.

  Identification of genomic DNA of LHR or other binding partners is accomplished by probing a specific genomic library with a cDNA or fragment thereof labeled with a detectable group (eg, a radioactive phosphate group) and recovering clones containing the gene It's no exception. If necessary, complete genes are joined together by “walking”. Typically, such probes encode sequences that are 70% or more homologous to HuLHR or MLHR and are approximately 10-100 bp in length. Homology and size for other binding partners can be determined without undue experimentation.

  In constructing vectors useful in the present invention, prokaryotes are generally used for cloning DNA sequences. For example, Escherichia coli K12-294 (ATCC No. 31446) is particularly useful. Other microbial strains that can be used include E. coli B and E. coli X1776 (ATCC No. 31537). These examples are merely illustrative and are not intended to be limiting. Alternatively, in vitro cloning methods (eg, polymerase chain reaction) are also suitable.

  The polypeptides of the present invention may be used as N-terminal methionyl analogs or with a heterologous polypeptide / signal sequence or other polypeptide having a specific cleavage site at the N-terminus of the hybrid / portion. It is expressed directly in recombinant cell culture as a fusion of For example, in constructing a prokaryotic expression vector for LHR, a natural LHR signal is used with a host that recognizes the signal. If the secretory leader is “recognized” by the host, the host signal peptidase can cleave the fusion of the leader polypeptide fused at the C-terminus to the mature LHR of interest. For host prokaryotes that do not process the LHR signal, the signal is replaced with a prokaryotic signal selected from, for example, alkaline phosphatase, penicillinase, lpp or a thermostable enterotoxin II leader. For yeast secretion, the human LHR signal can be replaced with yeast invertase, alpha factor or acid phosphatase leader. For mammalian cell expression, other mammalian secreted protein signals such as viral secretory leaders (eg, herpes, simplex, gD signal) are also suitable, but for mammalian LHR natural signals are sufficient.

The novel polypeptides of the present invention can be expressed in any host cell, but are preferably synthesized in a mammalian host. However, host cells derived from prokaryotes, molds, yeasts, insects, etc. are also used for expression. Representative prokaryotes are suitable for cloning strain and E. coli W3110 (F ^-, λ ^-, prototrophic, ATCC No. 27325), Serratia Marusesansu (Serratia marcescans) Enterobacteriaceae, such as Bacillus and various Pseudomonas family is there. Preferably, the amount of prokaryotic enzyme secreted by the host cell should be minimal.

  Typically, an expression host is transformed with DNA encoding a hybrid linked in an expression vector. Such vectors usually retain a replication site (although this is not necessary if chromosomal integration occurs). The expression vector also contains a labeling sequence that can provide phenotypic selection in the transformed cells as discussed below. For example, E. coli is typically transformed with plasmid pBR322 (Bolivar et al., Gene 2:95 (1977)) derived from E. coli species. pBR322 contains genes for ampicillin and tetracycline resistance and therefore provides an easy means to identify transformed cells, regardless of whether the purpose is cloning or expression. It would also be optimal if the expression vector contained sequences useful for transcription and translation control (eg, promoter and Shine Dalgarno sequences (for prokaryotes) or promoters and enhancers (for mammalian cells)). The promoter may be inducible but is not necessarily required. Surprisingly, it has been found that even a strong constitutive promoter, such as the CMV promoter for mammalian hosts, produces LHR without host cell toxicity. Although the expression vector may not need to contain any expression control, replication sequences or selection genes, these deletions will prevent the identification of hybrid transformants and the achievement of high levels of hybrid immunoglobulin expression .

  Suitable promoters for use with prokaryotic hosts include, for example, β-lactamase and lactose promoter systems (Chang et al., Nature, 275: 615 (1978) and Goeddel et al., Nature, 281: 544 (1979)), alkaline phosphatase Tryptophan (trp) promoter system (Goeddel, Nucleic Acids Res. 8: 4057 (1980) and European Patent Application No. 36776), and hybrid promoters such as the tac promoter (H. de Boer et al., Proc. Natl. Acad Sci. USA, 80: 21-25 (1983)). However, other functional bacterial promoters are also suitable. Since these nucleotide sequences are generally known, one skilled in the art can use a linker or adapter to provide the necessary restriction sites and operably link them to DNA encoding LHR (Siebenlist et al. Cell, 20: 269 (1980)). Promoters used in bacterial systems will also contain a Shine Dalgarno (SD) sequence operably linked to the DNA encoding LHR.

  In addition to prokaryotes, eukaryotic microorganisms such as yeast and filamentous fungi also give good results. Saccharomyces cerevisiae is the most commonly used eukaryotic microorganism, but several other strains are also commonly available. Plasmid YRp7 is a good expression vector in yeast (Stinchcomb et al., Nature 282: 39 (1979); Kingsman et al., Gene, 7: 141 (1979); Tschemper et al., Gene, 10: 157 (1980)). This plasmid provides a trp1 gene that provides a selection marker for mutants of yeast lacking the ability to grow in tryptophan (eg, ATCC No. 44076 or PEP4-1 (Jones, Genetics 85:12 (1977))) Already contained. The presence of trp1 deficiency as a characteristic of the yeast host cell genome provides an effective environment for detecting transformation by growth in the absence of tryptophan.

  Suitable facilitating sequences for use with yeast hosts include 3-phosphoglycerate kinase (Hitzeman et al., J. Biol. Chem. 255: 2073 (1980)) or other glycolytic enzymes (Hess et al., J. Adv. Enzyme Reg. 7: 149 (1968) and Holland, Biochemistry, 17: 4900 (1978)) (eg enolase, glyceraldehyde-3-phosphate dehydrogenase, hexokinase, pyruvate decarboxylase, phosphofructokinase, glucose -6-phosphate isomerase, 3-phosphoglycerate mutase, pyruvate kinase, triosephosphate isomerase, phosphoglucose isomerase, glucokinase, etc.) are included.

  Other yeast promoters, which are inducible promoters with the additional advantage that transcription can be controlled by growth conditions, include alcohol dehydrogenase 2, isocytochrome C, acid phosphatase, degrading enzymes related to nitrogen metabolism, metallothionein, glyceraldehyde- It is a promoter region of 3-phosphate dehydrogenase and an enzyme contributing to utilization of maltose and galactose. Vectors and promoters suitable for use in yeast expression are further described in R. Hizeman et al., European Patent Application No. 73657A.

  Eukaryotic expression control sequences are known. Almost all eukaryotic genes have an AT-rich region located approximately 25-30 bases upstream from the site where transcription begins. Another sequence found 70 to 80 bases upstream from the transcription start position of many genes is a CXCAAT region (X may be any nucleotide). The 3 'end of most eukaryotic genes is an AATAAA sequence, which will be a signal for adding a poly A tail to the 3' end of the coding sequence. All of these sequences are inserted into a mammalian expression vector.

  Suitable promoters for controlling transcription from vectors in mammalian host cells include various sources such as viruses (eg, polyomavirus, SV40, adenovirus, MMV (steroid-inducible), retrovirus (eg, HIV). LTR), hepatitis B virus, and most preferably cytomegalovirus) or from a heterologous mammalian promoter (eg, beta actin promoter). The early and late promoters of SV40 are conveniently obtained as an SV40 restriction fragment that also contains the SV40 viral origin of replication (Fiers et al., Nature, 273: 113 (1978)). The immediate early promoter of human cytomegalovirus is conveniently obtained as a HindIII E restriction fragment (Greenaway, P.J. et al., Gene 18: 355-360 (1982)).

  Transcription by higher eukaryotes of DNA encoding hybrid immunoglobulins and / or hybrid portions is increased by inserting an enhancer sequence into the vector. Enhancers are cis-acting elements of DNA, usually about 10-300 bp, that act on a promoter to increase its transcription. Enhancers are relatively orientation and position independent and the transcription units 5 '(Laimins, L. et al., PNAS, 78: 993 (1981)) and 3' (Lusky, ML et al., Mol. Cell Bio., 3 : 1108 (1983)) and within the intron (Banerji, JL et al., Cell, 33: 729 (1983)) and within the coding sequence itself (Osborne, TF et al., Mol. Cell Bio. 4: 1293 (1984)). Many enhancer sequences are now known from mammalian genes (globin, elastase, albumin, α-fetoprotein, and insulin). Typically, however, one will utilize an eukaryotic virus enhancer. Examples include the late SV40 enhancer (100-270 bp) of the origin of replication, the cytomegalovirus early promoter enhancer, the polyoma enhancer late of the origin of replication, and the adenovirus enhancer.

