RU2173167C2 - Method of direction of cellular immune response against hiv-infected mammalian cell, protein membrane-bound chimeric receptor, dna - Google Patents

Abstract

FIELD: medicine, genetic engineering, genetic therapy. SUBSTANCE: method involves the direction of cellular immune response against HIV-infected mammalian cells. Method involves effect of effective dose of cells expressing membrane-bound protein chimeric receptors on HIV-infected cells. Receptor has: (a) extracellular moiety which involves fragment CD4 that is able to recognize and bind HIV-infected cells but it does not mediate HIV infection and (b) intracellular moiety which is able to deliver a signal to indicated cells in order to destruct HIV-infected cells bound with receptor. Protein membrane- bound chimeric receptor has extracellular and intra-cellular components. Extracellular moiety has fragment CD4. Amino acid sequence is given in the invention description. Intracellular moiety is able to deliver a signal to cells carrying the indicated receptor. This signal induces destruction of receptor bound HIV or HIV-infected cells. Intracellular receptor is a moiety of signal-transmitting component of T-cellular receptor, B-cellular receptor or Fc- receptor. Fragment CD4 can be separated from intracellular moiety by transmembrane domain CD7 or CD5, or CD34. Structure of trans-membrane domain is given in the invention description. Invention presents also DNA encoding protein membrane-bound chimeric receptor. Invention provides to make functional chimeras that are able to direct immunocompetent cells for lysis of HIV-infected cells. EFFECT: improved method of direction. 15 cl, 28 dwg, 1 tbl

Description

 The invention was made with the support of the Government under Contract N A1 27849, awarded by the National Institutes of Health. The government has certain rights in this invention.

 The present invention relates to functional chimeras between CD4 fragments and receptors for immunocompetent cells (immunocytes), which are capable of directing immunocompetent cells to the lysis of HIV-infected cells, but do not make these immunocompetent cells susceptible to HIV infection. Thus, this invention provides a new and effective therapeutic agent against HIV.

BACKGROUND OF THE INVENTION
T-cell recognition of an antigen through a T-cell receptor is the basis of a number of immunological phenomena. T cells control the so-called cell-mediated immunity. It involves the destruction by cells of the immune system of foreign tissues or infected cells. There are various T cells, including helper and suppressor cells that modulate the immune response, and cytotoxic (or "killer") cells that can directly kill pathological cells.

 A T cell that recognizes and binds a unique antigen found on the surface of another cell becomes activated; it can then multiply and, if it is a cytotoxic cell, it can kill the cell bound by it.

HIV and immunopathogenesis
In 1984, it was shown that HIV is the etiological agent of AIDS. Since then, the definition of AIDS has been revised several times regarding what criteria should be included in the diagnosis. However, despite the instability in the diagnostic parameters, the simple common name for AIDS is HIV infection and the subsequent development of persistent constitutional symptoms and AIDS-determining diseases such as secondary infections, neoplasms and neurological disease. Harrison's Principles of Internal Medicine, 12th ed., McGraw Hill (1991).

 HIV is a human lentivirus retrovirus. Four recognized human retroviruses belong to two different groups: human T-lymphotropic (or leukemic) retroviruses, HTLV-1 and HTLV-2, and human immunodeficiency viruses, HIV-1 and HIV-2 (HIV-1 and HIV-2). The former are transformative viruses, while the latter are cytopathic viruses.

 HIV-1 has been identified as the most common cause of AIDS worldwide. The sequence homology between HIV-1 and HIV-2 is approximately 40%, with HIV-1 being closer to some members of the monkey immunodeficiency virus (SIV) group. CM Curran, J. et al. Science, 329: 1357-1359 (1985); Weiss, R. et al. Nature 324: 572-575 (1986).

, а также шесть экстрагенов, участвующих в репликации и других биологических активностях этого вируса. Как утверждалось ранее, общим определением СПИДа является глубокая иммуносупрессия преимущественно клеточно-опосредованного иммунитета. Эта иммуносупрессия (супрессия иммунного ответа) приводит к различным заболеваниям, вызываемым условно-патогенными микроорганизмами, в частности некоторым инфекциям и неоплазмам.HIV has the usual retroviral genes

as well as six extragens involved in replication and other biological activities of this virus. As stated earlier, the general definition of AIDS is deep immunosuppression of predominantly cell-mediated immunity. This immunosuppression (suppression of the immune response) leads to various diseases caused by opportunistic microorganisms, in particular some infections and neoplasms.

 As the main cause of immune deficiency in AIDS, quantitative and qualitative deficiency was identified in a subpopulation of lymphocytes formed by the thymus (T), the T4 population. This subpopulation of cells is determined phenotypically by the presence of a surface CD4 molecule, which has been shown to be a cellular receptor for HIV. Dalgleish et al. Nature 312: 763 (1984). Although a T4 cell is the main HIV-infected cell type, essentially any human cell expressing a CD4 molecule on its surface is able to bind to and become infected with HIV.

Traditionally, CD4 ⁺ T cells have been assigned the role of helper / inducer, indicating their function in providing an activating signal for B cells or in inducing T lymphocytes carrying the corresponding CD8 marker, converting them into cytotoxic / suppressor cells. Reinherz and Schlossman, Cell 19: 821-827 (1980); Goldstein et al., Immunol. Rev. 68: 5-42 (1982).

 HIV binds specifically and with high affinity, through a stretch of amino acids in the envelope of the virus (gp120), to part of region VI of the CD4 molecule located near its N-terminus. After binding, the virus fuses with the target cell membrane and is internalized. After internalization, he uses the enzyme reverse transcriptase to transcribe its genomic RNA into DNA, which integrates into cellular DNA, where it exists during the life of this cell as a “provirus”.

 Provirus can remain latent or be activated to transcribe mRNA and genomic RNA, which leads to synthesis, protein assembly, the formation of new virions and active virus replication (bading) as a result of penetration from the cell surface into the cell. Although the exact mechanism by which the virus induces cell death has not been established, it is believed that the main mechanism is massive viral badging from the cell surface, leading to the destruction of the plasma membrane and the disruption of the osmotic balance.

 During the infection process, the host organism produces antibodies against viral proteins, including the major gp120 and gp41 membrane glycoproteins. Despite this humoral immunity, the disease progresses, leading to lethal immunosuppression, characterized by multiple infections caused by opportunistic microorganisms, parasitemia, dementia and death. The inability of the host antiviral antibodies to stop the progression of the disease is one of the most unpleasant and disturbing aspects of this infection and gives poor predictions for vaccination efforts based on generally accepted approaches.

 Two factors may play a role in the effectiveness of the humoral response to immunodeficiency viruses. First, like other RNA viruses (and, in particular, like retroviruses), immunodeficiency viruses exhibit a high mutation rate in response to the immunological control of the host. Secondly, the envelope glycoproteins themselves are highly glycosylated molecules that provide few epitopes suitable for high-affinity antibody binding. The antigen-poor target, which is the viral envelope, provides little opportunity for the host to limit viral infection by producing specific antibodies.

HIV-infected cells express gp120 glycoprotein on their surface. gp120 mediates fusion events among CD4 ⁺ cells through a reaction similar to the reaction by which the virus enters uninfected cells, resulting in the formation of short-lived multinucleated giant cells. The formation of syncytium depends on the direct interaction of the gp120 membrane glycoprotein with the CD4 protein. Dalgleish et al., Supra; Klatzman, D. et al., Nature; 312: 763 (1984); McDougal, JS et al., Science 231: 382 (1986); Sodroski, J. et al. Nature 322: 470 (1986); Lifson JD et al. Nature 323: 725 (1986); Sodroski, J. et al., Nature 321: 412 (1986).

 The evidence that CD4-gp120 binding is responsible for viral infection of cells carrying the CD4 antigen includes the discovery that a specific complex forms between gp120 and CD4. McDougal et al., Supra. Other researchers showed that cell lines that were uninfected for HIV turned into infected cell lines after transfection and expression of the human CD4 cDNA gene. Maddon et al. Cell 46: 333-348 (1986).

 Therapeutic programs based on soluble CD4 as a passive agent to counteract viral uptake and syncytia-mediated cell transmission of the virus have been proposed and successfully demonstrated in vitro by a number of groups (Deen et al. Nature 331: 82-84 (1988); Fisher et al. Nature 331: 76-78 (1988); Hussey et al. Nature 331: 78-81 (1988); Smith et al. Science 238: 1704-1707 (1987); Traunecker et al. Nature 331: 84-86 (1988) ); and subsequently CD4 immunoglobulin fusion proteins with increased half-lives and moderate biological activity were provided (Capon et al. Nature 337: 525-531 (1989); Traunecker et al. Nature 339: 68-70 (1989); Byrn et al. , Nature 344: 667-670 (1990); Zettlmeissl et al., DNA Cell Biol. 9: 347-353 (1990)). Although CD4-immunotoxin conjugates or fusion proteins show strong cytotoxicity for infected cells in vitro (Chaudary et al. Nature 335: 369-372 (1988); Till et al. Science 242: 1166-1168 (1988)), the latent state of immunodeficiency syndrome makes it unlikely that any one type of therapy will be effective in eliminating the viral load, and the antigenicity of foreign fusion proteins appears to limit their acceptability in treatment requiring repeated dosing. Tests on monkeys infected with monkey immunodeficiency virus (SIV) have shown that soluble CD4 when administered without noticeable CD4 cytopenia can lower SIV titer and improve in vitro measurements of myeloid potential (Watanabe et al. Nature 337: 267-270 (1989)). However, after interruption of treatment, an immediate reappearance of the virus was observed, suggesting that lifelong administration may be necessary to prevent progressive depletion of the immune system.

T cell and Fc receptors
Cell surface expression of the most abundant form of T-cell antigen receptor (TCR) requires the simultaneous expression of at least 6 distinct polypeptide chains (Weiss et al., J. Exp. Med. 160: 1284-1299 (1984); Orloffhashi et al. Nature 316: 606-609 (1985); Berkhout et al., J. Biol. Chem. 263: 8528-8536 (1988); Sussman et al. Cell 52: 85-95 (1988)), α / β antigen binding chains , three polypeptides of the complex CD3 and ζ. If any of these chains is absent, then stable expression of the remaining members of the complex does not occur. ζ is a limiting surface expression polypeptide of the complete complex (Sussman et al. Cell 52: 85-95 (1988)), and it is believed that this chain mediates at least part of the cellular activation program triggered by ligand recognition by the receptor (Weissman et al., EMBO J. 8: 3651-3656 (1989); Frank et al. Science 249: 174-177 (1990)). The integral membrane homodimer 32 kDa type 1, ζ (zeta), has an extracellular domain of 9 residues without sites for adding (joining) N-linked glycan and an intracellular domain of 112 residues (mouse) or 113 residues (human) (Weissman et al. Science 238: 1018-1020 (1988); Weissman et al., Proc. Natl. Acad. Sci. USA 85: 9709-9713 (1988)). The ζ isoform called η (this) (Baniyash et al., J. Biol. Chem. 263: 9874-9878 (1988); Orloff et al., J. Biol. Chem. 264: 14812-14817 (1989)), which arises from the alternating mRNA splicing pathway (Jin et al., Proc. Natl. Acad. Sci. USA 87: 3319-3233 (1990)), is present in reduced amounts in cells expressing the antigen receptor. Ζ-η heterodimers are thought to mediate the formation of inositol phosphates, as well as the receptor-initiated programmed cell death called apoptosis (Mercep et al. Science 242: 571-574 (1988); Mercep et al., Science 246: 1162-1165 (1989 )).

Like ζ and η, the γ chain associated with the Fc receptor is expressed in cell surface complexes with additional polypeptides, some of which mediate ligand recognition, while others have an undefined function. γ (gamma) has a homodimeric structure and general organization very similar to the ζ structure and organization, and is a component of the high affinity mast cell / basophil IgE receptor, FceRI, which consists of at least three distinct polypeptide chains (Blank et al. Nature 337: 187 -189 (1989); Ra et al. Nature 241: 752-754 (1989)), as well as a component of the low affinity receptors for IgG present in Fc γ RII α mice (Ra et al., J. Biol. Chem. 264 : 15323-15327 (1989)), in humans, expression of the CD16 subtype by macrophages and natural killer cells, CD16 _TM (transmembrane CD16) (Lanier et al. Nature 342: 803-805 (1989); Ander son et al., Proc. Natl. Acad. Sci. USA 87: 2274-2278 (1990)) and an unidentified function polypeptide (Anderson et al., Proc. Natl. Acad. Sci. USA 87: 2274-2278 (1990) ) It has recently been reported that γ is expressed by the mouse T-cell line, CTL, in which it forms homodimers, as well as heterodimers γ-ζ and γ-η (Orloff et al. Nature 347: 189-191 (1990)).

 Fc receptors mediate phagocytosis of immune complexes, transcytosis, and antibody-dependent cell cytotoxicity (ADCC) (Ravetch and Kinet, Annu. Rev. Immunol. 9: 457-492 (1991); Unkeles et al., Annu. Rev. Immunol. 6 : 251-281 (1988); and Mellman, Curr. Opin. Immunol. 1: 16-25 (1988)). It has recently been shown that one of the murine isoforms of the low affinity Fc receptor, FcR γ IIIBl, mediates the internalization of Ig coated targets in clathrin-coated depressions and that another low affinity receptor, Fci γ lllA, mediates ADCC by binding to one or more members of a small family “trigger molecules” (Miettinen et al., Cell 58: 317-327 (1989); and Huziker and Mellman, J. Cell Biol. 109: 3291-3302 (1989)). These trigger (triggering) molecules, the T cell receptor (TCR), the ζ chain, the TCR η chain, and the Fc receptor γ chain interact with the ligand recognition domains of various receptors of the immune system and can autonomously initiate cellular effector programs, including cytolysis, after aggregation (Samelson et al. Cell 43: 223-231 (1985); Weissman et al. Science 239: 1018-1020 (1988); Jin et al., Proc. Natl. Acad. Sci. USA 87: 3319 -3323 (1990); Blank et al. Nature 337: 187-189 (1989); Lanier et al. Nature 342: 803-805 (1989); Kurosaki and Ravetch, Nature 342: 805-807 (1989); Hibbs et al. Science 246: 1608-1611 (1989); Anderson et al., Proc. Natl. Acad. Sci. USA 87: 2274-2278 (1990); and Irwing and Weiss, Cell 64: 891-901 (1991)) .

 However, when parallels are drawn between the Fc receptor families of low affinity in mice and humans, it becomes clear that FcR γ IIA and human C-isoforms do not have a copy in mice. Partly as a result of this, their function has yet to be determined.

 Due to the fact that humoral agents based on one CD4 have limited applicability in vivo, previous studies have identified the possibility of increasing cellular immunity to HIV. Protein chimera preparations were identified in which the extracellular domain of CD4 is fused to the transmembrane and / or intracellular domains of the T-cell receptor, Fc IgG receptor, or B-cell receptor signal transduction elements (U.S.S.N. 07/847566 and 07/665961). Cytolytic T cells expressing chimeras that include the extracellular domain of CD4 show strong MHC-independent destruction of cellular targets expressing HIV envelope proteins. An extremely important and new component of this approach was the identification of individual T-cell receptor, Fc receptor, and B-cell receptor chains, the aggregation of which is sufficient to initiate a cellular response. One particularly valuable application of this approach has been the invention of chimeras between CD4 and ζ, η, or γ, which target cytolytic T lymphocytes to recognize and kill cells expressing HIV gp120 (U.S.S.N. 07/847 566 and 07/665 961).