  Expression vectors used in eukaryotic host cells (nucleated cells from yeast, mold, insects, plants, animals, humans, or other multicellular organisms) are required to terminate transcription and affect mRNA expression. It will also contain possible sequences. These regions are transcribed as polyadenyl segments in the untranslated portion of the mRNA encoding hybrid immunoglobulin. This 3 ′ untranslated region also includes a transcription termination site.

An expression vector can also contain a selection gene (also called a selectable label). Examples of selectable labels suitable for mammalian cells include dihydrofolate reductase (DHFR), thymidine kinase (TK), or neomycin. If such a selectable label is successfully introduced into a mammalian host cell, the transformed mammalian host cell can survive when placed under selective pressure. There are two different types of selection schemes that are widely used. The first type is based on cell metabolism and the use of mutant cell lines that lack the ability to grow independently of supplemental media. CHO DHFR ^- cells and mouse LTK ^- cells is a second example. These cells lack the ability to grow under conditions that do not contain nutrients such as thymidine and hypoxanthine. These cells lack some genes required for the complete nucleotide synthesis pathway and cannot survive unless the deleted nucleotides are added to the supplemented medium. Instead of supplementing the medium, intact DHFR or TK genes can be introduced into cells lacking the respective gene, thereby changing its growth requirements. Individual cells that have not been transformed with DHFR or TK genes will not be able to survive in non-supplemented media.

  The second type of selection method is dominant selection, ie a selection method that can be used for any cell type and does not require the use of mutant cell lines. This method typically uses an agent to stop the growth of the host cell. Cells that have been successfully transformed with a heterologous gene express a protein conferring drug resistance and can therefore withstand this selection method. Examples of such dominant selection include the drugs neomycin (Southern et al., J. Molec. Appl. Genet. 1: 327 (1982)), mycophenolic acid (Mulligan et al., Science, 209: 1422 (1980)), or high gloss Mycin (Sugden et al., Mol. Cell. Biol. 5: 410-413 (1985)) is used. In the above three cases, by using a bacterial gene under eukaryotic control, resistance to an appropriate agent, ie G418 or neomycin (geneticin), xgpt (mycophenolic acid), or hygromycin, respectively, is obtained. To do.

  Eukaryotic host cells suitable for hybrid immunoglobulin expression include monkey kidney CV1 strain (COS-7, ATCC CRL 1651), human embryonic kidney strain (293 or suspension culture) transformed with SV40. 293 cells subcloned for growth, Graham, FL et al., J. Gen. Virol. 36:59 (1977)), juvenile hamster kidney cells (BHK, ATCC CCL 10), Chinese hamster ovary-cell-DHFR (CHO) Urlaub and Chasin, PNAS (USA), 77: 4216 (1980)), mouse Sertoli cells (TM4, Mather, JP, Biol. Reprom. 23: 243-251 (1980)), monkey kidney cells (CV1, ATCC CCL 70), African green monkey kidney cells (VERO-76, ATCC CRL-1587), human cervical cancer cells (HELA, ATCC CCL 2), canine kidney cells (MDCK, ATCC CCL 34), buffalo Rat hepatocytes (BRL 3A, ATCC CRL 1442), human lung cells (W138, ATCC CCL75), human hepatocytes (HepG2, HB 8065), mouse mammary tumor cells (MMT060562, ATCC CCL 51), and TRI cells (Mather, JP Et al., Annals NYAcad. Sci. 383: 44-68 (1982)).

  Standard ligation techniques are used to construct a suitable vector containing the desired coding and control sequences. The isolated plasmid or DNA fragment is cut, processed and religated into the desired form to form the required plasmid.

For analysis to confirm the correct sequence in the constructed plasmid, the ligation mixture was used to transform E. coli K12 294 strain (ATCC 31446) and a successful transformant, if appropriate, ampicillin. Alternatively, selection is based on tetracycline resistance. A plasmid was prepared from the transformant by the method of Messing et al., Nucleic Acids Res. 9: 309 (1981) or Maxam et al., Methods in Enzymology, 65: 499 (1980), and subjected to restriction analysis and / or DNA sequencing. analyse.
A host cell is transformed with the expression vector of the present invention and cultured in a conventional nutrient medium modified so as to be suitable for induction of a promoter, selection of a transformant, or amplification of a gene encoding a target sequence. Culture conditions, such as temperature and pH, will be used in the past for the host cell selected for expression, and these conditions will be apparent to those skilled in the art.

  As used herein, the term host cell encompasses cells in a host animal as well as cells in in vitro culture.

The term “transformation” means the introduction of DNA into an organism such that the DNA is replicated as an extrachromosomal element or by chromosomal integration. Unless otherwise specified, the host cell transformation method used in the present invention is the method of Graham, F. and van der Eb, A., Virology, 52: 456-457 (1973). However, other methods for introducing DNA into cells, such as by nuclear injection or protoplast fusion, can also be used. A preferred transfection method when using prokaryotic cells or cells with a robust cell wall construct is the calcium chloride described by Cohen, FN et al., Proc. Natl. Acad. Sci. (USA), 69: 2110 (1972). Is a calcium treatment.

The term “transfection” means the introduction of DNA into a host cell, regardless of whether any coding sequences are actually expressed. A number of transfection methods are generally known to those skilled in the art (eg, CaPO ₄ and electroporation). Host cell transformation is evidence of successful transfection.

  The novel polypeptides of the present invention are recovered from recombinant cell culture by known methods and purified. These methods include ammonium sulfate precipitation, ethanol precipitation, acid extraction, anion or cation exchange chromatography, phosphocellulose chromatography, immunoaffinity chromatography, hydroxylapatite chromatography and lectin chromatography. Other known purification methods within the scope of the present invention use immobilized carbohydrates, epidermal growth factors or complement domains. In addition, reverse phase HPLC, as well as chromatography using hybrid immunoglobulin ligands, are also useful for hybrid purification. It is desirable to have a low concentration (about 1-5 mM) of calcium ions present during purification. It is preferred to purify the LHR in the presence of a protease inhibitor such as PMSF.

  The therapeutic use of LHR-immunoglobulin hybrids competes with the normal binding of lymphocytes to lymphoid tissue. This hybrid is therefore particularly useful for organ or transplant rejection and is particularly useful for the treatment of patients with inflammation such as rheumatoid arthritis or other autoimmune diseases. LHR-immunoglobulin hybrids can also be applied to control lymphoma metastases and treat conditions where lymphocyte accumulation occurs.

  LHR-immunoglobulin hybrid heterodimers and heterotetramers are used in directing the therapeutic moiety to lymphoid tissue. For example, by using a hybrid immunoglobulin composed of one LHR-IgG chain and one CD4-IgG chain, CD4-IgG can be induced in a tissue infected with a virus such as human immunodeficiency virus (HIV). it can. This hybrid binds not only to the endothelial tissue in the lymph nodes, but also to secondary lymphoid organs such as Peyer's plaques and the endothelial tissue in the brain, so that the blood brain barrier is used for the treatment of HIV-related dementia. It can be used to deliver CD4-IgG through. Similarly, immunoglobulin heterotetramers with LHR-lysine- or CD4-lysine-immunoglobulin as described herein are used to deliver toxins such as ricin to the tissue of interest.

  In this way, selecting a ligand binding partner with specific affinity for a particular tissue clearly increases the ability to deliver a therapeutic agent with a stable and relatively long half-life, which can lead to inconvenient experiments. Accurate processing can be done even if it is not performed.

  The novel polypeptides of the present invention are placed in sterile isotonic formulations with the necessary cofactors and optionally administered by standard means well known in the art. This preparation is preferably a liquid, usually a physiological salt solution containing 0.5-10 mM calcium, non-phosphate buffer (pH 6.8-7.6), or a lyophilized powder. Also good.

  It is believed that intravenous delivery or delivery through a catheter or other surgical tube will be the primary therapeutic route of administration. Alternative routes include commercial nebulizers for liquid formulations, such as tablets, and inhalation of lyophilized or aerosolized receptors. Liquid formulations can be used after reconstitution from powder formulations.

  The novel polypeptides of the present invention may be administered by microspheres, liposomes, other particulate delivery systems, or by sustained release formulations in certain tissues, including blood. Suitable examples of sustained release carriers include semipermeable polymer matrices in the form of molded articles (eg, suppositories or microcapsules). The microcapsule sustained-release matrix or implantable sustained-release matrix includes polylactide (US Pat. No. 3,773,919, EP 58481), a copolymer of L-glutamic acid and γ-ethyl-L-glutamate (U. Sidman). Biopolymers 22 (1): 547-556 (1985)), poly (2-hydroxyethyl-methacrylate) or ethylene vinylacetic acid (R. Langer et al., J. Biomed. Mater. Res. 15: 167-277 ( 1981) and R. Langer, Chem. Tech. 12: 98-105 (1982)). Liposomes containing hybrid immunoglobulins are prepared by well known methods (DE 3218121A; Epstein et al., Proc. Natl. Acad. Sci. USA, 82: 3688-3692 (1985); Hwang et al., Proc. Natl. Acad. Sci. USA, 77: 4030-4034 (1980); EP 52322A; EP 36676A; EP 88046A; EP 143949A; Japanese Patent Application 83-11808; U.S. Patent Nos. 4,450,454 and 4,445,545; Liposomes are usually small (about 200-800 angstroms) monolayers, and their lipid component is greater than about 30 mol% cholesterol, at a ratio adjusted to optimize polypeptide leakage rate. is there.

  Sustained release polypeptide preparations are implanted or injected near the site of inflammation or treatment (eg, arthritic joints or peripheral lymph nodes).

  As will be readily apparent to the physician, the dosage of the novel polypeptide of the present invention is determined by the nature of the hybrid used (eg, binding activity and in vivo plasma half-life), the concentration of the hybrid in the formulation, the route of administration of the hybrid, the site of administration. , Depending on the rate of administration, the clinical tolerance of the patient involved, the pathological condition afflicting the patient, etc.

  The polypeptide of the present invention can also be administered together with other medicaments (eg, antibiotics, anti-inflammatory agents, anti-tumor agents) used to treat the above-mentioned symptoms. It may also be useful to administer the polypeptides of the invention with other therapeutic proteins such as γ-interferon and other immune modulators.

  In order to facilitate understanding of the following examples, some frequently used methods and / or terms are described.

  A “plasmid” is designated by a preceding lowercase letter p and / or followed by an uppercase letter and / or a number. The starting plasmids of the present invention are commercially available and can be used by anyone without limitation, or can be constructed from available plasmids according to published methods. Furthermore, plasmids equivalent to these are also known in the art and will be apparent to the general engineer.

  Specifically, these plasmids preferably have some or all of the following characteristics: (1) contain a minimal number of host organism sequences; (2) be stable in the intended host. (3) be able to exist in a high copy number form in the target host; (4) maintain a controllable promoter; (5) move away from the portion of the plasmid where the new DNA sequence is inserted. Having at least one DNA encoding a selectable feature in the part at the selected position. It is easy for those skilled in the art to modify a plasmid to meet the above criteria, given the available literature and the teachings herein. There exist or will be discovered in the future new cloning vectors having the properties described above and therefore suitable for use in the present invention, and these vectors are also considered to be within the scope of the present invention. Should be understood.

  The term “digestion” of DNA means catalytic cleavage of the DNA by a restriction enzyme that acts only on certain nucleotide sequences in the DNA. Various restriction enzymes used in the present invention are commercially available, and reaction conditions, cofactors, and other requirements were those that would be known to those skilled in the art. For analysis purposes, typically 1 μg of plasmid or DNA fragment is used with about 2 units of enzyme in about 20 μl of buffer solution. For the purpose of isolating DNA fragments for plasmid construction, typically 5-50 μg of DNA is digested with 20-250 units of enzyme in a larger volume. Suitable buffers and substrate weights for a particular restriction enzyme are specified by the manufacturer. Usually an incubation of about 1 hour at 37 ° C. is used, but may vary according to the supplier's instructions. After digestion, the target fragment is isolated by directly electrophoresis of the reaction solution on a polyacrylamide gel.

  The size separation of the cleaved fragments is performed using an 8% polyacrylamide gel described by Goeddel, D. et al., Nucleic Acids Res. 8: 4057 (1980).

  “Dephosphorylation” means removal of terminal 5 ′ phosphate by treatment with bacterial alkaline phosphatase (BAP). This operation prevents the two restriction cleavage ends of one DNA fragment from “cyclizing” or forming a closed circle to prevent insertion of another DNA fragment into that restriction site. Dephosphorylation procedures and reagents are conventional (Maniatis, T. et al., Molecular Cloning, pages 133-134 (1982)). Reactions using BAP are performed at 68 ° C. in 50 mM Tris to inhibit exonuclease activity present in the enzyme preparation. The reaction is carried out for 1 hour. After this reaction, the DNA fragment is gel purified.

  “Oligonucleotide” means either a single-stranded polydeoxynucleotide or two complementary polydeoxynucleotide strands, which can be chemically synthesized. Since such synthetic oligonucleotides do not have a 5 ′ phosphate, they do not ligate to another nucleotide unless the phosphate is added using ATP in the presence of a kinase. Synthetic oligonucleotides can be ligated to fragments that have not been dephosphorylated.

  “Ligation” refers to the process of formation of a phosphodiester bond between two double stranded nucleic acid fragments (Maniatis, T. et al., Ibid., P. 146). Unless otherwise stated, ligation is accomplished using 10 units of T4 DNA ligase (“ligase”) per 0.5 μg of approximately equimolar amounts of DNA fragment to be ligated, using known buffers and conditions.

“Filling” or “smoothing” means an operation of converting a single-stranded end in a sticky end of a nucleic acid cleaved with a restriction enzyme into a double strand. This removes sticky ends and produces blunt ends. This step converts a restriction cleavage end that attaches to an end that is generated only by a single restriction enzyme or a few other restriction enzymes into a compatible end for any restriction endonuclease that cuts smoothly or other packed attachment ends. Is a general purpose means. Typically, 2-15 μg of target DNA is transferred to 8 units of Klenow fragment of DNA polymerase I and 4 deoxy groups at about 37 ° C. in 10 mM MgCl ₂ , 1 mM dithiothreitol, 50 mM NaCl, 10 mM Tris buffer (pH 7.5). Smoothing is accomplished by incubating with each 250 μM nucleoside triphosphate. This incubation is generally stopped after 30 minutes, phenol and chloroform extracted and ethanol precipitated.

  It is presently believed that the three-dimensional structure of the composition of the present invention is important for its function as described herein. Accordingly, all related structural analogs that mimic the active structure formed by the compositions of the present invention are specifically encompassed within the scope of the present invention.

  It will be appreciated that applying the teachings of the present invention to a particular problem or situation will be within the scope of those skilled in the art given the teachings described herein. Examples of products of the present invention, and representative methods for isolating, using and producing them are described below, but these should not be construed as limiting the present invention.

  All references to “Mel14” monoclonal antibody or “Mel14” in these examples refer to monoclonal antibodies against lymphoid surface proteins considered to be murine, as described by Gallatin et al., Supra, Nature 303: 30 (1983). However, the use of Mel14 is no longer necessary for the present invention as the complete DNA and amino acid sequences of LHR are provided herein.

Example 1: Purification and cloning of MLHR Isolation of a cDNA clone encoding MLHR MLHR was isolated from detergent-treated murine spleen by immunoaffinity chromatography using the Mel14 monoclonal antibody.

  In a typical preparation, 300 spleens from ICR female mice (16 weeks of age) were minced and then potter's in 2% Triton X-100 in Dulbecco's PBS containing 1 mM PMSF and 1% aprotinin. Homogenization was performed using a Potter-Elvehjem tissue grinder. Dissolution was continued for 30 minutes at 4 ° C. on a permeator. The lysate was centrifuged at 2000 × G for 5 minutes and then at 40000 × G for 30 minutes.

  The supernatant was filtered through a Nitex screen and then pre-purified by adsorption to rat serum (10 ml packed gel) conjugated to cyanogen bromide activated Sepharose 4B. Rat serum was diluted 1:10 for coupling with the complex according to the manufacturer's instructions. The flow through was applied to a 3 ml column of MEL-14 antibody bound to Sepharose 4B at a rate of 0.5 mg per ml. Sodium azide was added at 0.02% concentration to all column buffers.