Summary of the invention
In general, this invention describes a method for directing a cellular immune response against an HIV-infected cell in a mammal. The method comprises administering to the mammal an effective amount of therapeutic cells expressing a membrane-bound protein chimeric receptor containing (a) an extracellular portion that includes a CD4 fragment capable of specifically recognizing and binding an HIV-infected cell but not mediating HIV infection, and (b) the intracellular part, which is capable of signaling a therapeutic cell about the destruction of an HIV-infected cell bound to a receptor.

 In a related method, the invention describes the use of therapeutic cells expressing a membrane-bound protein chimeric receptor comprising (a) an extracellular portion containing a CD4 fragment that is able to specifically recognize and bind an HIV-infected cell but does not mediate HIV infection, and (b) the intracellular portion, which is capable of signaling a therapeutic cell to destroy a receptor-bound HIV-infected cell, in the preparation of a medicament for the treatment of HIV-related diseases evaniy.

 In a second aspect, the invention describes a cell that expresses a protein membrane-bound chimeric receptor that contains (a) an extracellular portion that includes a CD4 fragment that is able to specifically recognize and bind an HIV-infected cell but does not mediate HIV infection, and ( b) an intracellular portion that is capable of signaling a therapeutic cell to destroy a receptor-bound HIV infected cell.

In preferred embodiments of both aspects, the CD4 fragment is amino acids 1-394 CD4 or amino acids 1-200 CD4; this CD4 fragment is separated from the intracellular portion by the CD7 transmembrane domain shown in FIG. 26, or the hinge region, CH2, and the CH3 domains of the human IgGl molecule shown in FIG. 25; the receptor includes the transmembrane portion of CD7; the receptor includes the transmembrane portion of CD5; the receptor includes the transmembrane portion of CD34; the CD4 fragment is separated from the membrane of the therapeutic cell by one or more protein α-helices; a CD4 fragment is separated from the therapeutic cell membrane by at least 48 angstroms (10 ^-10 m) or at least 72 angstroms; the intracellular portion is a signal transducing portion of a T-cell receptor protein (eg, ζ) of a B-cell receptor protein or Fc receptor protein; and therapeutic cells selected from the group consisting of: (a) T-lymphocytes; (b) cytotoxic T lymphocytes; (c) natural killer cells; (d) neutrophils; (e) granulocytes; (f) macrophages; (g) mast cells; (h) HeLa cells; and (i) embryonic stem cells (ES).

 In other aspects, the invention describes DNA encoding a chimeric receptor of the present invention; and a vector comprising chimeric receptor DNA.

 Although a characteristic embodiment of the invention is a chimera between CD4 and zeta, any receptor chain having a function similar to that of these molecules, for example, in granulocytes or B-lymphocytes, could be used for the purposes described here. Distinctive features of the desired trigger molecule of an immunocompetent cell include the ability to autonomously express (i.e., as a separate chain), the ability to fuse with the extracellular domain of CD4, so that the resulting chimera is present on the surface of the therapeutic cell, and the ability to initiate cell effector programs during secondary aggregation to collide with the target ligand.

 Currently, the most convenient way to deliver chimeric proteins to cells of the immune system is to use a form of gene therapy. However, reconstitution of immune system cells with chimeric receptors by mixing these cells with a suitably solubilized suitably purified chimeric protein would also result in a population of engineered cells capable of responding to HIV-infected targets. Similar approaches have been used, for example, to introduce a CD4 molecule into red blood cells for therapeutic purposes. In this case, the population of engineered cells would not be capable of self-renewal.

 The present invention relates to functional and simplified chimeras between fragments of a CD4 and a T-cell receptor, a B-cell receptor, and Fc receptor subunits that are capable of directing immunocompetent cells to recognize and lyse HIV-infected cells. A method for directing a cellular response in mammals involves administering an effective amount of therapeutic cells (eg, cytotoxic T lymphocytes) to a mammal, which cells are able to recognize and destroy the HIV-infected cell.

 The invention also includes chimeric receptor proteins that direct cytotoxic T lymphocytes for recognition and lysis of HIV-infected cells, host cells transformed with a vector containing these chimeric receptors, and antibodies directed against chimeric receptors.

 These and other non-limiting variations of this invention will be apparent to those skilled in the art from the following detailed description of the invention.

 In the following detailed description, reference will be made to various methodologies known to those skilled in the field of molecular biology and immunology. Publications and other materials representing such well-known referenced methodologies are hereby incorporated by reference in their entireties.

 Standard reference works outlining the basic principles of recombinant DNA technology include Watson et al. Molecular Biology of the Gene, Volumes I and II, the Benjamin / Cummings Publishing Company, Inc., publisher, Menio Park, CA (1987); Darnell et al. Molecular Cell Biology, Scientific American Books, Inc., Publisher, New York, N.Y. (1986); Lewin, Genesll, John Wiley & Sons, publishers, New York, N.Y. (1985); Old et al. Principles of Gene Manipulation: An Introduction to Genetic Ergineering, 2d edition. University of California Press, publisher, Berkeley, CA (1981); Maniatis et al. Molecular Cloning: A Laboratory Manual, 2nd Ed. Cold Spring Harbor Laboratory, publisher, Cold Spring Harbor, N.Y. (1989); and Ausubel et al. , Current Protocols in Molecular Biology, Wiley Press, New York, N.Y. (1989).

DEFINITIONS
By "cloning" is meant the use of in vitro recombination methods to insert a particular gene or other DNA sequence into a vector molecule.

 By "cDNA" is meant a complementary or copy-DNA produced from an RNA template by the action of an RNA-dependent DNA polymerase (reverse transcriptase). Thus, a “cDNA clone” refers to a duplex DNA sequence complementary to an RNA molecule of interest carried in a cloning vector.

 By "cDNA library" is meant a collection of recombinant DNA molecules containing inserted cDNAs that contain DNA that are copies of the mRNA expressed by the cell when the cDNA library is created. Such a cDNA library can be obtained by methods known to those skilled in the art and described, for example, in Ausubel et al., Supra and Maniatis et al. supra. In general terms, RNA is first isolated from the cells of an organism from the genome of which it is desirable to clone a particular gene. For the purposes of this invention, mammalian and, in particular, human lymphocyte cell lines are preferred. Currently, the preferred vector for this purpose is the vaccinia virus strain WR.

 “Vector” refers to a DNA molecule obtained, for example, from a plasmid, bacteriophage or mammalian virus or insect, into which DNA fragments can be inserted or cloned. The vector must contain one or more unique restriction sites and may be able to autonomously replicate in a particular host or carrier, such that the cloned sequence can propagate. Thus, by “expressing DNA vector” is meant any autonomous element capable of driving the synthesis of a recombinant peptide. Such DNA expression vectors include bacterial plasmids and phages and plasmins and viruses of mammals and insects.

 “Actually pure” refers to a compound, for example a protein, polypeptide or antibody, which does not actually contain the components that accompany them in nature. In general terms, the compound is actually pure if at least 60%, more preferably at least 75% and most preferably at least 90% of the total material in the sample is the target compound. Purity can be measured by any suitable method, for example, column chromatography, polyacrylamide gel electrophoresis, or HPLC analysis. In connection with a nucleic acid sequence, “substantially pure” means a nucleic acid sequence whose segment or fragment is not directly adjacent (i.e., covalently linked) to both coding sequences to which it is directly linked (i.e., to one at 5 '-end and the other at the 3'-end) in the naturally existing genome of the organism from which the DNA of the invention was obtained.

 By "fragment" of a molecule, such as any of the cDNA sequences of the present invention, is meant any portion of the adjacent nucleotides of the molecule. An “analog” of a molecule refers to a non-natural molecule that is substantially similar to either the whole molecule or its fragment. A molecule is said to be “substantially similar” to another molecule if the sequence of amino acids in both molecules is essentially the same. Essentially similar amino acid molecules will have similar biological activity. As used herein, a molecule is called a “chemical derivative” of another molecule when it contains chemical parts of a molecule that are not usually part of that molecule. Such parts of the molecule can improve the solubility of the molecule, absorption, biological half life, and the like. These parts of the molecule can, alternatively, reduce the toxicity of the molecule, eliminate or attenuate the unwanted side effect of the molecule, etc. Molecular moieties capable of mediating such effects are described, for example, in Remington's Phanmaceutical Sciences, 16th ed. Mack Publishing Co., Easton, Penn. (1980).

 "Functional derivative" of a chimeric receptor gene of the present invention refers to "fragments" or "analogues" of this gene that are "substantially similar" in nucleotide sequence and encode a molecule that has similar activity, for example, a chimera of a receptor for a T cell, B cell or Fc . Most preferably, this derivative has 90%, more preferably 70%, and preferably 40%, of the wild-type receptor chimera activity. The activity of a functional derivative of a chimeric receptor includes specific binding (with its extracellular portion of CD4) to an HIV-infected cell and ultimately destruction of that cell; in addition, the chimeric receptor does not make the receptor-bearing cell susceptible to HIV infection. Chimeric receptor activity may be determined, for example, by any of the methods described herein.

 The DNA sequence encoding the CD4 receptor chimera of the present invention or its functional derivative can be recombined with vector DNA in accordance with conventional methods, including blunt or step ends for ligation, restriction enzyme digestion to provide suitable ends, sticky ends, treatment with alkaline phosphatase to avoid undesired compounds; and ligation with suitable ligases. Methods for such manipulations are described by Maniatis, T., et al., Supra, and are well known in the art.

 A nucleic acid molecule such as DNA is said to be “capable of expressing” a polypeptide if it contains nucleotide sequences that include regulatory information for transcription and translation, and such regulatory sequences are “operably linked” to nucleotide sequences encoding this polypeptide. An “operative compound” is one in which the regulatory DNA sequences and the DNA sequences to be expressed are linked in such a way as to enable gene expression. The exact nature of the regulatory regions necessary for gene expression may vary from organism to organism, but in general terms will include a promoter (which controls the initiation of RNA transcription), as well as DNA sequences that, when transcribed into RNA, will signal the initiation of protein synthesis . Such regions typically include 5 ′ non-coding sequences involved in transcription and translation initiation, such as a TATA block, capping sequences, CAAT sequences, and the like.

 If desired, non-coding region 3 (to the right) of the protein coding gene sequence can be obtained by the methods described above. This region can be maintained for its regulatory transcription termination sequences. So, by preserving this 3′-region, in nature adjacent to the DNA sequence encoding the protein, transcription termination signals can be provided. In case these transcription termination signals do not function satisfactorily in the expressing host cell, a 3 ′ region functional in the host cell may be introduced instead.

 Two DNA sequences (such as a promoter region sequence and a CD4 chimeric receptor coding sequence) are called operably linked if the nature of the connection between these two DNA sequences (1) does not introduce a frame shift mutation, (2) does not interfere with the ability of the promoter region sequence direct transcription of the receptor chimera gene sequence or (3) does not interfere with the ability of the receptor chimera gene sequence to be transcribed by the promoter sequence district. The promoter region will be operatively linked to the DNA sequence if the promoter is able to initiate transcription of this DNA. Thus, transcription and translation signals recognized by the corresponding host are necessary for protein expression.

 The invention includes the expression of a chimeric CD4 receptor protein (or a functional derivative thereof) in prokaryotic or eukaryotic cells, although eukaryotic expression (and in particular in human lymphocytes) is preferred.

 Antibodies in accordance with this invention can be obtained in any of a variety of ways. For example, cells expressing a chimeric CD4 receptor protein or functional derivative thereof can be administered to an animal to induce the formation of serum containing polyclonal antibodies capable of binding these chimeric proteins.

 In a preferred method, the antibodies of this invention are monoclonal antibodies. Such monoclonal antibodies can be obtained using hybridoma technology (Kohler et al. Nature 256: 495 (1975); Kohler et al., Eur. J. Immunol. 6: 511 (1976); Kohler et al., Eur. J. Immunol. 6: 292 (1976); Haimnerling et al. In: Monoclonal Antibodies and T-cell Hybridomas, Elsevier, NY, pp. 563-684 (1981)). In general terms, such procedures include immunization of an animal with an antigen of a CD4 receptor chimera. The splenocytes of such animals are extracted and fused to a suitable myeloma cell line. Any suitable myeloma cell line may be used in accordance with this invention. After fusion, the resulting hybridoma cells are selectively maintained in a HAT medium and then cloned by serial dilution as described by Wands et al. (Gastroenterology 80: 225-232 (1981)). Hybridoma cells obtained by this selection are then analyzed to identify clones that secrete antibodies capable of binding to chimeras.

 Antibodies according to this invention can also be polyclonal or preferably region-specific polyclonal antibodies.

 Antibodies against the CD4 receptor chimera of the invention can be used to monitor the amount of a chimeric receptor (or cells carrying a chimeric receptor) in a patient. Such antibodies are very convenient for use in a standard immunodiagnostic assay known in the art, including immunometric or sandwich assays such as direct sandwich, reverse sandwich and simultaneous sandwich tests. These antibodies can be used in any number of combinations, which can be determined by experts without undue experimentation to conduct immunoassays of acceptable specificity, sensitivity and accuracy.

 Standard works setting forth general principles of immunology include Roitt, Essential Immunology, 6th ed., Blackwell Scientific Publications, Publisher, Oxford (1988); Kimball, Introduction to Immunology, 2d ed., Macrnillan Publishing Co., Publisher, New York (1986); Roitt et al. Immunology, GoWer Medical Publishing Ltd. Publisher, London (1985); Campbell, "Monoclonal Antibody Technology", in Burdon et al., Eds. Laboratory Techniques in Biochemistry and Molecular Biology, Volume 13, Elsevier, Publisher, Amsterdam (1984); Klein, Immunology: The Science of Self-Nonself Discrimination, John Wiley & Sons, Publisher, New York (1982); and Kennet et al., eds. Monoclonal Antibodies, Nyb-ridoma: A New Dimension In Biological Analyses, Plenum Press, Publisher, New York (1980).

 By "detection" ("detection") is meant the determination of the presence or absence of a substance or the determination of the amount of a substance. The term, therefore, refers to the use of the materials, compositions and methods of this invention for qualitative and quantitative determinations.

 Antibodies and actually purified antigen of the present invention are ideally suited for the manufacture of a whale (kit). Such a whale may contain a carrier (holder) in which one or more containers, such as bottles, tubes, etc., are placed, each container containing individual elements of the analysis that must be performed.

 Many types of tests can be included in whales, such as competitive and non-competitive analyzes. Typical examples of assays that can use the antibodies of this invention are RIA radioimmunoassay (RIA), enzyme immunoassays (EIA), enzyme-linked immunosorbent assays (ELISA) and immunoassays, or sandwich immunoassays.

 By the term “immunometric assay” (quantitative immunoassay), or “sandwich immunoassay”, is meant a simultaneous sandwich, direct sandwich, reverse sandwich immunoassay. These terms are well understood by those skilled in the art. They also understand that the antibodies in accordance with this invention will be applicable in other variations and forms of assays that are currently known and which may be developed in the future. It is implied that they are included in the scope of this invention.

 The phrase “specifically recognizes and binds” means that the antibody recognizes and binds a chimeric receptor polypeptide, but does not actually recognize and does not bind other molecules that are not related to this polypeptide in the sample, for example, in a biological sample.

 “Therapeutic cell” refers to a cell that has been transformed by the CD4 receptor chimera of this invention in such a way that it is able to recognize and destroy an HIV-infected cell; preferably such therapeutic cells are cells of the hematopoietic system.

 The term "extracellular" means that at least a portion of the molecule is exposed on the surface of the cell. "Intracellular" means having at least a portion of a molecule located in the cytoplasm of a therapeutic cell. "Transmembrane" means that at least a portion of the molecule is in the plasma membrane. The "extracellular part", "intracellular part" and "transmembrane part" may include flanking amino acid sequences that extend into adjacent cell compartments.

 “Oligomerization” means to complex with other proteins to form higher order dimers, trimers, tetramers or oligomers. Such oligomers may be homo-oligomers or hetero-oligomers. The "oligomerizing portion" is a region of a molecule that controls the formation of a complex (i.e., oligomer).

 “Cytolytic” means capable of destroying a cell (eg, an HIV-infected cell) or capable of destroying an infectious agent (eg, an HIV virus).

 By "immunodeficiency virus" is meant a retrovirus that, in the wild-type form, is capable of infecting T4 cells of a primate host and has viral morphogenesis and morphology characteristic of the lentivirus subfamily. This term includes, without limitation, all variants of HIV (HIV) and SIV, including HIV-1, HIV-2, SIVmac, SIVagm, SIVmnd, SIVsmm, SIVman, SIVmand and SIVcpz.

 "MHC-independent" means that the cellular cytolytic response does not require the presence of a major histocompatibility complex (MHC) class II antigen on the surface of the target cell.

 "Functional transmitting (transducing) cytolytic signal derivative" means a functional derivative (as defined above) that is capable of directing at least 40%, more preferably 70% or most preferably 90% of the biological activity of a wild-type molecule. As used herein, a “functional cytolytic signal transducing derivative” can act by directly signaling to a therapeutic cell to destroy an agent or cell bound to the receptor (for example, in the case of the intracellular portion of a chimeric receptor), or it can act indirectly by enhancing oligomerization with cytolytic signal-transmitting proteins of the therapeutic cell ( for example, in the case of a transmembrane domain). Such derivatives can be tested for their effectiveness, for example, using the in vitro tests described herein.

 A “functional HIV envelope binding derivative" is a functional derivative (as defined above) that is capable of binding any HIV envelope protein.

 Functional derivatives can be identified using, for example, the in vitro tests described herein.

THERAPEUTIC INTRODUCTION
Transformed cells of the present invention are used for the treatment of immunodeficiency virus. Current methods of administering such transformed cells include adoptive immunotherapy or cell transfer therapy. These methods allow the return of transformed cells of the immune system into the bloodstream. Rosenberg, Scientific American 62 (May 1009); Rosenberg et al. The New England Journal of Medicine 323 (9): 570 (1990).

 The pharmaceutical compositions of this invention can be administered to any animal that may experience the beneficial effects of the compounds of this invention. The main among these animals are humans, although the invention is not limited in this way.

Detailed description
First, drawings will be described.
Brief Description of the Drawings
FIG. 1A is the amino acid sequence near the fusion site between CD4 (residues 1-369) and various receptor chains. The underlined sequence indicates the position of the amino acids encoded within the BamHl site used to construct the fusion. The beginning of the transmembrane domain is marked with a vertical line. The sequence η is identical to the sequence ζ at the amino end, but differs at the carboxyl end (Jin et al., Proc. Natl. Acad. Sci. USA 87: 3319-3323 (1990)).

FIG. 1B is a flow cytometric analysis of surface expression of CD4, CD4: ζ CD4: γ, and CD4: η in CVI cells. Cells were infected with a virus expressing CD4 or CD16 _PI chimeras, incubated for 9 hours at 37 ^° C and stained with anti-CD4 MAb Leu3A conjugated with phycoerythrin.

FIG. 2 shows the surface expression of CD16 _™ after co-infection with one CD16 _™ alone (solid points) or co-infection with a virus expressing CD4: γ (streaks) or CD4: ζ (solid line). Rare points are cells infected with one CD4: ζ stained with 3G8 (Fleit et al., Proc. Natl. Acad. Sci. USA 79: 3275-3279 (1982)) (anti-CD16 MAb).

FIG. 3 shows the surface expression of CD16 _™ after co-infection with viruses expressing CD16 _™ and the following ζ chimeras: CD4: ζ (bold line), CD4: ζ CllG (solid line); CD4: ζ (dashed line); CD4: ζ C11G / D15G (dense points); no coinfection (CD16 _TM alone, sparse dots). Cells were incubated with anti-CD16 MAb 3G8 and goat anti-mouse IgG conjugated with phycoerythrin Fab'2. The expression level of ζ chimeras was essentially identical for the various mutants analyzed, and co-infection of cells with viruses expressing CD16 _TM and ζ chimeras did not significantly alter the surface expression of chimeras.

FIG. 4A-D show an increase in the intracellular free calcium ion after crosslinking of mutant ζ chimeras in the T cell line. Jurkat E6 cells (Weiss et al., J. Immunology 133: 123-128 (1984)) were infected with recombinant vaccinia viruses and analyzed by flow cytometry. The results for the CD4 ⁺ population are presented, so only cells expressing the corresponding chimeric protein are analyzed (by setting a discriminatory window in the sorter). The average ratio of violet to blue fluorescence Indo-I reflects the concentration of intracellular free calcium in the population as a whole, and the percentage of responding cells reflects a fraction of cells that exceeds a predetermined threshold ratio (set so that 10% of untreated cells are positive).

 FIG. 4A and 4B show Jurkat cells expressing CD4: ζ (solid line) or CD16: ζ (dashed line), which were incubated with anti-CD4 MAb Leu3a (conjugated to phycoerythrin) followed by crosslinking with goat anti-mouse IgG antibodies. A dotted line shows the response of uninfected cells to anti-CD3 MAb OCT3. FIG. 4C and 4D show Jurkat cells expressing CD4: ζ D15G (solid line); CD4: ζ C11G / D15G (dashes) or CD4: ζ C11G (dots) that were processed and analyzed as in FIG. 4A and 4B.

FIG. 5A-C show that receptors CD4: ζ, CD4: η and CD4: γ allow cytolytic T lymphocytes (CTL) to kill
targets expressing HIV-1 gp120 / 41. FIG. 5A: solid circles — CTL expressing CD4: ζ were incubated with HeLa cells expressing gp120 / 41; open circles - CTLs expressing CD4: ζ were incubated with uninfected HeLa cells; solid squares — uninfected CTLs incubated with HeLa cells expressing gp120 / 41; open squares - uninfected CTLs incubated with uninfected HeLa cells. FIG. 5B: solid circles — CTL expressing CD4: η incubated with HeLa cells expressing gp120 / 41; open circles - CTL expressing CD4: γ incubated with HeLa cells expressing gp120 / 41; open squares are CTLs expressing the C11G / D15G double mutant CD4: ζ chimera incubated with HeLa cells expressing gp120 / 41. FIG. 5C: Flow cytometric analysis of the CD4 CTL expression used in FIG. 5B. To adjust the target to effector ratios, the percentage of cells expressing the chimeric CD4 was determined by subtracting the scaled negative (uninfected) population by superimposing histograms; for comparison purposes, in this figure, uninfected cells were defined as an arbitrary threshold, which gives roughly the same fraction, positive for other cell populations, which would result in the subtraction of histograms.

FIG. 6A-B show the specificity of CD4-directed cytolysis. FIG. 6A: solid circles — CTL expressing CD4: ζ incubated with HeLa cells expressing CD16 _PI ; open circles are CTLs expressing CD4 incubated with HeLa cells expressing gp120 / 41. FIG. 6B: solid circles — CTL expressing CD4: ζ incubated with Raji cells (MHC class II ⁺ ); open circles - uninfected CTL cells incubated with RJ2.2.5 (MHC class II mutant ^- Raji) cells; solid squares - uninfected CTLs incubated with Raji cells (MHC class 11 ⁺ ); open squares are CTLs expressing CD4: ζ incubated with RJ2.2.5 cells (MHC class II ^- ). The scale of the ordinate is increased.

FIG. 7A-B characterize the chimeric receptor CD16: ζ. FIG. 7A is a schematic diagram of a CD16: ζ fusion protein. The extracellular portion of the phosphatidylinositol-conjugated form of monomeric CD16 was coupled to the dimeric ζ directly externally with respect to the transmembrane domain. The protein sequence of the compound is shown at the bottom. FIG. 7B shows a flow cytometric analysis of calcium mobilization after crosslinking of the CD16: ζ chimera in either a TCR-positive or TCR-negative cell line. The average ratio of violet to blue fluorescence (a measure of the relative concentration of calcium ion) is shown among cell populations treated with antibodies at time 0. Solid squares are the response of Jurkat cells to anti-CD3 MAb OKT3; solid triangles — CD16: ζ response to anti-CD16 MAb 3G8 crosslinked in the REX33A TCR ^- mutant; open squares - response to crosslinking of CD16: ζ in the TCR ^- mutant cell line Jurkat JRT3.T3.5; open triangles - response to crosslinking of CD16: ζ in Jurkat cells; crosses - response to non-chimeric CD16 in Jurkat cells; dots — response to non-chimeric CD16 in a REX33A TCR cell line.

FIG. 8A-B shows a deletion analysis of cytolytic potential. FIG. 8A shows the locations of the end points of the deletions ζ. Here, as elsewhere, mutations in ζ are represented as follows: the initial residue — location — residue of the mutant, so that D66 ^* , for example, means replacing Asp-66 with a termination codon. FIG. 8B shows the results of cytolysis analysis of undeleted CD16: ζ and secreting ζ deletions. Hybridoma cells expressing surface antibodies to CD16 were loaded with ⁵¹ Cr and incubated with increasing amounts of human cytolytic lymphocytes (CTL) infected with vaccinia virus recombinants expressing CD16 chimeras: ζ The percentage of isolated ⁵¹ Cr is plotted as a function of effector cell ratio (CTL) to target cells (hybridoma cells) (e / t). Solid circles — cytolysis mediated by cells expressing CD16: ζ (mfi 18.7); solid squares — cytolysis mediated by cells expressing CD16: ζ Asp66 ^* (mfi 840.2); open squares - cytolysis mediated by cells expressing CD16: ζ Glu60 ⁺ (mfi 16.0); open circles — cytolysis mediated by cells expressing CD16: ζ Tyr51 ^* (mfi 17.4); solid triangles - cytolysis mediated by cells expressing CD16: ζ Phe34 ^* (mfi 17.8); and open triangles — cytolysis mediated by cells expressing non-chimeric CD16 (mfi 591). Although in this experiment the expression of CD16: ζ Asp66 ^{* was} not comparable with the expression of other fusion proteins, cytolysis by cells expressing CD16: ζ at equivalent levels in the same experiment yielded results substantially identical to those found by cells expressing CD16: ζ Asp66.

FIG. 9A-D show that elimination of the potential for transmembrane interactions reveals a short ζ segment capable of mediating cytolysis. FIG. 9A is a schematic diagram of monomeric two and three part chimeras. At the top is the CD16: ζ construct, shortened at residue 65 and lacking the transmembrane residues Cys and Asp. Below are the constructs of CD16: CD5: ζ and CD16: CD7: ζ and the corresponding controls. The peptide sequences of the intracellular domains are shown below. FIG. 9B shows the cytolytic activity of monomeric chimeric deletion mutants. The cytolytic activity of cells expressing CD16: ζ (solid circles; mfi 495) was compared with the cytolytic activity of cells expressing CD16: ζ Asp66 ^* (solid squares; mfi 527) or CD16 mutants: ζ Cys11Gly / Asp15Gly / Asp66 ^* (open squares; mfi 338) and CD16: ζ Cys11Gly / Asp15Gly / Glu60 ^* (filled triangles; mfi 259). FIG. 9C shows cytolytic activity mediated by three-part fusion proteins. Solid triangles - CD16: ζ Asp66 ^* ; open squares - CD16: 5: ζ (48-65); solid squares - CD16: 7: ζ (48-65); open triangles - CD16: 7: ζ (48-59); open circles - CD16: 5; solid circles — CD16: 5. FIG. 9D shows the mobilization of calcium by mutants and three-part chimeras in the TCR ^, the Jurkat mutant cell line JRT3.T3.5. Open circles — response of cells expressing dimeric CD16: ζ Asp66 ^* ; solid squares — response of cells expressing CD16: ζ Cys11Gly / Asp15Gly / Asp66 ^* ; open squares - response of cells expressing CD16: ζ Cys11Gly / Asp15Gly / Glu60 ^* ; solid triangles - response of cells expressing CD16: 7: ζ (48-65); and open triangles - response of cells expressing CD16: ζ (48-59).

 FIG. 10A-F show the contribution of individual amino acids to the activity of a cytolytic signal-transmitting motif of 18 residues. FIG. 10A and 10B show cytolytic activity, and FIG. 10C shows the mobilization of calcium ion mediated by chimeras carrying point mutations near the carboxyl terminal tyrosine (Y62). FIG. 10A and 10B present data obtained on cells expressing low and high amounts, respectively, of CD16: ζ fusion proteins. Identical symbols are used to mobilize calcium and cytolysis assays and are shown in the single letter code on the right. Solid circles are cells expressing CD16: ζ (mfi in A, 21; B, 376); solid squares - cells expressing CD16: 7: ζ (48-65) (mfi A, 31; B, 82); open squares - CD16: 7: ζ (48-65) Glu60Gln (mfi A, 33; B, 92); crosses - CD16: 7: ζ (48-65) Asp63Asn (mfi A, 30; B, 74); solid triangles - CD16: 7: ζ (48-65) Tyr62Phe (mfi A, 24; B, 88); open circles - CD16: 7: ζ (48-65) Glu6lGln (mfi A, 20; B, 62); and open triangles - CD16: 7: ζ (48-65) Tyr62Ser (mfi B, 64). FIG. 10D and 10E show cytolytic activity, and FIG. 10F shows the mobilization of calcium ion by chimeras carrying point mutations near the amino terminal tyrosine (Y51). Identical symbols are used to mobilize calcium and cytolysis assays and are shown on the right. Continuous circles-cells expressing CD16: ζ (mfi in D, 21.2; in E, 672); solid square cells expressing CD16: 7: ζ (48-65) (mfi D, 31.3; E, 179); solid triangles - CD16: 7: ζ (48-65) -Asn48Ser (mfi D, 22.4; E, 209); open squares - CD16: 7: ζ (48-65) Leu50Ser (mfi D, 25.0; E, 142); and open triangles - CD16: 7: ζ (48-65) Tyr51Phe (mfi D, 32.3; E, 294).

 FIG. 11A-B show a comparative analysis of internal ζ repeats and a comparison of their ability to support cytolysis. FIG. 11A is a schematic diagram of chimeras formed by dividing the intracellular domain ζ into thirds and joining them to the transmembrane domain of the CD16: 7 chimera. The sequences of these intracellular domains are shown below, with common residues enclosed in blocks and close residues marked with asterisks. FIG. 11B shows the cytolytic activity of three subdomains of ζ. Solid circles are cells expressing CD16: ζ (mfi 476); solid squares - CD16: 7: ζ (71-104) (mfi 114); and solid triangles - CD16: 7: ζ (104-138) (mfi 104).

 FIG. 12 is a schematic of the CD16: FcRγII chimeras.

 FIG. 13A-B shows calcium mobilization after crosslinking of CD4: FcR γ II and CD16: FcRγII chimeras. FIG. 13A shows the ratio of violet to blue fluorescence emitted by cells loaded with calcium-sensitive Indo-1 fluorophore, presented as a function of time after crosslinking of the extracellular domain of CD16 with antibodies. FIG. 13B shows a similar analysis of the increase in the ratio of violet to blue fluorescence of cells bearing CD4: FcRγII chimeras after crosslinking with antibodies.