  The column was washed with 25 ml of 2% Triton X-100 in PBS followed by 25 ml of 10 mM CHAPS in the same buffer. Antigen was released by adding 10 ml of 10 mM CHAPS in 100 mM glycine, 200 mM NaCl, pH 3, and neutralized by collecting it in 1 ml of 1 mM TRIS HCl (pH 7.6). The column was washed with 20 mM triethylamine, 200 mM NaCl, pH 11, pre-equilibrated with 10 mM CHAPS in PBS, neutralized and re-added with antigen diluted in 100 ml column buffer, and the washing and releasing procedure was repeated. .

  The purified protein was concentrated with Centricon 30 (Amicon Incorporated) and analyzed by SDS-PAGE (7.5% acrylamide) using silver staining for visualization. A typical purification yielded 30-40 μg of antigen per 300 mice based on comparison to the orosomucoid standard.

  Although no data is shown here, the polyacrylamide gel of the purified material showed a diffusion band that migrated to about 90,000 daltons and a higher molecular weight protein around 180000 daltons. The ratio of 90000 dalton component to 180000 dalton component was 10: 1 or more in all of the many preparations. This material was visualized by silver staining of a 10% polyacrylamide gel.

  Vapor phase Edman degradation of the 90000 dalton band resulted in the identification of a single N-terminal sequence (FIG. 9A) containing the N-terminal amino acid itself. N-terminal 38 amino acids with 4 spaces (X) at positions 1, 19, 33 and 34 were identified. Asparagine at position 22 by combining the deletion of the amino acid signal at position 22 with subsequent tyrosine residues (Y) and threonine residues (T) devoid of N-linked glycosylation site consensus sequences (NXT / S) Inferred (N).

  The 13 sequence residues in FIG. 9A shown above the 38-residue N-terminus were estimated in the past by Siegelman et al. (Supra) using a radiolabeled amino acid sequencing method. It shows a high degree of homology (11 out of 13 residues) with the determined LHR sequence.

  No ubiquitin sequence was obtained in any of the three sequencing experiments performed with two separate MLHR preparations. Perhaps this modification is missing in mouse splenocytes, or the N-terminus of ubiquitin is shielded against Edman degradation in LHR obtained from this source.

  The amino acid sequences of FIGS. 4-6 were compared to known sequences in the Dayhoff protein database using the algorithm of Lipman, D. et al., Science 227: 1435-1441 (1981).

  Residues from amino acids 7 to 15 underlined in FIG. 9A were selected to create the oligonucleotide probe shown in FIG. 9B. A 26-mer oligonucleotide probe degenerated 32 times from these residues was designed and synthesized on an Applied Biosystems oligonucleotide synthesizer. This probe contains all possible codon degeneracy, except for the proline at position 9, which selected the codon CCC based on mammalian codon usage rules.

A single hybridizing cDNA clone was isolated by screening a murine spleen cDNA library obtained from minced mouse spleens with this probe. As an operation, 600000 plaques derived from oligo dT-primed gt10 murine spleen cDNA library prepared from mRNA isolated from murine spleen cells treated with 5 μg / ml concanavalin A for 6 hours were used at a rate of 12 000 phage per plate. Plates were inoculated and 20% formamide, 5 × SSC (150 mM NaCl, 15 mM trisodium citrate), 50 mM sodium phosphate (pH 7.6), 5 × Denhardt's solution, 10% dextran sulfate, 20 μg / ml in denatured sheared salmon sperm DNA 42 overnight at ° C., it was hybridized with a 32-fold degenerate 26-mer oligonucleotide probe labeled with P ³² shown in FIG. 9B. These parameters are referred to herein as “strict conditions”. The filters were washed twice for 30 minutes at 42 ° C. in 1 × SSC, 0.1% SDS and autoradiographed overnight at −70 ° C. One replication positive clone was rescreened and the EcoR1 insert was isolated and inserted into the M13 or PUC 118/119 vector and its nucleotide sequence determined from the single stranded template using sequence specific primers.

  The complete DNA sequence of the 2.2 kilobase EcoR1 insert contained in this bacteriophage is shown in FIGS. The longest open reading frame begins with a methionine codon at positions 106-108. A Kozak box homology is found around this methionine codon, suggesting that this codon probably functions at the initiation of protein translation. A protein sequence containing 373 amino acids with a molecular weight of about 42200 daltons is encoded in this open reading frame. The translated protein has a sequence of residues 40 to 76 that exactly corresponds to the N-terminal amino acid sequence determined from the isolated MLHR.

This result suggests that the mature N-terminus of MLHR begins with a tryptophan residue at position 39. However, it is believed that some proteolytic processing of the actual N-terminus of LHR would have occurred during the isolation of this protein.

  The hydrophobic characterization of this protein reveals a hydrophobic domain located at the N-terminus that can serve as a signal sequence for insertion into the lumen of the endoplasmic reticulum. The deduced sequence at positions 39-333 is predominantly hydrophilic followed by a 22 residue hydrophobic domain characteristic of stop transfer or membrane anchoring domains.

  The region thought to be the intracellular region at the very C-terminus of this protein is very short, only 17 residues long. There are several basic amino acids immediately C-terminal to the predicted transmembrane domain, a property typical of the junction between the cell surface receptor membrane anchor and the cytoplasmic domain (Yarden Etc., Nature). One serine residue, a potential phosphorylation site, is present in the putative cytoplasmic domain.

  This protein contains 10 potential N-linked glycosylation sites, all of which are in domains thought to be extracellular domains. The lack of asparagine at position 60 (residue 22 of the mature protein) in peptide sequencing analysis confirms glycosylation at this site and reveals the extracellular orientation of this region. The decoding region contains a total of 25 cysteine residues, and 4 of these cysteine residues are located within the sequence considered to be the leader sequence.

Protein motifs within the MLHR As shown in FIGS. 11-13, the putative MLHR amino acid sequence is compared with other proteins in the Dayhoff protein sequence databank and the fastp program (Lipman, D. and Pearson, W., Science 227: 1435 -1441 (1985)), some interesting sequence homologies were found.

  The protein having the highest sequence homology value is shown by surrounding the region showing the maximum sequence homology with a solid line. The numbers listed at the beginning of the sequence indicate the position of these homologous sequences within the protein.

  FIG. 11 shows that the N-terminal motif of LHR (residues 39-155) has several carbohydrate binding protein homologies listed below (% homology of these sequences to MLHR) Are listed after the example column): Drikamer; amino acid residues discovered by Drickamer et al. (1), MuLHR; MLHR sequence, Hu.HepLec (27.8%); human Hepatitis lectin (2), Barn. Lec (25%); Barnacle lectin (3), Ra. HepLec (23.5%); Rat hepatitis lectin (4), Ch. HepLec (27.5%); Lectin (5), Hu. IgERec (28.6%); Human IgE receptor (6), RaHepLec2 (22.6%); Rat hepatitis lectin 2 (7), Ra.ASGRec (22.6%); Rat・ Asialoglycoprotein receptor (8), Ra IRP (25.6%); rat islet regeneration protein (9), Ra.MBP (26.1%); rat mannose binding protein (10), Ra.MBDA (26.1%); rat mannose binding Protein precursor A (11), Ra. KCBP (27%); Rat Kuppel cell binding protein (12), FlyLec (23.1%); Drosophila lectin (13), and Rab. Surf (20 .9%); rabbit lung surfactant (14).

  As can be seen from FIG. 11, the LHR motif, mostly localized at the N-terminus, shows a high degree of homology with some of the calcium-dependent animal lectins or C-type lectins (1). These include various hepatitis glycobinding proteins from chicken, rat and human, soluble mannose-binding lectin, Kuppel cell-derived lectin, asialoglycoprotein receptor, cartilage proteoglycan nucleoprotein, cardiopulmonary surfactant apoprotein and fly fly Two types of invertebrate lectins derived from barnacles are included, but are not limited to these. Not all of the “non-mutated” amino acids recognized by Drickamer and their collaborators (above) for the first time to be common to C-type animal lectins are fully conserved in the carbohydrate binding domain of MLHR. The degree of homology between these residues and other positions is obvious. Known lectins belonging to the C-type family show a range of sugar binding specificities including oligosaccharides (1) with terminal galactose, N-acetylglucosamine and mannose.

  The fact that there are many residues found to be invariant in all these carbohydrate binding proteins strongly suggests that this region functions as a carbohydrate binding domain in MLHR, and that lymphocytes are lymphoid tissue It clearly explains the observation that it can bind to the specialized endothelium in a sugar- and calcium-dependent manner. In some embodiments, the invention is practiced using the carbohydrate binding domain of LHR alone without an adjacent LHR region.