FIG. 14A-B shows an analysis of the cytolysis of chimeras CD4: FcR γ II and CD16: FcR γ II. FIG. 14A shows the percentage of ⁵¹ Cr isolated from anti-CD16 hybridoma cells upon incubation of cells with increasing amounts of cytotoxic T lymphocytes expressing CD16 chimeras: FcR γ II (effector cells). FIG. 14B shows a similar analysis of the cytotoxicity mediated by CD4: FcR γ II chimeras against target cells expressing HIV envelope glycoproteins.

 FIG. 15A-E shows the identification of residues in the tail of FcR γ II A, which is important for cytolysis. FIG. 15A is a diagram of deletion constructs. FIG. 15B and 15C show calcium mobilization and cytolysis of the carboxyl-terminal deletion variants of CD16: FcR γ II A. FIG. 15B and 15E show the mobilization of calcium and cytolysis by three-part chimeras carrying the progressively decreasing amino end of the intracellular tail, CD16: FcR γ II A.

 FIG. 16 (SEQ ID N 24) shows the amino acid sequence of the CD3 delta receptor protein; the block sequence is a preferred signal transmitting portion.

 FIG. 17 (SEQ ID N 25) shows the amino acid sequence of the T3 gamma receptor protein; the boxed sequence is a preferred cytolytic signal transmitting portion.

 FIG. 18 (SEQ ID N 26) shows the amino acid sequence of the mbi receptor protein; the boxed sequence is a preferred cytolytic signal transmitting portion.

 FIG. 19 (SEQ ID 27) shows the amino acid sequence of the B29 receptor protein; the boxed sequence is a preferred cytolytic signal transmitting portion.

FIG. 20A-E shows a diagram of the CD4 chimeras. Molecule "A" -CD4 (D1-D4): Ig: CD7; molecule "B" - CD4 (D1, D2): Ig: CD7; molecule "C" - CD4 (D1-D4): Ig: CD7: ζ; molecule "D" - CD4 (D1, D2): Ig: CD7: ζ; and the E molecule is CD4: ζ. The extracellular domain of the human CD4 molecule, corresponding to amino acids 1-394 of the precursor, was connected using the BamHI site with a hinge
region, CH1 and CH2 domains of human IgG1, as previously described (Zettlmeissl et al., DNA Cell Biol. 9: 347 (1990)), except that the cDNA version of human Ig sequences was used to allow expression in recombinant cowpox viruses. A two-domain version of the CD4 chimeras was created by embedding the BamHI adapter at the unique NheI site (corresponding to amino acid 200) in the cDNA of the CD4 precursor. Membrane accession sequences consisted of 22 residues from the first exon of human membrane-bound IgG1, followed by CD7 residues 146-203. Amino acids 55-163 ζ served as the trigger motif of four-part constructs (C and D). In four-part constructs containing the ζ chain, intracellular expression of ζ with commercially available antibodies against the intracellular domain (Coulter) has been documented.

FIG. 21 shows the cytolysis of target cells expressing an HIV envelope glycoprotein mediated by a clone of cytotoxic T cells, WH3 expressing various chimera derived from CD4 as effector molecules. For cytotoxicity assays, CD8 ⁺ CD4 ^- HLA B44, the restricted human T cell line, WH3, was maintained in IMDM supplemented with 10% human serum as previously described here. Cells were stimulated with irradiated gamma rays (3000 rad) carrying B44 mononuclear cells and phytohemagglutinin (PHA) at a concentration of 1 μg / ml. After one day of stimulation, the PHA was diluted to 0.5 μg / ml by the addition of fresh medium; after 3 days, the medium was completely replaced. Cells were grown for at least 10 days before being used in cytotoxicity assays. Cells were infected with suitable recombinant vaccinia viruses as described here for VPE16. Infections occurred for an additional 3-4 hours in complete medium, after which the cells were collected by centrifugation and resuspended at a density of 1 • 10 ⁷ cells / ml. 100 μl was added to each well of a microtiter plate with a U-shaped lower part containing 100 μl per well of complete medium and diluted in 2-fold serial steps. Two wells for each sample did not contain lymphocytes to measure spontaneous chromium excretion and total chromium uptake. Target cells, HeLa subline S3 (HeLa-S3, ATCC), were infected as described above, in 10 cm plates with vPE16 • 10 ⁶ infected cells were separated using PBS and 1 mM EDTA, centrifuged and resuspended in 100 μl ⁵¹ Cr- sodium chromate (1 mCi / ml in PBS) and kept for 1 hour at 37 ^° C and then washed 3 times with PBS. 100 μl of labeled target cells were added to each well. The microtiter plate was unscrewed at 750 x g for 1 minute and incubated for 4 hours at 37 ^° C. At the end of the incubation period, the cells in each well were resuspended by careful pipetting, a sample was taken to determine the total number of pulses included, and the microtiter plate was unscrewed at 750 x g within 1 minute. Aliquots (100 μl) of the supernatant were taken and counted in a gamma ray scintillation counter. The effector: target ratio was corrected for the percentage of infected cells measured by flow cytometry.

FIG. 22 shows HIV-1 replication in transfected cell lines. Cell lines stably expressing wild-type CD4 and various recombinant chimeras were created in the subline of the 293 human embryonic kidney cell line. The stock solution of the HIV-1IIIB isolate virus was prepared with a titer of approximately 10 ⁶ infectious particles / ml as measured by analysis of the final dilution results using the human T-cell line C8166 as an indicator. Infections were carried out with an approximate multiplicity of infection 1 for a period of 8-12 hours at 37 ^° C. The next day, the cells were washed with PBS three times, trypsinized, reseeded in new plates and the culture supernatant was taken to determine the titer (designated as day 0). At intervals of 3-4 days thereafter, cell culture supernatants were collected and stored for p24 analysis. The cells were supplied with fresh medium containing hygromycin B at a concentration of 100 μg / ml. Culture supernatants were analyzed using a commercial kit for ELISA-based analysis of the p24 HIV-1 antigen (Coulter) in accordance with the instructions supplied by the manufacturer. The results are representative of two independent experiments of similar duration.

 FIG. 23 shows the nucleic acid sequence and amino acid sequence of the D1-D4 CD4 domains (CD4 Bam).

 FIG. 24 shows the nucleic acid sequence and amino acid sequence of the D1-D2 CD4 domains (CD4 Nhe).

 FIG. 25 shows the nucleic acid sequence and the amino acid sequence of the hinge region, CH2 and CH3 domains of human IgG1 (Igh23 Bam).

 FIG. 26 shows the nucleic acid sequence and amino acid sequence of the transmembrane domain of CD7 (TM7 Barn Mlu).

 FIG. 27 shows the nucleic acid sequence and amino acid sequence of the intracellular domain of zeta (Zeta Mlu Not).

 FIG. 28 shows a DNA sequence and a primary amino acid sequence of a synthetic alpha helix.

EXAMPLE 1
Construction of human IgGl chimeras: receptor
Human IgGl heavy chain sequences were obtained by connecting sequences in the CH3 domain with a cDNA fragment derived from the 3 ′ end of the transmembrane form of an antibody mRNA. The 3'-terminal fragment was obtained by polymerase chain reaction using the tonsil cDNA library as a substrate and oligonucleotides having the sequence:
CGC GGG GTG ACC GTG CCC TCC AGC AGC TTG GGC (SEQ ID N 7)
CGC GGG GAT CCG TCG TCC AGA GCC CGT CCA GCT CCC CGT CCT CCT CA (SEQ ID N 8),
corresponding to the 5'- and 3'-ends of the desired DNA fragments, respectively, the 5'-oligo is complementary to the site in the CH1 domain of human IgGl, and the 3'-oligo is complementary to the site located directly 5 'from sequences encoding the membrane-crossing domain. The PCR product was digested with BstXI and BamHI and ligated between the BstXI and BamHI sites of the semi-synthetic IgGl gene of the antibody carrying the variable and constant regions. After inserting the BstXI-BamHI fragment, the amplified parts of this construct were introduced into the Smal site in CH3 by exchange of restriction fragments, so that only the part between the SmaI site and the 3'-oligo was produced from the PCR reaction.

 To create a chimeric receptor, human IgGl: ζ heavy chain gene ending at the BamHI site was linked to the BamHI site of the ζ chimera described below, so that the antibody sequences formed the extracellular part. Flow cytometry of COS cells transfected with a plasmid encoding this chimera showed high levels of expression of the determinate antibodies upon co-transfection of the expression plasmid encoding the light chain cDNA and moderate expression of the determinants of the antibody in the absence of the light chain expressing plasmid.

 Such chimeras, including human IgG1 fused to η or γ (see below) or any signal transducing portion of a T-cell receptor or Fc receptor protein, can be constructed generally, as described above, using standard molecular biology methods .

To create a separate transcriptional unit that would allow both heavy and light chains to be expressed from a single promoter, a plasmid encoding a bicistronic mRNA was created from sequences encoding a heavy chain and light chain, and a 5'-untranslated part of this mRNA encoding glucose-regulated 78 kDa protein, otherwise known as grp78 or BiP. The grp78 sequences were obtained by PCR from human genomic DNA using primers having the sequences:
CGC GGG CGG CCG CGA CGC CGG CCA AGA CAG CAC (SEQ ID N 9) and
CGC GTT GAC GAG CAG CCA GTT GGG CAG CAG CAG (SEQ ID IP 10)
at the 5'- and 3'-ends, respectively. Polymerase chain reactions were performed with these oligonucleotides in the presence of 10% dimethyl sulfoxide. The fragment obtained by PCR was digested with NotI and HincII and inserted between NotI and HpaI during transcription from sequences encoding human IgGl. Subsequently, sequences encoding human IgG kappa light chain cDNA were inserted upstream of the grp78 leader using the HincII site and another site in the vector. The expression plasmid obtained as a result of these manipulations consisted of a semi-synthetic heavy chain gene, followed by grp78 leader sequences, followed by kappa light chain cDNA sequences and polyadenylation signals obtained from the SV40 DNA fragment. Transfection of COS cells with this expression plasmid yielded markedly improved expression of heavy chain determinants compared to transfection of a plasmid encoding only heavy chain determinants.

 To create a bicistronic gene containing a heavy chain / receptor chimera and a light chain, the heavy chain upstream sequences can be replaced by any heavy chain / receptor chimera gene described herein.

Е. coli в качестве селектируемого маркера, и встраивали в геном штамма WR вируса коровьей оспы посредством гомологичной рекомбинации и отбора на рост в микофенольной кислоте (Falkner et al., J. Virol. 62:1849-1854 (1988); Boyle et al. Gene 65:123-128 (1988)). Проточная цитометрия показала, что рекомбинанты вируса коровьей оспы направляют обильное образование слитых белков CD4: ζ и CD4: γ при клеточной поверхности, тогда как экспрессия CD4: η была значительно более слабой (фиг. 1B). Последнее открытие согласуется с недавним сообщением, что трансфекция экспрессионной плазмиды кДНК η в мышиную гибридомную клеточную линию дала значительно меньшую экспрессию, чем трансфекция сравнимой экспрессионной плазмиды ζ (Clayton et al., J. Exp. Med. 172:1243- 1253 (1990)). Иммунопреципитация клеток, инфицированных рекомбинантами вируса коровьей оспы, выявила, что эти слитые белки образуют ковалентные димеры в отличие от природно встречающегося антигена CD4. Было найдено, что мол. массы мономерных слитых белков CD4: ζ и CD4: γ и нативного CD4 равны 63, 55 и 53 кД соответственно. Большие массы слитых белков приблизительно согласуются с большей длиной внутриклеточной части, которая превышает длину внутриклеточной части нативного CD4 на 75 (CD4: ζ) или 5 (CD4: γ ) остатков.EXAMPLE II
Construction of chimeras CD4 receptor
ζ cDNA (Weissman et al., Proc. Natl. Acad. Sci. USA 85: 9709-9713 (1988)) and γ (Kuster et al., J. Biol. Chem. 265: 6448-6452 (1990)) were isolated using polymerase chain reaction from libraries obtained from the HPB-ALL tumor cell line (Aruffo et al., Proc. Natl. Acad. Sci. USA 84: 8573-8577 (1987b)) and from natural human killer cells, then As η cDNA (Jin et al., Proc. Natl. Acad. Sci. USA 87: 3319-3323 (1990)) was isolated from a mouse thymocyte library, ζ, η and γ cDNA were connected to the extracellular domain of the designed form CD4 with the BamHI site immediately opposite the transcription from the membrane-crossing domain (Aruffo et al., Proc. Natl. Acad. Sci. USA 8 4: 8573-8577 (1987b); Zettlmeissl et al., DNA Cell Biol. 9: 347-353 (1990)), which was linked to the BamHI site naturally present in the ζ and η cDNA in a similar position at a distance of several residues upstream transcription from the membrane crossing domain (SEQ ID N 1, N 3, N 4 and N 6). To form a fusion protein with γ, the BamHI site was constructed in sequence at the same approximate position (FIG. 1; SEQ ID N 2 and 5). These gene fusions were introduced into a cowpox virus expression plasmid carrying the gene

E. coli as a selectable marker, and was inserted into the genome of the WR strain of vaccinia virus by homologous recombination and selection for growth in mycophenolic acid (Falkner et al., J. Virol. 62: 1849-1854 (1988); Boyle et al. Gene 65: 123-128 (1988)). Flow cytometry showed that vaccinia virus recombinants direct the abundant formation of CD4: ζ and CD4: γ fusion proteins at the cell surface, while CD4: η expression was significantly weaker (Fig. 1B). The latter finding is consistent with a recent report that transfection of the expression plasmid η cDNA into a murine hybridoma cell line yielded significantly less expression than transfection of a comparable expression plasmid ζ (Clayton et al., J. Exp. Med. 172: 1243-1253 (1990)) . Immunoprecipitation of cells infected with vaccinia virus recombinants revealed that these fusion proteins form covalent dimers, in contrast to the naturally occurring CD4 antigen. It was found that pier. the masses of the monomeric fusion proteins CD4: ζ and CD4: γ and native CD4 are 63, 55 and 53 kD, respectively. The large masses of fusion proteins are approximately consistent with the greater length of the intracellular part, which exceeds the length of the intracellular part of native CD4 by 75 (CD4: ζ) or 5 (CD4: γ) residues.

EXAMPLE III
CD4 chimeras can bind to other receptor chains
Cell surface expression of macrophage / natural killer cells of human FcγRIII (CD16 _TM ) on transfectants is facilitated by co-transfection with murine (Kurosaki et al. Nature 342: 805-807 (1989)) or human (Hibbs el al. Science 246 : 1608-1611 (1989)) γ as well as human ζ (Lanier et al. Nature 342: 803-805 (1989)).

In accordance with these reports, the expression of chimeras also made possible the surface expression of CD16 _TM upon delivery to the target cell either by co-transfection or by co-infection with recombinant vaccinia viruses (Fig. 2). The enhancement of surface expression of CD16 _TM by ζ was more pronounced than the amplification by γ (Fig. 2) in the tested cell lines, while native CD4 did not enhance the surface expression of CD16 _TM .

EXAMPLE IV
Asp ζ mutants do not bind to the Fc receptor
To create chimeras that would not bind to an existing antigen or Fc receptor, mutant ζ fusion proteins were obtained that did not contain either an intram Membrane Asp, an Intram Membrane Cys, or both. Flow cytometry showed that the surface expression intensity of the various mutant chimeras was not significantly different from the unmuted predecessor, and immunoprecipitation experiments showed that the total expression of the chimeras was the same. As expected, it was found that mutant chimeras lacking transmembrane cysteine residues do not form disulfide-bonded dimers. Two mutant chimeras lacking Asp were unable to support surface expression of CD16 _TM , while monomeric chimeras lacking Cys but bearing Asp made co-expression of CD16 _TM possible, but at a lower efficiency than the parent dimer (FIG. . 3).

EXAMPLE V
Mutant receptors retain the ability to initiate a calcium response
To determine whether cross-linking of fusion proteins would allow accumulation of free intracellular calcium, similar to that occurring in the case of a T-cell antigen receptor, human T-cell leukemia cell lines, Jurkat E6 (ATCC Accession N TIB 152, American Type Culture Collection, Rockville, MD) were infected with vaccinia virus recombinants and the relative concentration of calcium in the cytoplasm was measured after crosslinking of the extracellular domain with antibodies. Flow cytometry measurements were performed with cells loaded with calcium-sensitive Indo-1 dye (Grynkiewicz et al., J. Biol. Chem. 260: 3340-3450 (1985); Rabinovitch et al., J. limnunol. 137: 952-961 (1986)). FIG. 4A-D shows the results of experiments with the release of calcium by cells infected with CD4: ζ and Asp ^- and Cys ^- ζ mutants. Crosslinking of chimeras reproducibly increased intracellular amounts of calcium. CD4: η and CD4: γ also made possible the accumulation of intracellular calcium in infected cells. Jurkat cells express low levels of CD4 on the cell surface, however, crosslinking of native CD4 in the presence or absence of CD16: ζ does not alter intracellular calcium levels (Fig. 4A-B).

EXAMPLE VI
Chimeras CD4: ζ, η and γ mediate cytolysis of target cells expressing HIV gp 120/41
To determine whether chimeric receptors trigger cytolytic effector programs, a system was created based on recognition of the CD4 complex of the gp120 / gp41 HIV envelope. HeLa cells were infected with recombinant vaccinia viruses expressing gp120 / gp41 (Chakrabarti et al., Nature 320: 535-537 (1986); Earl et al., J. Virol. 64: 2448-2451 (1990)) and ⁵¹ Cr. Labeled cells were incubated with cells from the line of allospecific (CD8 ⁺ , CD4 ^- ) cytotoxic T lymphocytes that were infected with vaccinia virus recombinants expressing CD4: ζ CD4: η or CD4: γ chimeras or double mutant CD4: ζ CysllGly: Asp15Gly chimeras . FIG. 5A-C shows that HeLa cells expressing gp120 / 41 were specifically lysed by cytotoxic T lymphocytes (CTLs) expressing CD4 chimeras. Uninfected HeLa cells were not CTL targets carrying CD4: ζ chimeras, and HeLa cells expressing gp120 / 41 were not recognized by uninfected CTL. To compare the effectiveness of different chimeras, the effector: target ratios were adjusted for the CTL fraction expressing CD4 chimeras and the fraction of NeLa cells expressing gp120 / 41, measured by flow cytometry. FIG. 5C shows a cytometric analysis of CT4 CD4 expression used in the cytolysis experiment shown in FIG. 5A and 5B. Although the average density of surface CD4: ζ was significantly higher than the average density of CD4: η, the cytolytic efficiencies of cells expressing any form were similar. When corrected for the fraction of target cells expressing gp20, the efficiency of cytolysis mediated by CD4 proteins; ζ and CD4: η, was comparable to the highest efficiencies reported for target: effector specific T-cell receptor pairs (average effector: target ratio for 50% excretion by T-cells expressing CD4: ζ was 1.9 ± 0.99, n = 10 The CD4: γ fusion protein was less active, as was the CD4: ζ fusion protein lacking Asp and Cys transmembrane residues, however, significant cytolysis was observed in both cases (Fig. 5B-C).

To control that vaccinia virus infection can enhance CTL artifact recognition, similar cytolysis experiments were performed with target cells infected with vaccinia virus recombinants expressing phosphatidylinositol bound form CD16 (CD16 _PI ) and labeled with ⁵¹ Cr, and with CTL, infected control recombinants expressing either CD16 _PI or CD16: ζ FIG. 6A shows that T cells expressing non-CD4 chimeras do not recognize HeLa cells expressing other surface proteins encoded by vaccinia virus. In addition, CTLs expressing non-chimeric CD4 did not significantly lyse HeLa cells expressing gp120 / 41 (FIG. 6A).

EXAMPLE VII
Cells carrying MHC class II are not targets of chimeras
CD4 is believed to interact with a non-polymorphic sequence expressed by the MHC class II antigen (Gay et al. Nature 328: 626-629 (1987); SlecKman et al., Nature 328: 351-353 (1987)). Although the specific interaction between CD4 and class II antigen has never been documented with purified proteins, under certain conditions, adhesion can be demonstrated between cells expressing CD4 and cells expressing class II molecules (Doyle et al. Nature 330: 256-259 (1987) ; Clayton et al., J. Exp. Med. 172: 1243-1253 (1990); Lamarre et al., Science 245: 743-746 (1989)). Then it was necessary to test whether it was possible to detect the killing of cells carrying the class II antigen. FIG. 6B shows that there is no specific cytolysis directed by the CD4: ζ chimera against the Raji B cell line, which expresses large amounts of class II antigen. Although moderate cytolysis (approximately 5%) was observed, a class II antigen-free Raji mutant, RJ2.2.5 (Accolla, J. Exp. Med. 157: 1053-1058 (1983)) showed a similar sensitivity as incubated Raji cells with uninfected T cells.

Е. coli в качестве селектирующего маркера, и встраивали в геном штамма WR вируса коровьей оспы гомологичной рекомбинацией и отбором на рост в микофенольной кислоте (Falkner and Moss, J. Virol. 62:1849 (1988); Boyle and Coupar, Gene 65:123 (1988)).EXAMPLE VIII
Sequence Requirements for the Induction of Cytolysis of the Chimeric T-Cell Antigen / Zeta Fc Receptor Chain
Although chimeras between CD4 and ζ can arm cytotoxic T lymphocytes (CTLs) to kill target cells expressing HIV gp120, alternatives have been sought for CD4 to unambiguously compare the properties of the zeta chimeras introduced into human T cell lines. Such lines can express CD4, making it difficult to specifically determine the relationship between the type or degree of calcium mobilization and the cytotoxic potential of various chimeras. To overcome this, chimeras between ζ and CD16 were created in which the extracellular domain of CD16 is attached to the transmembrane and intracellular ζ sequences (Fig. 7A). These gene fusions introduced the vaccinia virus carrying the gene into the expression plasmid

E. coli as a selection marker, and was inserted into the genome of the WR strain of vaccinia virus by homologous recombination and selection for growth in mycophenolic acid (Falkner and Moss, J. Virol. 62: 1849 (1988); Boyle and Coupar, Gene 65: 123 (1988)).

 T cell lines were infected with vaccinia virus recombinants and the relative concentrations of the free calcium ion in the cytoplasm were measured after crosslinking of the extracellular domains with antibodies. Both spectrofluorimetric (general population) and flow cytometric (single cells) measurements were performed with cells loaded with Indo-I dye (Glynkiewicz et al., J. Biol. Chem. 260: 3440 (1985); Rabinovitch et al., J. Immmunol. 137: 952 (1986)). FIG. 7B shows an analysis of data collected from human Jurkat T cell leukemia cells infected with vaccinia virus recombinants expressing a CD16: ζ fusion protein. Crosslinking of these chimeras reproducibly increased intracellular calcium, while a similar treatment of cells expressing non-chimeric CD16 gave little or no effect. When expressing this chimera in mutant cell lines lacking an antigen receptor, either REX33A (Breitmeyer et al., J. Immunol. 138: 726 (1987); Sancho et al., J. Biol. Chem. 264: 20760 (1989) ), or Jurkat mutant JRT3. T3.5 (Weiss et al., J. Immunol. 135: 123 (1984)), a strong response to crosslinked with antibody CD16 was observed. Similar data were obtained on the mutant cell line REX20A (Breitmeyer et al., Supra, 1987; Blumberg et al., J. Biol. Chem. 265: 14036 (1990)) and the CD3 / Ti negative Jurkat cell mutant obtained in this laboratory . Infection with recombinants expressing CD16: ζ did not restore the response to anti-CD3 antibodies, indicating that this fusion protein did not act by rescuing the intracellular chains of the CD3 complex.

 To assess the ability of these chimeras to change the direction of cell-mediated immunity, CTLs were infected with vaccinia virus recombinants expressing CD16 chimeras and used for specific lysis of hybridoma cells expressing membrane-bound anti-CD16 antibodies. This assay is a continuation of the hybridoma cytotoxicity assay, originally developed to analyze the effector mechanisms of cells carrying Fc receptors (Graziano and Fanger, J. Immunol. 138: 945, 1987; Graziano and Fanger, J. Immunol. 129: 35-36, 1987 ; Shen et al., Mol. Immunol, 26: 959, 1989; Fander et al., Immunol. Today 10:92, 1989). FIG. 8B shows that the expression of CD16: ζ cytotoxic T lymphocytes allows CTL with a receptor to kill 3G8 (anti-CD16; Fleit et al., Proc. Nail. Acad. Sci. USA 79: 3275, 1982) hybridoma cells, whereas CTL, Expressing the phosphatidylinositol-associated form of CD16 are inactive. CTLs coupled to CD16: ζ also do not kill hybridoma cells expressing non-relevant antibodies.

To identify the minimal ζ sequences necessary for cytolysis, a series of deletion mutants were obtained in which a successively larger part of the intracellular domain ζ was removed from the carboxyl terminus (Fig. 8A). Most of the intracellular domain of zeta could be removed with little effect on the cytolytic potential; the full-length chimera CD16: ζ was essentially equal in efficiency to the chimera deleted to residue 65, CD16: ζ Asp66 ^* (Fig. 8B). A significant decrease in cytotoxicity was observed during deletion to residue 59 ζ (chimera CD16: ζ Glu60 ^* ) and further deletion to residue 50 resulted in slightly lower activity. However, a complete loss of activity was not observed, even with a decrease in the intracellular domain to three residues of the transmembrane anchor (Fig. 8B).

Since ζ is a disulfide-bonded dimer, one of the explanations for the conservation of cytolytic activity was that endogenous ζ formed heterodimers with a chimeric ζ deletion, thereby restoring activity. To test this idea, residues 11 and 15 ζ were replaced (Gly was introduced instead of Asp and Cys) (CysllGly / Asp15Gly) and immunoprecipitation was carried out as follows. About 2 x 10 ⁶ CVI cells were infected for 1 hour in serum-free DME with recombinant vaccinia virus with a multiplicity of infection (moi) of at least 10. 6-8 hours after infection, cells were separated from the plates with PBS / 1 mM EDTA and superficially labeled 0.2 mCi ¹²⁵ I on 2 • 10 ⁶ cells using lactoperoxidase and H ₂ O ₂ according to the method of Clark and Einfeld (Laucocyte Typing II, pp 155-167, Springer-Verlag, NY, 1986). Labeled cells were collected by centrifugation and lysed in 1% NP-40, 0.1% SDS, 0.15 M NaCl, 0.05 M Tris, pH 8.0, 5 mM MgCl ₂ , 5 mM KCl, 0.2 M iodoacetamide and 1 mM PMSF. Nuclei were removed by centrifugation and CD16 proteins were immunoprecipitated with agarose with 3G8 antibody (Fleit et al., Supra, 1982; Medarex) and anti-mouse IgG (Cappel, Durham, NC). Samples were subjected to electrophoresis on an 8% polyacrylamide gel with SDS under non-reducing conditions or on a 10% gel under reducing conditions. These immunoprecipitations confirmed that the CD16: ζ CysllGly / Asp15Gly chimera was not associated with disulfide-bonded dimeric structures.

The cytolytic activity of mutant receptors was also tested. The mutated chimera deleted to residue 65 (CD16: ζ CysllGly / Asp15Gly / Asp66 ^* ) was, depending on the test conditions, 2-8 times less active in the analysis of cytolysis than the comparable non-mutated chimera (CD16: ζ Asp66 ^* ), which was usually about 2 times more active or indistinguishable in activity from CD16: ζ (Fig. 9B). The decrease in the activity of mutant chimeras is comparable to the decrease observed with CD4 chimeras of a similar structure (see above) and is most likely due to the lower efficiency of ζ monomers compared to dimers. In contrast, the Asp ^- , Cys ^- mutated chimera deleted to residue 59 did not have cytolytic activity, confirming the hypothesis that the association with other chains mediated by transmembrane residues of Cys and / or Asp was responsible for the poor persistence of cytolytic activity in deletion amino end further than to residue 65.

Studies using flow cytometry showed that deletion mutants lacking Asp and Cys transmembrane residues could still accelerate the increase in the amount of free calcium ion inside the cell in response to cross-linking with antibodies in the TCR of the mutant Jurkat cell line (Fig. 9D). Similar results were obtained for chimeras expressed in the Jurkat parent line. In the case of CD16: ζ CysllGly / Asp15Gly / Glu60 ^*, these data show that the ability to mediate a “calcium” response can be mutationally separated from the ability to maintain cytolysis.

To specifically exclude the possible contribution of transmembrane ζ residues, the transmembrane and first 17 cytoplasmic ζ residues were replaced by sequences encoding the CD5 or CD antigens passing through the membrane and the first 14 or first 17 cytoplasmic residues, respectively (Fig. 9A). The obtained three-part fusion proteins CD16: 5: ζ (48-65) and CD16: 7: ζ (48-65) did not form disulfide-bound dimers, which form simpler CD16: ζ chimeras, since they do not contain a cysteine residue in the transmembrane domain ζ. Both consist of three parts chimeras were able to mobilize calcium in Jurkat and TCR lines cells ^- -cells (. Figure 9D) and increase the cytolytic response in CTL (Fig 9C and data not shown.). However, shortening the ζ part to residue 59 in the CD16: ζ chimera (48-59) cancels the ability of the three-part fusion protein to trigger a calcium mobilization reaction in TCR positive or negative Jurkat cells or cytolysis in mature CTLs (Figs. 9C and 9D and not shown data).

 To study the contributions of individual residues within the motif from 18 residues, we obtained a number of mutant variants by site-specific mutagenesis and evaluated their ability to mediate receptor-directed killing under conditions of low (Fig. 10A and 10D) or high (Fig. 10B and 10E) chimeric receptor expression. FIG. 10A-F shows that although a number of relatively conservative substitutions (i.e., substituting acidic residues with their related amides or replacing tyrosine with phenylalanine) at residues 59-63 gave a moderate decrease in cytolytic efficacy, in general these options retained the ability to mobilize calcium. However, when taken together, these residues contain an important submotif because their deletion excludes cytolytic activity. The conversion of Tyr 62 to either Phe or Ser excluded both the cytolytic response and the calcium mobilization reaction. At the amino terminus of a segment of 18 residues, the replacement of Tyr 51 with Phe abolished both the mobilization of calcium and cytolytic activity, while the replacement of Leu with Ser at position 50 excluded the calcium response, while only partially weakened cytolysis. Without being tied to a specific hypothesis, it is assumed that the inability of the Leu50Ser mutant to mobilize calcium in short-term flow cytometric analyzes does not fully reflect its ability to mediate a significant increase in free intracellular calcium over a longer period of time for cytolysis analysis. However, calcium insensitive cytolytic activity has been reported for some cytolytic T-cell lines, and it is possible that a similar phenomenon underlies the results described here. Replacing Asn 48 with Ser weakened cytotoxicity in some experiments, while in other experiments it had a weak effect.