  The next motif (residues 160-193) found almost immediately after completion of the carbohydrate binding domain shows a high degree of homology to the epidermal growth factor (egf) family. FIG. 12 shows epidermal growth factor (egf) homology: MLHR; MLHR sequence, Notch (38.5%); Drosophila notch locus (15), S. purp (31.7%); Strongycenturur purpuratus egf-like protein (16), Pro.Z (34.1%); bovine protein Z (17), Fact.X (34.2%); coagulation factor X (18), Fact .VII (27.3%); Coagulation factor VII (19), Fact.IX (33.3%); Coagulation factor IX (20), Lin-12 (32.1%); Caenorhadoditis elegance ( Caenorhabditis elegans) Lin-12 locus (21), Fact. XII (26%); coagulation factor XII (22), and Mu.egf (30%); murine egg (23).

  As can be seen in FIG. 12, the maximum homology in this region of MLHR is found in the Drosophila neurogenic locus notch, but there is also considerable homology in some of the other members of this large family. Is recognized. The diversity of the location of this domain among the members of this family suggests that this region may exist in genomic fragments that can be interchanged between different proteins for different functions.

  In addition to six cysteine residues, almost all members of this family share three glycine residues. This conservation of cysteine and glycine residues is consistent with a possible structural role for this region in LHR. This domain is thought to position the carbohydrate binding region located at the N-terminus in an orientation suitable for ligand interaction. This domain is also thought to act to enhance the interaction between lymphocytes and the endothelium by binding to the egf-receptor homologue on the endothelial surface.

  The last protein motif in the extracellular region of MLHR encodes amino acids 197 to 328. This region of the glycoprotein contains two tandem repeats of a 62 residue sequence containing an amino acid motif (FIG. 13) with a high degree of homology with several complement factor binding proteins.

  FIG. 13 shows complement binding protein homology: MLHR; MLHR sequence, HuComH (31.9%); human complement protein H precursor (24), MuComH (28.9%); murine complement protein H precursor (25), HuBeta (25.6%); human β-2-glycoprotein I (26), HuCR1 (29.9%); human CR1 (27), EBV / 3d (25%) 6; Human Epstein-Barr virus / C3d receptor (28), HuC2 (27.1%); human complement C2 precursor (29), HuB (23.1%); human complement factor B (30), MuC4b (22%); murine C4b binding precursor (31), HuC1s (29.2%); human C1s zymogen (32), HuC4b (26.1%); human C4b binding protein (33), HuDAF (27. 1%); human decay-accelerating factor (34), VacSecP (26.2%); vaccinia virus secretory peptide (35).

  These proteins encoding multiple broad domains of this repeat domain include, among others, human and murine complement H precursor, human beta2 glycoprotein, Epstein-Barr virus / C3d receptor, human C4b Binding proteins, decay accelerating factors and vaccinia virus secreted polypeptides are included.

  FIG. 7C shows the homology between the two tandem repeats in MLHR and the tandem repeats found in proteins belonging to the complement-binding family. Many of the amino acids conserved in this group of complement binding proteins are also found in two repeats in this region of MLHR, including the number of conserved cysteine residues.

  Interestingly, the two repeats contained in the MLHR not only strictly replicate each other at the amino acid level, but also exhibit strict homology at the nucleotide sequence level (nucleotide residues 685-865 and 866-1056). Yes. Although this result may be due to cloning artifacts (artificial products), it may be derived from a separate cDNA library generated from the MLHR expressing cell line 38C13 (available from Stanford University (Palo Alto, California, USA)). This replication region is also found in some other isolated clones, and in the human MLHR congener (see below). In addition, several other genes (most notably the Lp (a) gene) show a higher degree of intragenic repeat conservation of this domain. These results suggest that MLHR, like other members of the complement binding family, contains multiple repeats of this binding domain.

  In conclusion, the extracellular region of MLHR appears to contain three distinct protein motifs that perform one or more new functions by binding to each other. A summary of protein motifs contained in this glycoprotein is shown in FIG.

Example 2: In general, as described in the cloning above example of HuLHR, the 2.2kbEcoR1 insert of the aforementioned murine Mel14 antigen cDNA clones were ^isolated by random primed DNA polymerase synthesis with ^{P 32} triphosphates Labeled with high specific activity, this was used to screen 600000 clones from oligo dT primed lambda gt10 cDNA library derived from human peripheral blood lymphocyte mRNA obtained from primary cells. Filters in 40% formamide, 5 × SSC (1 × SSC is 30 mM NaCl, 3 mM trisodium citrate), 50 mM sodium phosphate (pH 6.8), 10% dextran sulfate, 5 × Denhardt's solution and 20 μg / ml shear-boiled salmon sperm DNA Hybridization was performed overnight at 42 ° C. These were washed twice for 40 minutes at 55 ° C. in 0.2 × SSC, 0.1% sodium dodecyl sulfate. Twelve clones (about 1 positive per plate of 50000 phage) were selected, the largest EcoR1 insert (˜2.2 kilobases) was isolated, and its DNA sequence was determined in bacteriophage m13 using sequence specific primers. Determined by dideoxynucleotide sequencing.

  This ~ 2.2 kb clone encoded a 372 amino acid open reading frame with a molecular weight of approximately 42200 daltons starting with methionine followed by the Kozak box homology region. This encoded protein contained 26 cysteine residues and 8 potential N-linked glycosylation sites. The N-terminal highly hydrophobic region of this protein (residues 20-33) is considered the signal sequence, another 22 amino acid long highly hydrophobic region (residue 335) located at the C-terminus. ~ 357) was considered a stop transfer or membrane anchoring domain. This C-terminal hydrophobic region was followed by a charged, possibly cytoplasmic region.

  When the nucleotide sequence of this human clone was compared with the already known MLHR sequence, a high degree of total DNA sequence homology (˜83%) was observed. The relative amino acid sequence conservation of MLHR and HuLHR in each LHR domain is as follows: carbohydrate binding domain—83%; egf-like domain—82%; complement binding repeat 1--79%; complement binding Repeat 2—63%; total complement binding domain—71%; transmembrane domain—96%.

  Comparing the published Hermes sequence (Jalkanen, supra) with the HuLHR sequence of FIGS. 1-3 reveals no sequence homology.

Example 3: Expression of MLHR To conclusively verify that the murine cDNA clone isolated in the present invention encodes MLHR, this clone was inserted into an expression vector and a transient cell transfection assay. Analyzed with Expression of HuLHR was performed in the same manner.

An EcoR1 fragment containing the above open reading frame (˜2.2 kilobase EcoR1 fragment shown in FIGS. 4-6) was isolated and the pRK5 vector containing the cytomegalovirus promoter (Eaton, D. et al., Biochemistry 25 : 8343-8347 (1986); European Publication No. 307247 (published March 15, 1989)). The insert cDNA to select the plasmid containing the correct orientation relative to the promoter, was introduced into 293 human embryonic kidney cells using CaPO ₄ precipitation.

Two days later, cells were incubated with 500 μCi each of S ³⁵ cysteine and methionine. Lysates and supernatants were prepared as previously described (Lasky, L. et al., Cell 50: 975-985 (1987)) and sandwiched between Mel14 monoclonal antibody and protein A sepharose. Immunoprecipitation with Mel14 monoclonal antibody (purified by immunoaffinity chromatography) was performed by using IgG polyclonal antibody.

At the same time, cells known to express MLHR, B-cell lymphoma 38C13, are metabolically labeled with methionine or cysteine to analyze supernatant MLHR, or to analyze LHR bound to cells. the cell surface glycoproteins were labeled with I ¹²⁵ and lactoperoxidase was analyzed by Mel14 antibody immunoprecipitation.

  The resulting immunoprecipitate was analyzed on a 7.5% polyacrylamide SDS gel and autoradiographed overnight at -70 ° C.

The results of these tests are shown in FIG. The meaning of lanes A to F in this figure will be described next:
A. Lysate of 293 cells transfected with MLHR expression plasmid immunoprecipitated with Mel14 monoclonal antibody;
B. Supernatant of 293 cells transfected with MLHR expression plasmid immunoprecipitated with Mel14 monoclonal antibody;
C. Lysate of 293 cells transfected with a plasmid expressing HIV gp120 coat glycoprotein immunoprecipitated with Mel14 monoclonal antibody;
D. Supernatant of 293 cells transfected with HIV coat plasmid, immunoprecipitated with Mel14 monoclonal antibody;
E. 38C13 cell supernatant immunoprecipitated with Mel14 monoclonal antibody;
F. Surface labeled with I ^125, 38C13 cell lysate immunoprecipitated with Mel14 monoclonal antibody.