 To study the potential role of excess elements of the sequence, the intracellular domain ζ was divided into three segments, covering residues 33–65, 71–104, and 104–138. Each of these segments was attached to the CD16: CD7 chimera via a MluI site inserted distally to the main membrane anchor sequences of the CD7 intracellular domain (see below; FIG. 11A). A comparison of the cytolytic efficacy of these three elements showed that they are essentially equipotent (Fig. 11B). A comparison of the sequences (FIG. 11A) shows that the second motif carries 11 residues between tyrosines, while the first and second motifs carry 10.

 Although an accurate assessment of the T-cell activation process has not been made, it is clear that aggregation of the antigen receptor or receptor chimeras carrying intracellular ζ sequences triggers the mobilization of calcium, the release of cytokines and granules, and the appearance of surface activation markers. The active site ζ, a short linear peptide sequence, perhaps too small to have its own enzymatic activity, apparently interacts with one or at most several proteins to mediate cell activation. It is also clear that mobilization of free calcium alone is insufficient for cellular activation, because the ability to mediate cytolysis can be separated by mutations from the ability to mediate calcium accumulation.

 As shown here, the addition of 18 residues from the intracellular domain ζ to the transmembrane and intracellular domain of two unrelated proteins allows the obtained chimeras to redirect cytolytic activity against target cells that bind to the extracellular portion of fusion proteins. Although chimeras carrying a motif of 18 residues are approximately 8 times less active than full-length ζ chimeras, reduced activity can be attributed to the loss of transmembrane interactions, which usually allow wild-type ζ to form dimers bound by disulfide bonds. Those. deletion constructs ζ, which have the same carboxyl end as this motif and do not have transmembrane residues Cys and Asp, usually exhibit slightly lower activity than chimeras bearing only the motif of 18 residues.

 The element of cytolytic competence on which we have concentrated attention has two tyrosines and does not have serine or threonine residues that exclude possible phosphorylation contributions to activity. A mutation of any tyrosine disrupts activity, and although preliminary experiments do not show significant tyrosine phosphorylation after crosslinking with chimeric surface antigens bearing a motif of 18 residues, the possible participation of such phosphorylation at a low level cannot be ruled out. In addition to the noted effects with respect to the two tyrosine residues, a number of amino acid substitutions at the amino and carboxy-terminus of this motive weaken activity in conditions of low receptor density.

 Sequences similar to the ζ active motif can be found in the cytoplasmic domains of some other transmembrane proteins, including CD3 δ and γ molecules, mb1 and B29 proteins located on the cell surface and associated with IgM, and β and γ IgE high affinity receptor chains, Fc ∈ RI (Reth, Nature 338: 383, 1989). Although the function of these sequences is unclear, when expressed efficiently, each of them may be capable of autonomously activating T cells, and such activity may explain the residual TCR response observed in the zeta-negative mutant cell line (Sussman et al. Cell 52:85 1988).

 The zeta itself carries three such sequences, at approximately the same distance from one another, and a rough dissection into three parts of the intracellular domain shows that each of them is capable of initiating a cytolytic response. η formed by splicing the ζ isoform (Jin et. al., supra, 1990; Clayton et al., Proc. Natl. Acad. Sci. USA 88: 5202, 1991) does not have a carboxyl half of the third motive. Since the removal of the carboxyl half of the first motive destroys the activity, it seems likely that the bulk of the biological effectiveness of η may be associated with the first two motives. Although η is as active as ζ in various dimensions, in enhancing antigen-mediated cytokine release (Bauer et al., Proc. Natl. Acad. Sci. USA 88: 3842, 1991) or in changing the direction of cytolysis (see above), η not phosphorylated in response to receptor stimulation (Bauer et al., supra, 1991). Thus, either all three motifs are required for phosphorylation, or the third motif is the preferred substrate for the unidentified tyrosine kinase.

EXAMPLE IX
Transduction (transmission) of cytolytic signal by human Fc receptor
To evaluate the actions of the various subtypes of the human Fc receptor, chimeric molecules were created in which the extracellular domain of the human CD4, CD5, or CD16 antigens were connected to the transmembrane and intracellular domains of the B1, B2, and C subtypes of FcR11 γ A (nomenclature Ravetch and Kinet, Ann. Rev. Ann. Immunol. 9: 457, 1991). In particular, cDNA sequences corresponding to the transmembrane and cytoplasmic domains of the previously described B1 and B2 FcR11A isoforms were amplified from a preexisting PC23 clone or from a human minadine cDNA library (constructed by standard methods) using the following synthetic oligonucleotide primers:
CCC GGA TCC CAG CAT GGG CAG CTC TT (SEQ ID N 18; FcR11 A direct);
CGC GGG GCG GCC GCT TTA GTT ATT ACT GTT GAC ATG GTC GTT (SEQ ID N 19; FcR11 A reverse);
GCG GGG GGA TCC CAC TGT CCA AGC TCC CAG CTC TTC ACC G (SEQ ID N 20; FcR11 B1 and FcR11 B2 reverse);
GCG GGG GCG GCC GCC TAA ATA CGG TTC TGG TC (SEQ ID N 21; FcR11 B1 and FcR11 B2 reverse).

These primers contained cleavage sites for the BamHI and NotI enzymes, respectively, protruding 6 residues from the 5 ′ end. The NotI site was directly followed by the antisense stop codon, either CTA or TTA. All primers contained 18 or more residues complementary to the 5 ′ and 3 ′ ends of the desired fragments. The cDNA fragment corresponding to the cytoplasmic domain of FcR11 γ C, which differs from the 11A isoform in only one amino acid (L instead of P at residue 268), was obtained by site-specific mutagenesis using overlapping PCR using primers having the sequences:
TCA GAA AGA GAC AAC CTG AAG AAA CCA ACA A (SEQ ID N 22) and
TTG TTG GTT TCT TCA GGT TGT GTC TTT CTG A (SEQ ID N 23).

Е. coli для облегчения идентификации желаемых рекомбинантов. Идентичность всех изоформ (показанных в фиг. 12) подтверждали дидезокси-секвенированием.These PCR fragments (PCR) were inserted into expression vectors (vaccinia viruses) that contained the extracellular domains of CD16 or CD4, respectively, and then inserted into wild-type vaccinia virus by recombination at the thymidine kinase locus using co-integration selection

E. coli to facilitate identification of the desired recombinants. The identity of all isoforms (shown in FIG. 12) was confirmed by dideoxy sequencing.

The preparation of chimeric receptor proteins was further confirmed by immunoprecipitation studies. Approximately 10 ⁷ JRT3.T3.5 cells were infected for 1 hour in serum-free IMDM medium with recombinant vaccinia virus with a multiplicity of infection (moi) of at least 10. 12 hours after infection, cells were harvested and superficially labeled with 0.5 mCi ¹²⁵ I at 10 ⁷ cells using the lactoperoxidase / glucose oxidase method (Clark and Einfeld, supra). Labeled cells were collected by centrifugation and lysed with 1% NP-40, 0.1 mM MgCl ₂ , 5 mM KCl, 0.2 M iodoacetamide and 1 mM PMSF. Nuclei were removed by centrifugation and CD16 fusion proteins were immunoprecipitated with agarose with 4G8 antibodies and anti-mouse IgG. Samples were subjected to electrophoresis under reducing conditions. All immunoprecipitated chimeric receptor molecules had the expected molecular weights.

To test the ability of the chimeric receptors to mediate an increase in cytoplasmic free calcium ion, the recombinant viruses were used for infection TCR ^- mutant cell line Jurkat JRT3.T3.5 (as described above) and cytoplasmic free calcium was measured in the cells (as described herein) following crosslinking of the extracellular domains monoclonal receptor 3G8 or Leu-3A (as described here). These experiments found that the intracellular domains of FcR γ 11 A and C were able to mediate an increase in the cytoplasmic free calcium ion after crosslinking of the extracellular domains, while the intracellular domains FcR γ 11 B1 and B2 were inactive under comparable conditions (Figs. 13A and 13B). Hybridomas CD4, CD5 and CD16 FcR γ 11 A shared virtually equal ability to enhance the calcium response (Fig. 13A-B). Other cell lines, both monocytic and lymphocytic differentiation lines, were able to respond to a signal initiated by crosslinking of extracellular domains.

To determine the involvement of various intracellular domains of FcR γ 11 in cytolysis, human cytotoxic T lymphocytes (CTR) were infected with vaccinia virus recombinants expressing CD16 chimeras: FcR γ 11 A, B1, B2 and C. Then, infected cells were co-cultured with ⁵¹ Cr loaded hybridoma cells (i.e., 3G8 10-2 cells) that expressed an anti-CD16 antibody on the cell surface. In this assay, CTLs carrying the CD16 chimera killed the hybridoma target cells (allowing free ⁵¹ Cr to be secreted) if the extracellular domain of the chimera was connected to an intracellular segment capable of activating an effector lymphocyte program; this cytolysis assay is described in detail below. FIG. 14A shows that CTLs coupled to CD16: FcR γ 11A and C, but not FcR γ 11 B1 or B2, are capable of lysing target cells expressing surface anti-CD16 antibodies.

 To exclude the possibility that this specific cytolysis was associated with interaction with the CD16 part, cytolysis experiments were carried out in which the intracellular domains of FcR11 were joined to the extracellular domain of CD4. In this case, the target cells were HeLa cells expressing HIV membrane gp120 / 41 proteins (in particular, HeLa cells infected with the vPE16 vaccinia virus vector (available from the National Institute of Allergy and Infections Disease AIDS Depository, Bethesda, MD). Like in the CD16 system, target cells expressing HIV envelope antigens were susceptible to lysis by T cells expressing the CD4 chimera: FcR γ 11 A but not FcR γ 11 B1 or B2 (Fig. 14B).

 The intracellular domains FcR γ 11 A and C do not have a measurable sequence homology with any other protein, including members of the extended FcR γ / TCR ζ family. To determine the sequence elements responsible for the induction of cytolysis, 5'- and 3'-deletions of the sequences encoding the intracellular domain (described below and shown in Fig. 15A) were obtained and evaluated for their effectiveness in calcium mobilization and cytolysis (as described here). In experiments in which the amino-terminal part of the intracellular domain was removed, the FcR γ 11 transmembrane domain was replaced with the transmembrane domain of the unrelated CD7 antigen to exclude the possible contribution of interactions mediated by the membrane-crossing domain.

 FIG. 15B and 15C show that the removal of 14 carboxyl-terminal residues, including tyrosine 298, resulted in a complete loss of cytolytic ability and a significant decrease in calcium mobilization potential. Further, a deletion to a position in front of tyrosine 282 gave an identical phenotype (Fig. 15B and 15C). Deletion from the N-terminus of the intracellular domain to residue 268 had no significant effect on either the calcium profile or cytolytic activity, while deletion to residue 275 markedly attenuated the release of free calcium, but had little effect on cytolysis (Figs. 15D and 15E). A further deletion, up to residue 282, gave FcR γ 11 tails, which did not have the ability to either mobilize calcium or trigger cytolysis (Fig. 15D and 15E). The "active element" defined by these crude measurements is relatively large (36 amino acids) and contains two tyrosines, separated by 16 residues.

EXAMPLE X
Directed cytolysis by lymphocytes carrying chimeric CD4 receptors that do not support infection
As discussed above, effector molecules can be constructed that direct the cytolytic activity of CTL in an MHC-independent manner. For example, a chimera composed of the extracellular domain of CD4 fused to the ζ chain in the human CTL clone, WH3, specifically kills target cells that detect HIV-1 surface glycoprotein, gp120. However, since the extracellular domain of the CD4 molecule confers sensitivity to HIV infection, such CTLs with immunoglobulin receptors transferred to them can become targets for the virus, resulting in a decrease in their activity (Dalgleish et al. Nature 312: 767 (1984); Klatzman et al. Nature 312: 767 (1984)). To prevent such a result, CD4-based effector molecules were designed that are effective in specifically targeting HIV-infected cells for cell-mediated killing, but which do not confer sensitivity to HIV infection.

 A three-part fusion protein was created by genetic apposition of the extracellular domain of CD4 (FIG. 23) to the hinge region, the second and third constant domains of the human IgG1 heavy chain (Zettlmeissl et al., DNA Cell Biol. 9: 347 (1990)) (FIG. 25), which in this case were attached to a portion of the first transmembrane exon of a membrane-bound human IgGI, followed by a portion of the human CD7 antigen consisting of sequences between a single Ig-like domain and a stop transmission sequence after the transmembrane domain (Aruffo and Seed, EMBO J. 6: 3313 (1987)) (f 26 g). The primary amino acid sequence of the extracellular portion of the CD7 segment consisted of a proline-rich region suggesting a stem-like structure that pushes the Ig-like domain away from the cell surface (Aruffo and Seed EMBO J. 6: 3313 (1987)) (FIG. 26). Recombinant vaccinia viruses were prepared for expression of this fusion protein and related chimeras, as described herein. In particular, recombinant vaccinia viruses were obtained by homologous recombination in CV-1 cells. At least two plaque visualization cycles were performed for each strain before obtaining high titer strains in CV-1 cells.

 A three-part chimera (CD4 (D1-D4): lg: CD7) (Fig. 20, molecule “A”) showed effective surface expression and was tested for its ability to act as an HIV receptor in a vaccinia virus test for measles pox virus (Lifson et al. Nature 323: 725 (1986)); Ashorn et al., J. Virol. 64: 2149 (1990)). HeLa cells infected with recombinant vaccinia virus (VPE16) encoding HIV-1 sheath glycoprotein (Earl et al., J. Virol. 64: 2448 (1990)) were co-cultured with HeLa cells infected with either CD4, CD4: ζ or CD4 (D1-D4): Iq: CD7. 6 cm plates with HeLa cells (ATCC, Rockville, MD) at 50% confluency were infected in serum-free medium for 1 hour with an approximate multiplicity of infection (moi) 10. Cells were incubated for another 5-6 hours in complete medium and then separated containing phosphate saline (PBS) with 1 mM EDTA. Cells expressing the membrane and CD4 chimera were mixed in a 1: 1 ratio and subcultured into 6 cm plates with complete medium. Syncytia was scored at 6-8 hours of co-cultivation and photographed.

Co-cultures of CD4 and vPE16 led to the formation of easily detectable multinucleated giant cells. Also, a chimera consisting of the extracellular domain of CD4 fused to the TCR ζ chain (FIG. 27) (CD4: ζ) was able to mediate the formation of syncytia, while cells expressing CD4 (D1-D4): Ig: CD7 are not showed signs of cell fusion. We also tested a construct expressing only the first and second domains of CD4 (Fig. 24), CD4 (D1, D2): Ig: CD7 (Fig. 20, molecule “B”), since in another connection it was shown that amino terminal two CD4 domains are necessary for HIV infectivity (Landau et al. Nature 334: 159 (1988)). This molecule also showed immunity to HIV-induced syncytia formation. A study of binding to soluble ¹²⁵ I-labeled gpl20 showed that both CD4 (D1-D4): Ig: CD7 and CD4 (D1, D2): Ig: CD7 had undisturbed affinity for gp120.

 We then determined whether chimeric molecules that resisted the formation of syncytia would be able to target killing cells while providing them with a trigger part, as described here. We fused the intracellular domain ζ (Fig. 27) with the 3'-end of CD4 (D1-D4): Ig: CD7 and CD4 (D1, D2): Ig: CD7 and prepared the corresponding vaccinia viruses. These constructs, CD4 (D1-D4): Ig: CD7: ζ CD4 (D1, D2): Ig: CD7: ζ (Fig. 20, molecules "C" and "D") were expressed in human WH3 clone CTL and tested for their ability to target and kill HeLa cells expressing the surface envelope glycoprotein of the HIV membrane (using the methods described here). FIG. 21 shows that the intracellular domain ζ, fused with either CD4- (D1-D4): Ig: CD7 or CD4 (D1, D2): Ig: CD7, can report killing ability; constructs lacking a ζ chain were unable to mediate this activity. CD4: ζ, a positive control, mediated slightly more effective cytotoxicity, and CD4 (D1, D2): Ig: CD7: ζ - a slightly less effective cytotoxicity than CD4 (D1-D4): Ig: CD7: ζ (Fig. 21) . However, it is clear that both CD4 (D1-D4): Ig: CD7: ζ and CD4 (D1, D2): Ig: CD7: ζ have the ability to mediate the specific killing of cells expressing HIV envelope proteins on their surface. These four-part chimeras are persistently unable to mediate the formation of syncytia in the vaccinia virus test. We also demonstrated that the only ζ motif shown in FIG. 11A is sufficient to impart cytolytic activity to the CD4 chimera (D1-D4).

Radioimmunoprecipitation experiments showed that fusion molecules were predominantly (if not always) dimers. In these experiments, protein A agarose granules were used to immunoprecipitate a solubilized extract of metabolically labeled HeLa cells infected with recombinant vaccinia virus expressing chimeras CD4- (D1-D4): Ig: CD7: ζ and CD4 (D1, D2): Ig: CD7: ζ. The immunoprecipitated material was fractionated by polyacrylamide gel electrophoresis under reducing and non-reducing conditions. In particular, approximately 5 x 10 ⁶ HeLa-S3 cells were infected, as described above for vPE16, with a suitable strain of vaccinia virus. Cells were metabolically labeled with 200 μCi / ml Tran ³⁵ S-Label (ICN Radiochemicals, Irvine, Calif.) For 6-8 hours in cysteine and methionine-free medium. They were separated using PBS containing 1 mm EDTA. Then the cells were besieged and lysed in 150 mM NaCl, 50 mM Tris pH 7.5, 5 mM EDTA, 0.5% NP-40, 0.1% SDS, 5 mM EDTA, 1 mM PMSF.

After removing the nuclei by centrifugation, one fifth of each cell extract was adsorbed on washed granules of protein-A conjugated agarose for 2 hours at 4 ^° C. Then, the granules were washed with PBS containing 1% NP-40 and eluted in SDS sample buffer in the presence or absence of mercaptoethanol. The results of these experiments showed that most of the immunoprecipitated CD4 (D1, D2): Ig: CD7: ζ chimeras migrated as dimers of the expected molecular weight under non-reducing conditions.

 To directly assess the ability of cells expressing CD4 fusion molecules to support HIV infection, we conducted long-term studies of infectivity on transfectants expressing CD4 (D1-D4): Ig: CD7 and CD4: (D1, D2): Ig: CD7. Stable transfectants CD4 (D1-D4): lg: CD7 and CD4 (D1, D2): lg: CD7 and CD4 were prepared in a subline of 293 cells, an easily transfected human embryonic kidney cell line. Chimeric molecules were subcloned in bidirectional vectors in which the hygromycin B gene was controlled by the herpes simplex virus thymidine kinase promoter. 10 cm plates with 60-70% confluency were transfected with 10 μg of this plasmid DNA co-precipitated with calcium phosphate. Before transfection, plasmids were linearized at the unique Sfi 1 site and the ends were complemented with T4 DNA polymerase. At 24 hours after transfection, the cells were diluted 4 times, and at 48 hours after transfection, hygromycin B (Sigma, St. Louis, MO) was selected at 400 μg / ml. Every 3-4 days, the cells were supplied with fresh medium containing hygromycin.

 Resistant colonies were scrubbed, propagated, and their expression was evaluated by indirect immunofluorescence using fluorescein conjugated anti-human Fc IgG antibodies (Organon Teknika, West Chester, PA) or Q4120, an antibody reactive with human CD4 (Sigma), followed by flow cytometry (Coul Hialeah, FH). Two independent clones from each construct were selected for analysis with surface CD4 levels comparable to those detected by other cell lines. FIG. 22 shows that after exposure to HIV, p24 was detected in stable CD4 transfected cultures as early as 3 days after infection. The presence of multinucleated giant cells and characteristic bloating was observed as early as 5 days after infection in these cultures. No significant levels of p24 or multinucleated giant cells were found in the non-transfected parent cells or in either of the two independently obtained transfectant isolates CD4 (D1-D4): Ig: CD7 and CD4 (D1, D2): Ig: CD7 after 32 days of cultivation (Fig. 22).

 After completion of the infectivity studies, the cells were analyzed for expression of CD4 on the cell surface. The density of the surface CD4 epitope was significantly reduced in infected cultures expressing CD4, which is consistent with viral negative modulation, but was undisturbed in cultures expressing CD4 (D1-D4): Ig: CD7 and CD4 (D1, D2): Ig: CD7. These experiments show that it is possible to create chimeric molecules that carry two apical CD4 domains, which, when fused with the ζ-chain of the T-cell receptor, can target HIV-infected cells and kill them, while not supporting a CD4-mediated HIV infection.

 Additional experiments suggest that it is the physical distance between the extracellular domain of the CD4 molecule and the lipid bilayer that makes it possible to resist HIV infection. In the first experiment, we constructed a chimeric molecule having a deletion of the stem and transmembrane domain of CD7; this division removed the proline-rich region of the transmembrane portion of CD7. Upon fusion of this domain with the extracellular domain of CD4, it retained its ability to efficiently attach to the extracellular domain of the CD4 molecule, as measured by surface expression of the CD4 molecule (as described here). However, the ability to withstand the formation of syncytia induced by the HIV envelope glycoprotein has been lost. Thus, the deletion of the proline-rich part of the CD7 molecule, a region capable of apparently forming an α-helical twisted structure, effectively reduced the distance between the extracellular domain of CD4 and the lipid bilayer and destroyed the ability of the chimera to resist the formation of syncytia.

 In the second experiment, we showed that the ability to withstand HIV-induced syncytia formation can be provided for the CD4 / CD5 chimera, which, as shown earlier, serves as a transmembrane anchor for the extracellular domain of CD4, but is not able to withstand the HIV-induced syncytium formation. In this experiment, the hinge region, the CH2 and CH3 domains of the human IgG1 heavy chain were inserted into the CD4 / CD5 molecule; the resulting chimera resisted the formation of syncytium, which again suggests that the distance provided by the immunoglobulin domains is sufficient to confer resistance to HIV-induced syncytium formation.

 In the third experiment, the CD4 domain was removed at varying distances from the cell membrane using synthetic α-helices of various lengths. In particular, synthetic oligonucleotides were constructed that carried repeating α-helical motifs of lysine and glutamic acid residues flanked by two alanine residues (see FIG. 28 for primary nucleic acid and amino acid sequences). In earlier studies, it was found that such amino acid sequences occur with high frequency in alpha helices, suggesting that such repeating motifs might take an alpha-helical conformation and that placing such alpha-helices between the transmembrane domain and the extracellular domains of CD4 advanced would CD4 away from the cell membrane. By varying the length of the alpha-helical segment, the extension distance necessary to withstand HIV entry was calculated based on the known alpha-helix and turn heights. These results are presented in Table 1.

 In this table, “CD4” means CD4 (D1-D4), unless otherwise indicated; "H", "CH2" and "CH3" denote the hinge region, regions of the human IgG1 heavy chain CH2 and CH3, respectively; "CD7-tm" and "stk" denote the transmembrane region and the region of the CD7 stem; "CD7tm (long version)" and "CD7tm (short version)" respectively denote the transmembrane region of CD7 and the transmembrane region of CD7 having a deletion of a proline-rich domain, as discussed above; "CD5tm" refers to the transmembrane region of CD5; and "CD34tm" refers to the transmembrane region of CD34. In lines J-L, the length of the alpha-helical region is given in angstroms; these values are based on the fact that 3.6 residues per turn of the alpha helix, which corresponds to 5.4 angstroms (or 1.5 A per residue).

 Accordingly, an alpha helix of 16 residues should extend the extracellular domain of CD4 by approximately 24 angstroms. Alpha helices 48 and 72 angstroms were constructed by sequentially concatemerizing the BstYl fragment to the unique site of the BamH1 fragment (see Fig. 28), followed by selection of clones with the correct orientation.

 Syncytia formation was evaluated in co-culture tests with HeLa cells expressing HIV-1 sheath glycoprotein from the vPE-16 vaccinia virus construct (see above).

Thy-1 expression was measured as follows. A live retroviral vector based on HIV-1 clone hxb.2 was constructed. In this vector, the non-essential nef gene was replaced by the rat thy-1 coding sequence, which is efficiently expressed on the cell surface of the molecule, which attaches to the membrane with a phosphatidylinositol bond. The virus derived from this molecular clone, designated hxb / thy-1, was infectious, as shown by cytopathological effects and p24 production in the culture supernatants of infected C8166 cells (human CD4 ⁺ leukemia T cell line). In addition, upon exposure to hxb / thy-1, HeLa cells temporarily transfected with CD4 showed signs of thy-1 expression already at 18 hours after infection, as was expected from a signal regulated by the method of regulation of the nef gene. Signals encoded by the nef gene typically belong to the class of viral regulatory proteins that undergo multiple splicing and do not contain a rev-responsive element. Such signals can accumulate constitutively in the cytoplasm in the form of products of early viral genes. Thy-1 signals were expected to be regulated in this way, i.e. appear early in the life cycle of the virus. In short, this system facilitated the analysis of HIV entry, since thy-1 expression was used as a substitute for virus entry. Various CD4-based chimeras were unstably transfected into HeLa cells using standard DEAE-dextran methods. Transfected cells were exposed to hxb / thy-1 virus at 48 hours after transfection and evaluated for thy-1 expression at 24-48 hours after infection. In the results shown in table 1, the expression of thy-1 was measured at 24 hours after infection using commercially available monoclonal antibodies against Thy-1.

 Based on the data presented in Table 1, we concluded that the extracellular domains of CD4 should optimally be moved away from the cell membrane by at least 48 angstroms and preferably at least 72 angstroms to counter HIV-1 infection.

 Using a strategy similar to the general strategy described here, chimeras can be constructed based on anti-HIV membrane antibodies that target HIV-infected cells. Examples of such antibodies are described in Gorny et al., Proc. Natl. Acad Sci. USA 86: 1624 (1989) and Marasco et al., J. Clin. Invest. 90: 1467 (1992).

EXAMPLE XI
Additional trigger proteins of the T-cell receptor and B-cell receptor
Other intracellular and transmembrane signal transduction domains in accordance with this invention can be derived from T-cell receptor, CD3 delta and T3 gamma proteins, and B-cell receptor proteins, mbi and B29. The amino acid sequences of these proteins are shown in FIG. 16 (CD3 delta; SEQ ID N 24), FIG. 17 (T3 gamma; SEQ ID N 25), FIG. 18 (mD1; SEQ ID N 26) and FIG. 19 (B29; SEQ ID N 27). Parts of these sequences sufficient for transduction of the cytolytic signal (and, therefore, preferably included in the chimeric receptor of this invention) are shown in brackets (blocks). Chimeric receptors comprising these protein domains are constructed and used in the therapeutic methods of this invention, as described in general terms above.

EXAMPLE XII
EXPERIMENTAL METHODS
Infection and radioimmunoprecipitation for vaccinia virus
About 5 x 10 ⁶ CVI cells were infected for 1 hour in serum-free DME with recombinant vaccinia virus with a multiplicity of infection (moi) of at least 10. Cells were placed in fresh medium after infection and metabolically labeled with 200 μCi / ml ³⁵ S-methionine plus cysteine (Tran- ³⁶ S-label, ICN; Costo Mesa, CA) in methionine and cysteine-free DMEM (Gibco; Grand Island, NY) for 6 hours. Labeled cells were separated by PBS containing 1 mM EDTA, collected by centrifugation and lysed in 1% NP-40, 1% SDS, 0.15 M NaCl, 0.05 M Tris pH 8.0, 5 mM EDTA and 1 mM PMSF. Nuclei were removed by centrifugation and CD4 proteins were immunoprecipitated with agarose with OKT4 antibodies and anti-mouse IqG (Cappel, Durham, NC). Samples were subjected to electrophoresis on 8% SDS gels with non-reducing (NR) and reducing (R) conditions. Gels containing ³⁵ S-labeled samples were impregnated with En ³ Hance (New England Nuclear, Boston, MA) before radiography. Facilitated expression of the transmembrane form of CD16, CD16 _TM , was measured by comparing its expression in CVI cells individually infected with CD16 _TM with expression in cells co-infected with viruses encoding CD16 _TM and ζ or γ chimeras. After infection and incubation for 6 hours or more, cells were separated from the dishes using PBS, 1 mM EDTA, and the expression of CD16 _™ and chimeras was measured by indirect fluorescence and flow cytometry.

Calcium yield test
Jurkat E6 subline cells (Weiss et al., J. Immunol. 133: 123-128 (1984)) were infected with recombinant vaccinia viruses for 1 hour in serum-free IMDM at a multiplicity of infection (moi) of 10 and incubated for 3-9 hours in IMDM, 10% fetal calf serum (FBS). Cells were harvested by centrifugation and resuspended at 3 x 10 ⁶ cells / ml in complete medium containing 1 mM Indo-1 acetomethoxy ester (Grynkiewicz et al., J. Biol. Chem. 260: 3340-3450 (1985)) (Molecular Probes) and incubated at 37 ^o C for 45 minutes. Loaded Indo-I cells were pelleted and resuspended at 1 x 10 ⁶ cells / ml in serum-free IMDM and stored at room temperature in the dark. Cells were analyzed for free calcium ion by simultaneously measuring violet and blue fluorescence using flow cytometry (Rabinovitch et al., J. Immunol. 137: 952-961 (1986)). To initiate calcium release, Leu-3A antibodies (anti-CD4 antibodies) (Becton Dickinson, Lincoln Park, NJ) conjugated with phycoerythrin (PE) were added to the cell suspension at 1 μg / ml followed by 10 μg ml of unconjugated goat anti-mouse IgG at time 0 or unconjugated monoclonal antibody 3G8 (anti-CD16) at 1 μg / ml, followed by addition of 10 μg / ml phycoerythrin (PE) Fab ₂ , goat anti-mouse IgG conjugated at time 0. Histograms of the violet to blue radiation ratio were obtained from PE positive (infected) glue intramural population, which typically was 40-80% of all cells. The response of the T-cell antigen receptor in uninfected cells was triggered by OKT3 antibody, without crosslinking. For experiments involving CD16 chimeric receptors, samples showing a baseline shift toward lower intracellular calcium (without antibodies) were excluded from the analysis. The histogram data was then analyzed by converting binary data to ASC11 using Write Hand Man software (Cooper City, FL), followed by analysis with the FORTRAN software collection. The violet-to-blue emission ratio before adding the second antibody was used to establish the normalized initial ratio, taken as a unit, and the threshold resting ratio, set so that 10% of the resting population exceeded the threshold value.