As can be seen from FIG. 10, cells transfected with this construct produce two types of cell binding proteins that react specifically with the Mel14 antibody. These cell binding proteins migrate to about ˜70,000 and ˜85,000 daltons, suggesting that the ˜42200 dalton nucleoprotein becomes glycosylated in the transfected cells. The sialidase treatment changes the molecular weight of the larger band (data not shown), suggesting that the protein is a relatively mature glycoprotein, while the lower molecular weight band Resistance to this enzyme indicates that the protein may be in the precursor form.

  FAC analysis of the transiently transfected cell line with Mel14 antibody revealed that some of the LHR expressed in these cells was detectable on the cell surface (data not shown).

  Higher molecular weight glycoproteins produced in this transfection cell line are the results obtained with other transfection cell lines, ie those produced by peripheral lymph node-induced B-cell lymphoma 38C13 (FIG. 10 lane F). ) Was found to be slightly smaller. This may be due to cell-specific differences in glycosylation.

  Interestingly, both 38C13 cells and transfected human cells appear to release lower molecular weight forms of MLHR into the medium (Figure 10 lanes B and E). The nature of this released molecule is not clear, but its low molecular weight suggests that this is a cell surface type cleavage product produced by proteolysis near the membrane anchor.

  In conclusion, these results clearly demonstrate that the cDNA clones we isolated encode MLHR.

Example 4: Construction, purification and analysis of a truncated MLHR-IgG chimera FIG. 15 shows the construction of MLHR-IgG chimeras containing lectin, lectin-egf and lectin-egf-complement regulatory motifs. Yes. The top of the figure shows the N-terminal signal sequence (SS), lectin domain, epidermal growth factor (egf) domain and overlapping complement regulatory domain (CDB) along with a transmembrane anchor domain (TMD) and a short cytoplasmic sequence. Fig. 4 shows the protein domain of the murine lymphocyte-derived receptor (MLHR) containing. Three leading deletion MLHR-IgG chimeras containing lectin (MLHR-L + IgG), lectin and egf (MLHR-LE + IgG), and lectin, egf and two complement regulatory motifs (MLHR-LEC + IgG) are also shown in FIG. Is shown in All of these truncated proteins have hinge domains (H) such that these chimeras, together with CH2 and CH3 constant regions, contain the two cysteine residues (C) of the hinge that contribute to immunoglobulin dimerization. ) In the human heavy chain gamma 1 region immediately upstream. A human heavy chain IgG1 constant region cassette characterized in the past (Caponet et al., Supra (1989)) was used. The junction site between the LHR sequence and the human IgG sequence was selected so that the chimeric molecule was efficiently synthesized and dimerized under conditions where no light chain was produced by the binding of these molecules near the hinge region. Furthermore, the use of human IgG1 constant regions can avoid problems due to cross-reactivity with endogenous murine IgGs in immunohistochemical experiments described below.

  As can be seen from FIG. 16, these chimeras were efficiently synthesized and secreted in these transient transfection assays. The reactivity of these chimeras with protein A sepharose under conditions where no antibody is added demonstrates that these constant region domains are normally folded. FIG. 16 shows that these molecules dimerize under non-reducing conditions, demonstrating that the hinge region is fully functional in these chimeras. Finally, protein A reactivity also allows for nearly homogeneous purification of these chimeras on protein A sepharose columns. The results of this example demonstrate the production of antibody-like substances where the constant domain is derived from human IgG gamma 1 heavy chain while the “variable” domain is derived from MLHR.

Construction of Chimera Previously described MLHR-PRK5 expression plasmid (Eaton et al. (1986); Lasky et al., Cell 50: 975-985 (1987)) and cDNA copy of human heavy chain IgG (Capon et al., Nature 337: 525- Starting from 531 (1989), a 1100 bp HindIII fragment encoding the CH1-CH3 region of the human IgG1 constant region was inserted 3 ′ of the polyA site of the MLHR cDNA. After converting this plasmid to a single-stranded template using the m13 origin of replication and KO7 helper phage, the lectin, egf and the second complement-binding repeat and hinge on the N-terminal side of the region presumed to be the transmembrane fixed region The region between and was looped out with a 48-mer oligonucleotide by in vitro mutagenesis (Zoller and Smith (1982)). The resulting mutants were screened with a 21 P oligonucleotide labeled with ³² P across the deletion junction and the isolated mutants were sequenced using supercoil sequencing.

Correct mutants were tested for expression by transfection of human kidney 293 cells using previously described methods. ^The supernatant labeled with ³⁵ S methionine and cysteine was analyzed by immunoprecipitation with protein A sepharose beads under the condition where no antibody was added. Precipitated proteins were analyzed on 7.5% polyacrylamide-SDS gels with or without β-mercaptoethanol reduction. A plasmid harboring a correctly expressed chimera was introduced into 293 cells by transfection in the presence of a selection plasmid encoding resistance to G418 and dihydrofolate reductase. Clones were selected in G418 and the inserted plasmid was amplified in the presence of methotrexate. Permanent cell lines expressing high levels of each construct were grown in large quantities in T-flasks and the cell supernatant was clarified by centrifugation and filtration. The resulting supernatant was concentrated by Amicon filtration, passed through a standard protein A sepharose column, washed with PBS and eluted with 0.1 M acetic acid, 0.15 M NaCl (pH 3.5). The eluted material was immediately neutralized with 3M Tris (pH 9) and quantified by SDS gel electrophoresis and ELISA assay.

  The result of gel electrophoresis described in the previous paragraph is shown in FIG. The reduced protein is shown in lanes A to F, the non-reduced protein is shown in lanes GI, and the purified protein is shown in lanes J to L. The molecular weight of the marker is shown in kilodaltons. The identification of each lane is described below: Secreted MLHRLEC-IgG, B. Intracellular MLHRLEC-IgG, C.I. Secreted MLHRLE-IgG, D. Intracellular MLHRLE-IgG, E. coli. Secreted MLHRL-IgG, F.I. Intracellular MLHRL-IgG, G. Secreted MLHRLEC-IgG, H.I. Secreted MLHRLE-IgG, I.V. Secreted MLHRL-IgG, J. Purified MLHRLEC-IgG, K.I. Purified MLHRLE-IgG, L. Purified MLHRL-IgG.

  Isolated LHR-IgG chimeras were quantified using an ELISA consisting of wells coated with anti-human IgG1-specific mouse monoclonal antibodies. An unknown sample and highly purified human CD4-IgG1 immunoadhesin standard is incubated with the antibody-coated plate, the plate is washed, and the bound material is combined with horseradish peroxidase to goat anti-human IgG1 And then further washing and adding the substrate. This quantitative assay allowed measurement of sub-nanogram quantities of isolated LHR-IgG chimeras.

Analysis of MLHR-IgG chimeric PPME reactivity by ELISA The ability of various IgG chimeras to recognize yeast cell wall carbohydrates, polyphosphomannan esters or PPME was analyzed by ELISA in a previously described format (Imai et al. (1989)). . Briefly, about an equal volume of purified chimera is coated in microtiter wells overnight at 4 ° C. After blocking non-specific sites with BSA, the bound antigen was reacted with 5 μg / ml PPME solution. Bound carbohydrate was detected using a polyclonal antibody against this and a standard (Vector) immunohistochemical staining reagent. Inhibition by Mel14 was performed by preincubating MLHRLEC-IgG containing wells with this monoclonal antibody prior to addition of PPME, while addition of 10 mM EGTA during the binding reaction, to induce this induced receptor-carbohydrate interaction. Calcium dependence was proved. As an assay for inhibition, various other additives were added prior to PPME incubation. After 1 hour at 22 ° C, the plates were washed and incubated with a rabbit polyclonal antibody against PPME for 1 hour at 22 ° C. Plates were washed and incubated with vector ABC-AP for 30 minutes, washed and developed. The resulting assay was measured with a plate reader. The carbohydrate used in the inhibition assay was obtained from Sigma Chemical Co., St. Louis, MO.

  The result of the PPME binding analysis is shown in FIG. Each lane contains the following MLHR-IgG chimera: Binding of PPME to MLHRL-, MLHRLE- and MLHRLEC-IgG chimeras; Inhibition of MLHRLEC-IgG-PPME binding by Mel14 monoclonal antibody and EGTA, C.I. Inhibition of MLHRLEC-IgG-PPME binding by other carbohydrates.