Cytolysis assay
The human WH3 T cell line, CD8 ⁺ CD4 ^- HLA B44 restricted cytolytic line was maintained in IMDM with 10% human serum with 100 E / ml interleukin-2 (IL-2) and periodically stimulated or non-specifically irradiated (3000 rad) not containing HLA with peripheral blood lymphocytes and 1 μg / ml phytohemagglutinin or specifically irradiated with B44-bearing mononuclear cells. After one day of non-specific stimulation, PHA (PHA) (phytohemagglutinin) was diluted to 0.5 μg / ml by adding fresh medium and after 3 days the medium was replaced. Cells were grown for at least 10 days after stimulation before use in cytotoxicity assays. These cells were infected with recombinant vaccinia virus with a multiplicity of infection of at least 10 for 1 hour in serum-free medium, followed by incubation in complete medium for 3 hours. Cells were harvested by centrifugation and resuspended at a density of 1 x 10 ⁷ cells / ml. 200 μl was added to each well of a microtiter plate with a U-shaped bottom containing 100 μl / well of complete medium. Cells were diluted in serial 2-fold dilutions. Two wells for each sample did not contain lymphocytes in order to obtain spontaneous release of chromium and measure the total absorption of chromium. Target cells, from the S3HeLa subline, were infected in 6.0 or 10.0 cm plates at a multiplicity of 10 infection for 1 hour in serum free medium and then incubated in complete medium for 3 hours. Then they were separated from the cells using PBS, 1 mm EDTA and counted. An aliquot of 10 ⁶ target cells (HeLa, Raji or RJ2.2.5 for experiments with the CD4 chimeric receptor and 3G8 10-2 cells (Shel et al., Mol. Immunol. 26: 959 (1989) for experiments with the CD16 chimeric receptor ) were centrifuged and resuspended in 50 μl of sterile ⁵¹ Cr-sodium chromate (1 mCi / ml, Dupont Wilmington, DE) for 1 hour at 37 ^° C. with periodic stirring, then washed three times with PBS. 100 μl of labeled cells resuspended in medium at 10 ⁵ cells / ml were added to each well. Raji and RJ2.2.5 target cells were labeled in the same way as HeLa cells. The microtiter plate was untwisted at 750 x g for 1 minute and incubated for 4 hours at 37 ^° C. At the end of the incubation period, the cells in each well were resuspended by careful pipetting, a sample was taken to determine the total incorporation, and the microtiter plate was unscrewed at 750 x g for 1 minute. and counted in a gamma ray scintillation counter. The kill percentage was adjusted for the fraction of infected target cells (usually 50-90%) measured by flow cytometry. For infected effector cells, the effector / target ratio was adjusted by the percentage of infected cells (usually 20-50% for experiments with the CD4 chimeric receptor and more than 70% for experiments with the CD16 chimeric receptor).

In vitro mutagenesis of the ζ (zeta) sequence
To create point mutations in amino acid residues 11 and / or 15 of the ζ sequence, synthetic oligonucleotide primers were obtained extending from the BamHI site
against the transcription from the transmembrane domain ζ and converting residue 11 of native ζ from Cys to Gly (CllG) or residue 15 from Asp to Gly (D15G) or both (CllG / D15G). They were used in PCR reactions to generate mutated fragments that were inserted into wild-type CD4: ζ constructs.

To create ζ deletions, ζ cDNA sequences were amplified by PCR using synthetic oligonucleotide primers designed to create a stop codon (UAG) after residues 50, 59 or 65. These primers contained a cleavage site for NotI enzyme in the form of protruding 5 or 6 residues from 5'-end, usually in the sequence of the form CGC GGG CGG CCG CTA (SEQ ID N 11), where the last three residues correspond to stop anticodon. The NotI and stop anticodon sequences were followed by 18 or more residues complementary to the target 3 ′ end of this fragment. The resulting chimeras were named CD16: ζ Y51 ^* , CD16: ζ E60 ^* and CD16: ζ D66 ^*, respectively. The BamHI site against the transcription from the transmembrane domain and the NotI site were used to generate fragments that were reintroduced into the wild-type CD16: ζ construct. Monomeric ζ-chimeras were created by the release of trans-membrane and membrane-proximal intracellular ζ sequences by cleavage using the BamHI and SacI constructs Asp ^- and Cys ^- CD4: ζ described above and incorporating this fragment into the constructs CD16: ζ E60 ^* and CD16: ζ D66 ^* respectively.

Three-part chimeric constructs CD16: 7: ζ (48-65) and CD16: 7: ζ (48-59)
To obtain the CD16 construct: ζ D66 ^* , the ζ cDNA sequence corresponding to the transmembrane domain and the 17 following residues of the cytoplasmic domain was replaced with the corresponding transmembrane and cytoplasmic domain obtained from CD5 and CD7 cDNA. Fragments of CD5 and CD7 were obtained by PCR using direct oligonucleotides, including the BamHI cleavage site and corresponding to the region opposite the transcription from the transmembrane domain of CD5 and CD7, respectively, and the following reverse oligonucleotides that overlap the CD5 and CD7 sequences, respectively, and the ζ sequence containing the site cleavage of SacI.

SD5: ζ: CGC GGG CTC CTT ATA GAG CTG GTT CTG GCG CTG CTT CTT CTG (SEQ ID N 12)
CD7: ζ: CGC GGG GAG CTC GTT ATA GAC CTG GTT TGC CGC CGA ATT CTT ATC CCG (SEQ ID N 13).

PCR products CD5 and CD7 were digested with BamHI and SacI and ligated with digested BamHI and SacI CD16: ζ E60 ^* and the ζ sequence from BamHI to SacI was replaced with a CD7 fragment. To obtain CD16: CD5 and CD16: CD7 constructs, CD5 and CD7 fragments were obtained by PCR using an oligonucleotide containing a NotI cleavage site and an encoding stop codon (UAA) after residue Gln416 and Ala193 CD5 and CD7, respectively. The PCR fragment of CD5 and CD7 was digested with BamHI and NotI and built into the CD16: ζ Asp66 ^* construct.

In vitro mutagenesis of N-terminal residues within the transducing cytolytic signal of the ζ motif (zeta)
Synthetically, oligonucleotide primers extending from the SacI site within the ζ motif and converting native residue 48 from Asn to Ser (N48S), residue 50 from Leu to Ser (L50S) and residue 51 from Tyr to Phe (Y51F) were synthesized and used in the reaction PCR to generate fragments that were reintroduced into the wild-type CD16: 7: ζ (48-65) construct.

In vitro mutagenesis of C-terminal residues within the transducing cytolytic signal of the ζ motif (zeta)
Synthetic oligonucleotide primers extending from the NotI 3 ′ site with respect to the stop codon and converting the native residue 60 from Glu to Gln (E60Q), residue 61 from Glu to Gin (E61Q), residue 62 from Tyr to Phe or Ser (Y62F or Y62S) and residue 63 from Asp to Asn (D63N) were synthesized and used in PCR to generate fragments that were subcloned into the wild-type CD16 construct: ζ D66 ^* from the BamHI site to the NotI site.

Chimeric constructs CD16: 7: ζ (33-65), CD16: 7: ζ (71-104), DC16: 7: ζ (104-137)
The CD7 transmembrane fragment carrying the MluI and NotI sites at the junction between the transmembrane and intracellular domains was obtained by PCR using the oligonucleotide of the following sequence: CGC GGG GCG GCC ACG CGT CCT CGC CAG CAC ACA (SEQ ID N 14). The resulting PCR fragment was digested with BamHI and NotI and re-integrated into the CD16: 7: ζ construct (48-65). Fragments ζ ,, encoding residues 33-65, 71-104 and 104-137, were obtained by PCR using a pair of primers containing MluI sites at the 5'-end of the direct primers and stop codons, followed by NotI sites at 5'- end of reverse primers. In each case, the restriction sites had 6 residues from the 5'-end of the primer to ensure restriction enzyme digestion.

ζ 33: CGC GGG ACG CGT TTC AGC CGT CCT CGC CAG CAC ACA (SEQ ID N 15)
ζ 71: CGC GGG ACG CGT GAC CCT GAG ATG GGG GGA AAG (SEQ ID N 16) and
ζ 104: CGC GGG ACG CGT ATT GGG ATG AAA GGC GAG CGC (SEQ ID N 17).

Construction of deletion mutants FcR γ 11A
The carboxyl-terminal deletion mutants FcR γ 11A were constructed by PCR in the same way as full-length constructs, converting the sequences encoding tyrosine at positions 282 and 298 into stop codons (TAA). N-terminal deletions were generated by amplification of fragments encoding a smaller part of the intracellular domain by PCR using oligonucleotides that allow the fragments to be inserted between the MluI and NotI restriction sites in a predesigned expression plasmid encoding the CD16 extracellular domain fused to the transmembrane domain the latter ends at the MluI site and the junction between the transmembrane and intracellular domains.

OTHER OPTIONS
The examples described above show that the aggregation of ζ, η or γ chimeras is sufficient to initiate a cytolytic effector cell response in T cells. The well-known expression range of ζ, η and γ, which includes T-lymphocytes, natural killer cells, basophilic granulocytes, macrophages and mast cells, suggests that conservative sequence motifs can interact with a sensory apparatus common to cells of hematopoietic origin, and that is important a component of the body's immune defenses in the immune system can be mediated by receptor aggregation events.

 The activity of the cytolytic response and the lack of response to target cells carrying MHC class II receptors demonstrate that ζ, η or γ-based chimeras form the basis for a genetic effect on AIDS through adoptive immunotherapy. The widespread occurrence of endogenous ζ and γ and the evidence that Fc receptors associated with γ mediate cytotoxicity in various types of cells (Fanger et al., Immunol. Today 10: 92-99 (1989)) allow us to consider many cells as suitable perhaps for this purpose. For example, neutrophilic granulocytes, which have a very short lifespan (approximately 4 hours) in the circulation and are intensely cytolytic cells, are attractive target cells for the expression of chimeras. Infection with HIV neutrophils does not appear to result in virus isolation, and the abundance of these cells (most preferably white blood cells) should facilitate the body's immune defense. Another tempting opportunity for host cells is mature T cells, a population currently available for retroviral genetic engineering (Rosenberg, S. A. Sci. Am. 262: 62-69 (1990)). Using recombinant IL-2, T-cell populations can be propagated in culture with relative ease, and multiplied populations have a limited lifespan during reinfusion (Rosenberg et al., N. Engl. J. Med. 323: 570-578 (1990) )

 Under suitable conditions, HIV recognition by cells expressing CD4 chimeras should also provide mitogenic stimuli, enabling the cell population with immunoglobulin receptors to respond dynamically to viral load. Although we have focused on the behavior of fusion proteins in cytolytic T lymphocytes, the expression of these chimeras in helper lymphocytes can provide an HIV-mobilized source of cytokines that could counteract the collapse of the helper cell subpopulation in AIDS. A recent description of several schemes for constructing resistance to infection at stages other than virus penetration (Friedman et al. Nature 335: 452-454 (1988); Green et al. Cell 58: 215-223 (1989); Malim et al. Cell 58: 205-214 (1989); Trono et al. Cell 59: 113-120 (1989); Buonocore et al. Nature 345: 625-628 (1990)) suggests that cells carrying CD4 chimeras could be designed in such a way as to reduce the formation of the virus by expression of appropriate agents having an intracellular site of action.

 The ability to transmit signals to T-lymphocytes through autonomous chimeras also provides the ability to regulate lymphocytes constructed using retroviruses in vivo. Crosslinked stimuli mediated, for example, with specific IgM antibodies designed to remove complement-binding domains, can allow such lymphocytes to increase in number in situ, while treatment with similar specific IgM antibodies (for example, recognizing the amino acid variation constructed in the chimeric chain) could selectively deplete the engineered population. In addition, anti-CD4 IgM antibodies do not require additional crosslinking to mobilize calcium in Jurkat cells expressing CD4: ζ chimeras. The ability to regulate cell populations without resorting to repeated extracorporeal amplifications can significantly expand the range and effectiveness of current applications proposed for genetically engineered T cells.

 Although this invention has been described in connection with its characteristic variants, it should be understood that it can be modified further, and this application is intended to cover the variations, applications or adaptations of this invention, and includes such deviations from this description that are included in the field to which this invention relates, and can be attributed to the main features set forth herein, as follows from the scope of the attached claims.

Claims (14)

 1. A method of directing a cellular immune response against an HIV-infected mammalian cell, comprising exposing it to an effective amount of cells expressing a membrane-bound protein chimeric receptor containing (a) an extracellular portion that includes a CD4 fragment consisting of amino acids 1-200 (SEQ ID N0: 31) or amino acids 1-394 (SEQ ID N0: 29), which is able to specifically recognize and bind said HIV-infected cell, but which does not mediate HIV infection, and (b) the intracellular part, which is capable of Vat signal said cell to destroy a receptor associated with HIV-infected cell, and which is part of the transmit signal T-cell receptor, B cell receptor, or Fc-receptor.  2. The method according to claim 1, characterized in that the CD4 fragment consists of amino acids 1 to 394 or amino acids 1 to 200.  3. The method according to claim 1, characterized in that the CD4 fragment is separated from the intracellular part by the CD7 transmembrane domain shown in FIG. 26, or the hinge region, by the CH2 and CH3 domains of the human IgGI molecule shown in FIG.  4. The method according to claim 1, characterized in that the CD4 fragment is separated from the cell membrane by at least 48 angstroms or at least 72 angstroms.  5. The method according to claim 1, characterized in that the T-cell receptor is ξ.  6. The method according to claim 1, characterized in that said receptor comprises a transmembrane portion of CD7, a transmembrane portion of CD5, or a transmembrane portion of CD34.  7. The method according to claim 1, characterized in that the CD4 fragment is separated from the cell membrane by one or more protein alpha helices.  8. Protein membrane-bound chimeric receptor containing (a) the extracellular part, representing a CD4 fragment consisting of amino acids 1 to 200 (SEQ ID N0: 31) or amino acids 1 to 394 (SEQ ID N0: 29), which is able to specifically recognize and bind with HIV or HIV-infected cells, but which does not mediate HIV infection, and (b) the intracellular part, which is capable of signaling to cells carrying the indicated receptor, causing destruction of the receptor-linked HIV or HIV-infected cells and which is the signal transducing part of T- crate Nogo receptor, B cell receptor, or Fc-receptor.  9. The chimeric receptor of claim 8, wherein the CD4 fragment consists of amino acids 1 to 394 or amino acids 1 to 200.  10. The chimeric receptor of claim 8, wherein the CD4 fragment is separated from the intracellular portion by the CD7 transmembrane domain shown in FIG. 26, or the hinge region, by the CH2 and CH3 domains of the human IgGI molecule shown in FIG. 25.  11. The chimeric receptor of claim 8, wherein the CD4 fragment is separated from the cell membrane by at least a distance of 48 angstroms or at least a distance of 72 angstroms. 12. The chimeric receptor of claim 8, wherein the T-cell receptor is ξ.
13. The chimeric receptor of claim 8, characterized in that the receptor includes the transmembrane part of CD7, the transmembrane part of CD5 or the transmembrane part of CD34.  14. The chimeric receptor of claim 8, wherein the CD4 fragment is separated from the specified cell membrane by one or more protein alpha helices.  15. DNA encoding a chimeric receptor of claim 8. Priority on points:
02/14/1994 according to claims 1-3, 5, 6, 8-10, 12, 13, 15;
08/02/1994 according to claims 4, 7, 11, 14.

Images