  Previous studies have shown that LHR can bind yeast cell wall mannans, phosphomannan esters or PPMEs (Yednock et al., J. Cell Biol. 104: 725-731 (1987)) and this binding is It has been demonstrated to inhibit the ability of human lymph nodes to adsorb to high endothelial vesicles, consistent with the hypothesis that peripheral lymph node LHR lectin domains will recognize carbohydrates on peripheral lymph node endothelium. Yes. It was further discovered that the Mel14 antibody inhibits PPME binding to the lymphocyte surface (Yednock et al., Supra (1987)), consistent with the understanding that this carbohydrate binds within the lectin domain of the peripheral lymph node LHR. I'm doing it.

  Chimeras containing lectins, egf and overlapping complement-binding repeat structures were found to bind PPME. This binding can be inhibited by the Mel14 antibody, which is consistent with data demonstrating that the MLHRLEC-IgG chimera is recognized by this antibody (data not shown). This binding is also quantitatively equivalent to that previously observed using MLHR isolated from splenocytes (Imai et al., In press (1989)), indicating that this is on the lymphocyte surface. This suggests the same protein-carbohydrate interaction observed in LHR (Yednock et al., Supra (1987)). Furthermore, this binding was also found to be calcium dependent (Stoolman and Rosen, J. Cell Biol. 96: 722-729 (1983)), as already shown for lymphocyte binding receptors. , Suggesting that C-type or calcium-dependent lectin domains (Drickamer, J. Biol. Chem. 263: 9557-9560 (1988)) contribute at least in part to this interaction (FIG. 16b).

  Previous studies have demonstrated that various carbohydrates other than PPME can be recognized by MLHRs derived from the spleen (Yednock et al., Supra (1987); Imai et al., Supra (1989)). These include fucoidins, dextran sulfate and sulfatides from brain. To compare the ability of these carbohydrates to inhibit the interaction of MLHRLEC-IgG chimera with PPME, the specificity of this molecule with previously described spleen-derived glycoproteins (Imai et al., Supra (1989)). Tested. As can be seen from FIG. 16, fucoidin, dextran sulfate and sulfatide can all inhibit the interaction of PPME and MLHRLEC-IgG, indicating that the carbohydrate specificity of this protein induced by recombinant methods is naturally reduced. It is very similar to what has been described in the past for the proteins present in. Not being inhibited by the other two charged carbohydrates, chondroitin sulfate and heparin, this inhibition is due to specific carbohydrate recognition and due to non-specific interactions due to the highly charged nature of these compounds. It suggests that it is not a thing.

Cell block assay with MLHR-IgG chimera Previously described (Geoffrey and Rosen, J. Cell Biol., In press (1989)) using Peyer's plaques and cryostat sections of mouse peripheral lymph nodes, stamp fur A Woodferf (Wampruff) cell block assay (Stamper and Woodruff, J. Exp. Med. 144: 828-833 (1976)) was performed. Briefly, frozen tissue sections were incubated with mesenteric lymphocytes in the presence of MLHR-IgG chimera, isolated spleen-derived MLHR or buffer alone. MLHR-IgG chimera was added at a concentration on the order of 10 μg / section and pre-incubated on frozen sections before the addition of 1 × 10 7 cells / ml. The slides were washed and lymphocyte adhesion was measured by digital biomorphometry as the number of lymphocytes bound to HEV in these lymphoid organs per unit area.

  Although data are not shown, MLHRLEC-IgG chimera inhibits lymphocyte binding to peripheral lymph node HEV at a level of about 75% inhibition, whereas in this assay, spleen-derived MLHR blocks at a level of about 50%. I understood. This inhibition was calcium dependent and was blocked by the addition of MEL14 monoclonal antibody (data not shown).

Immunohistochemical analysis of MLHR-IgG chimeras Isolated MLHR-IgG chimeras were used for immunohistochemical experiments using the same method as used for monoclonal antibodies. 8-10 micron tissue sections were cut in a cryostat and fixed with 0.1 M cacodylate, 1% paraformaldehyde for 30 minutes at 4 ° C. The sections were washed in Dulbecco's PBS and stained with various amounts of MLHR-IgG chimera in 5% normal mouse serum for 30 minutes at 4 ° C. These sections were then washed and incubated with a second stage containing a biotinylated goat anti-human Fc specific antibody (vector). Endogenous peroxidase was removed by treating the sections with hydrogen peroxide-methanol after the addition of the second stage reagent and before the addition of the vector ABC complex. Sections were washed and incubated with substrate (AEC) for 5-10 minutes. Finally, the sections were counterstained with aqueous hematoxylin (Biomedia) and examined with a Zeiss Axioplot.

  In these immunohistochemical analyzes of the three MLHR-IgG chimeras, peripheral lymph nodes were used as tissue source. The selection of peripheral lymph nodes as a histological source has been demonstrated by many previous literatures that lymphocytes bind to HEV of this lymphoid tissue in a manner that can be blocked by Mel14, This suggests that the maximum level of ligand recognized by MLHR should be present in this tissue (Gallatin et al., Nature 304: 30-34 (1983)). MLHRLEC-IgG chimera was able to stain peripheral lymph node HEV. This staining was found exclusively on high walled endothelial cells and no extraluminal or near luminal areas were stained. Furthermore, this staining could be blocked by MEL14 antibody and was dependent on the presence of calcium. This suggests that MLHRLEC-IgG binding to peripheral lymph node HEV is similar to adhesion between lymphocytes and HEV. Consistent with the PPME binding data, staining of peripheral lymph node HEV with MLHRLEC-IgG can be inhibited by fucoidin and dextran sulfate (FIG. 10), whereas chondroitin sulfate and simple mannans cannot inhibit this staining reaction. (Data not shown). This also suggests that this staining reaction is due to recognition of carbohydrate ligands expressed on this peripheral lymph node HEV. These data demonstrate that this type of immunohistochemical reagent can be used to examine the tissue distribution of endothelial molecules that can interact with peripheral lymph node LHR.

MLHR ligand is present in Peyer's plaque Although not shown here, as a result of immunohistochemical assay, we unexpectedly found that the MLHRLEC-IgG chimera can specifically recognize the endothelium of Peyer's plaque I discovered that. This chimera appears to stain the high wall endothelium of Peyer's plaques containing lymphocytes. This staining could be inhibited by MEL14 antibody and was also calcium dependent. Interestingly, Peyer's plaque HEV staining appeared somewhat weaker compared to peripheral lymph node HEV staining, suggesting lower levels of MLHR ligand expressed in this lymphoid organ. Yes. These results indicate that although other adhesion systems may also be involved in this organ (Holzman et al., Cell 56: 37-46 (1989)), a ligand for peripheral lymph node LHR is expressed, and thus this ligand is It has been demonstrated to be involved in the binding of lymphocytes to the endothelium of the organs.

Example 5: Construction of CD4-IgG-MLHR-IgG chimera Two previously constructed PRK plasmids were used for direct expression of MLHR-IgG and human CD4-IgG. The MLHR plasmid is as described in the above example. The CD4-Ig plasmid is the one described in Capon et al. (Supra), modified by deleting the C _H1 domain and the region encoding the hinge region part up to the first cysteine residue. These plasmids using standard calcium phosphate method described above, selecting two genes together with PSV ^{T Antigen} to create a temporary expressing cells at a high level, or two genes of cell clones stably expressing Co-transfected into human 293 cells with PSV ^neo to confer neomycin resistance. Expression was analyzed by radioimmunoprecipitation; since CD4-IgG, LHR-IgG and CD4-IgG-LHR-IgG all contain an IgG Fc portion, they can all be directly precipitated by protein A in a standard manner . Three types of molecules were detected: CD4-IgG homodimer, LHR-IgG homodimer and CD4-IgG-LHR-IgG heterodimer. These molecules separate into their respective monomer components upon reduction, indicating that the components of each dimer are covalently linked to each other by disulfide bonds, including heterodimers. .

References for Examples
1. Drikamer, K., J. Biol. Chem. 263: 9557 (1988); Drickamer, K., Kidney Int. 32: S16
7 (1987)
2.Spiess, M. et al., Proc. Natl. Acad. Sci., USA 82: 6465 (1985)
3.Muramoto, K. et al., Biochem Biophys. Acta 874: 285 (1986)
4. Leung, J. et al., J. Biol. Chem. 260: 12523 (1985); Holland, E. et al., Proc. Natl. Acad. Sc
i., USA 81: 7338 (1984)
5.Drikamer, K, J. Biol. Chem. 256: 5827 (1981)
6.Kikutani, H. et al., Cell 47: 657 (1986)
7.McPhaul, M. et al., Molec. Cell. Biol. 7: 1841 (1987)
8.Halberg, D. et al., J. Biol. Chem. 262: 9828 (1987)
9.Terzaono, K. et al., J. Biol. Chem. 263: 2111 (1988)
10.Drickamer, K. et al., J. Biol. Chem. 262: 2582 (1987)
11.Drikamer, K. et al., J. Biol. Chem. 261: 6878 (1986)
12.Hoyle, G. et al., J. Biol. Chem. 263: 7487 (1988)
13.Takahashi, H. et al., J. Biol. Chem. 260: 12228 (1985)
14.Boggaram, V. et al., J. Biol. Chem. 263: 2939 (1988)
15.Kidd, S. et al., Mol. Cell. Biol. 6: 3094 (1986)
16.Hursh, D. et al., Science 237: 1487 (1987)
17.Hojrup, P. et al., FEBS Lett. 184: 333 (1985)
18.Fung, M. et al., Nucl. Acids Res. 12: 4481 (1984)
19.Takeya, H. et al., Proc. Natl. Acad. Sci., USA 76: 4990 (1979)
20.McMullen, B. et al., Biochem. Biophys. Res. Commun. 115: 8 (1983)
21.Greenwald, I., Cell 43: 583 (1985)
22.Cool, D. et al., Biol. Chem. 260: 13666 (1985)
23.Gray, A. et al., Nature 303: 722 (1983)
24.Schlz, T. et al., Eur. J. Immunol. 16: 1351 (1986)
25.Kristensen, T. et al., Proc. Natl. Acad. Sci., USA 83: 3963 (1986)
26.Lozier, J. et al., Proc. Natl. Acad. Sci., USA 81: 3640 (1984)
27 Bentley, D. Biochem. J. 239: 339 (1986)
28. Moore, M. et al., Proc. Natl. Acad. Sci., USA 84: 9194 (1987)
29.Bentley, D. et al., Proc. Natl. Acad. Sci., USA 81: 1212 (1984)
30.Mole, J. et al., J. Biol. Chem. 259: 3407 (1984)
31.DiScipio, R. et al., J. Biol. Chem. 263: 549 (1988)
32.Kusomoto, H. et al., Proc. Natl. Acad. Sci., USA 85: 7307 (1988)
33.Lintin, S. et al., FEBS Lett. 204: 77 (1986)
34.Caras, I. et al., Nature 325: 545 (1987)
35. Kotwal, G. et al., Nature 335: 176 (1988).

FIG. 1 represents the amino acid sequence and DNA sequence of human LHR (HuLHR). FIG. 2 is a continuation of FIG. FIG. 3 is a continuation of FIG. FIG. 4 represents the amino acid sequence and DNA sequence of murine LHR (MLHR). FIG. 5 is a continuation of FIG. FIG. 6 is a continuation of FIG. FIG. 7 shows a comparison between the amino acid sequences of mature HuLHR and MLHR. FIG. 8 is a continuation of FIG. Figures 9A-9B show the results of MLHR isolation and N-terminal sequencing. FIG. 9A represents the gas phase Edman degradation of a 90000 dalton band obtained from an SDS-polyacrylamide gel of material purified by Mur 14 monoclonal antibody affinity chromatography from a detergent extract of murine spleen. The underlined residues of amino acids 7-15 were selected to create the oligonucleotide probe shown in FIG. 9B. FIG. 9B shows a 26-mer (26-mer) nucleotide probe degenerate 32 times. FIG. 10 represents the transient expression of the MLHR cDNA clone. Lanes A to F will be described next. A: Lysate of 293 cells transfected with MLHR expression plasmid immunoprecipitated with Mel14 monoclonal antibody. B: Supernatant of 293 cells transfected with MLHR expression plasmid immunoprecipitated with Mel14 monoclonal antibody. C: Lysate of 293 cells transfected with a plasmid expressing HIV gp120 coat glycoprotein immunized with Mel14 monoclonal antibody. D: Supernatant of 293 cells transfected with HIV coat expression plasmid immunoprecipitated with Mel14 monoclonal antibody. E: Supernatant of 38C13 cells immunoprecipitated with Mel14 monoclonal antibody. ^F: surface labeled with ^{I 125,} and immunoprecipitated with Mel14 monoclonal antibody, 38C13 cells lysates. FIG. 11 shows a protein sequence that is heterologous to MLHR but functionally similar. The column labeled “MLHR” corresponds to the MLHR in FIGS. FIG. 11 compares carbohydrate binding domains. FIG. 12 shows a protein sequence that is heterologous to MLHR but functionally similar. The column labeled “MLHR” corresponds to the MLHR in FIGS. FIG. 12 compares the epidermal growth factor domains. FIG. 13 shows a protein sequence that is heterologous to MLHR but functionally similar. The column labeled “MLHR” corresponds to the MLHR in FIGS. FIG. 13 compares complement binding factor domains. FIG. 14 is a schematic diagram of protein domains found in LHR, which includes a signal sequence, carbohydrate binding domain, epidermal growth factor (egf) domain, two complement binding domain repeats (arrows), transmembrane binding domain. (TMD), and charged intracellular domains are included. FIG. 15 represents the construction of MLHR-IgG chimeras containing lectin, lectin-egf and lectin-egf-complement regulatory motifs. FIG. 16 shows gel electrophoresis of expression products and purified products of MLHR-IgG chimeras. FIG. 17 shows polyphosphomannan ester (PPME) binding analysis of various MLHR-IgG chimeras.

Claims (21)

  A polypeptide comprising an immunoglobulin heavy chain dimer without an immunoglobulin light chain, the receptor, a carrier protein, a hormone, a growth factor, an enzyme or a nutrient, comprising a lymphocyte-derived receptor, an immunoglobulin The amino acid sequence of a ligand binding partner that is not a component of the gene superfamily, a homologous protein, or a multi-subunit polypeptide encoded by a separate gene replaces at least one variable region of an immunoglobulin heavy chain A polypeptide whose binding partner is fused to the amino acid sequence of an immunoglobulin constant region at the C-terminus and retains binding properties.   The polypeptide of claim 1, wherein the immunoglobulin heavy chain dimer retains at least a functionally active hinge, CH2 and CH3 domain of an immunoglobulin heavy chain constant region.   The polypeptide of claim 2, wherein the polypeptide is deficient in the CH1 domain.   The polypeptide according to any one of claims 1 to 3, wherein the ligand binding partner is a receptor.   The polypeptide according to any one of claims 1 to 4, wherein the immunoglobulin heavy chain is obtained from IgG1, IgG2, IgG3, IgG4, IgA, IgE, IgD, or IgM.   The polypeptide according to claim 5, wherein the immunoglobulin heavy chain is a constant domain of an IgG heavy chain.   The polypeptide according to any one of claims 1 to 6, wherein the immunoglobulin heavy chain is human.   The polypeptide according to any one of claims 1 to 7, wherein the ligand binding partner is human.   The polypeptide according to any one of claims 1 to 8, wherein the ligand binding partner is a single chain. 10. A polypeptide fusion wherein the ligand binding partner is a first ligand binding partner, wherein the fusion further comprises an additional fusion of a second ligand binding partner and an immunoglobulin. Polypeptide fusion.   11. The polypeptide fusion of claim 10, wherein the second ligand binding partner is different from the first ligand binding partner.   12. A polypeptide according to any one of claims 1 to 11 in a physiologically acceptable carrier.   The polypeptide according to claim 12, wherein the carrier is a sterile isotonic solution.   A nucleic acid encoding the polypeptide according to any one of claims 1 to 11.   A replicable expression vector comprising the nucleic acid according to claim 14.   A composition comprising cells transformed with the expression vector according to claim 15.   The composition according to claim 16, wherein the cell is a mammalian cell.   18. The composition of claim 17, wherein the cell is a Chinese hamster ovary cell line. The method for culturing a cell according to any one of claims 16 to 18, comprising culturing the transformed cell and recovering the polypeptide from the cell culture.   20. The method of claim 19, wherein the polypeptide is recovered from the host cell.   21. The method of claim 20, wherein the polypeptide is secreted into the culture medium and recovered from the culture medium.

Images