WO1991009123A1

WO1991009123A1 - Soluble cd4 having a decreased affinity for mhc class ii antigens

Info

Publication number: WO1991009123A1
Application number: PCT/US1989/005625
Authority: WO
Inventors: Ellis L. Reinherz; Linda K. Clayton
Original assignee: Dana Farber Cancer Institute
Priority date: 1989-12-15
Filing date: 1989-12-15
Publication date: 1991-06-27

Abstract

The present invention pertains to biologically active, modified soluble human CD4 surface glycoprotein fragments having a decreased affinity (compared to naturally occurring human CD4) for major histocompatibility class II antigens but which are capable of binding to the HIV gp120 envelope protein. The soluble CD4 fragments have the ability to inhibit HIV binding to naturally occurring CD4 surface receptors but will not inhibit class II MHC recognition events.

Description

Soluble CD4 having a decreased affinity for MHC class II antigens.

Background of the Invention

The CD4 (T4 molecule, which is a surfacs glycoprotein on a subset of T lymphocytes (referred to as T4 lymphocytes) is involved in class II (la) major histocompatability (MHC) recognition and appears to be the physiological receptor for one or more monomorphic regions of class II MHC. Meuer, S. et al., Proceedings of the National Academy of

Sciences, U.S.A., 79:4395-4399 (1982); Biddison, W. et al., J. Exp. Med., 156:1065-1076 (1982); Gay, D. et al., Nature, 328:626-629 (1987).

MHC class II restricted regulatory and effector function of this subset of mature T cells is dependent on CD4-class II interactions. Such interactions appear to be crucial during thymic ontogeny for development of CD4+CD8- T lymphocytes with helper and cytotoxic functions (Kruisbeek, A.M. et al., J. Exp. Med. 161:1029-1047 (1985)). It has been shown that monoclonal antibodies directed either at CD4 or MHC class II antigen can profoundly interfere with T cell receptor-mediated activation (Krensky, A.M. et al., Proc. Natl. Acad. Sci. USA 79:2365-2369 (1982); Meuer S.C. et al., Proc. Natl. Acad. Sci. USA 79 : 4395 -4399 (1982); Biddison W. et al., J. Exp. Med. 165:1065-1076 (1982); Marrack, P. et al., J. Exp. Med. 158:1077-1091 (1983).

Human CD4 is also the receptor for the gpl20 envelope glycoprotein of the human immunodeficiency virus (HIV) and is essential for virus entry into the host cell, and for membrane fusion, which both contribute to cell-to-cell transmission of the virus and to its cytopathic effects (Klatzmann, D. et al.,Science 225:59-63 (1984); Dalgleish, A.G. et al., Nature 312:763-766 (1984); Sattentau, Q. et al.,

Science 234:1120-1123 (1986); McDougal, J.S. et al., Science 231:382-385 (1986); Maddon, P.J. et al.,

Cell 47:333-348 (1986)). Residues involved in gp120 binding have been localized to a region within the immunoglobulin-like domain I of CD4 corresponding to CDR2 of an Ig variable region (Peterson, A. et al., Cell 54:65-72 (1988); Landau, N.R. et al., Nature 334:159-162 (1988); Clayton, L.K. et al., Nature

335:363-366 (1988)).

Considerable effort has been expended in studying the CD4-gp120 interaction and in trying to interfere with or inhibit that interaction, in an attempt to provide a means by which the life

threatening effects of HIV infection can be slowed or reversed. Several groups have focused their efforts on the ability of soluble CD4 (T4) protein to interfere with infection of cells by HIV and its subsequent effects. (Hussey, R.E. et al., Nature,

331:78-81 (1988) ; Fisher, R.A. et al., Nature,

331:76-78 (1988) ; Deen, K.C. et al ., Nature,

331:82-84 (1988) ; Traunecker, A. et al. , Nature, 331:84-86 (1988)) . A means by which to prevent HIV infection of T4 lymphocytes (i.e., helper and inducer T lymphocytes), which make up approximately 60-80% of the total circulating T lymphocyte population, would be of great value, particularly in light of the fact that HIV infection of such cells can cause total collapse of the immune system.

(Curran, J. et al., Science, 229:1352-1357 (1985); Weiss, R. et al., Nature, 324:572-575 (1986)). Summary of the Invention

The present invention pertains to biologically active, modified soluble human CD4 surface glycoprotein fragments having a decreased affinity, compared to naturally occurring CD4 molecules, for major his tocompatability class II antigens, but which are capable of binding to the human immunodeficiency virus (HIV) gp120 envelope protein. The soluble CD4 fragments correspond to a portion of the extracellular domain of human CD4 that is capable of binding HIV gp120. The fragments, however, are mo di f i e d in a manner ( e . g . , by s ub s t i tut i on , deletion, additions to the amino acid sequence) so as to alter the class II MHC binding affinity, compared to naturally occurring CD4 molecules.

The binding region for MHC class II antigens broadly spans the extracellular domain and includes residues involved in gp120 binding. Therefore, the portion of the CD4 molecule involved in MHC binding can be modified to decrease binding affinity for class II MHC antigens without substantially altering its affinity for gp120. These modified soluble CD4 fragments have the ability to inhibit HIV binding to the CD4 receptors of T lymphocytes and to deplete circulating levels of HIV gp120, but will not substantially bind class II MHC antigen bearing B lymphocytes. As a result, the CD4 fragments do not interfere with the ability of CD4 T lymphocyte surface receptors to bind class II MHC antigen bearing cells.

The soluble CD4 fragments of this invention can be used to treat a patient infected with HIV. In addition, the soluble CD4 fragments can be used to screen for substances which alter or block CD4-MHC class II interactions. They can also be used as a tool for studying binding affinity of molecules to CD4 surface receptors and/or class II MHC antigen bearing B cells.

Brief Description of the Figures

Figure 1 is the nucleotide sequence of T4 SEC1 cDNA (referred to as T4 . sequence), which encodes

370 amino acids of soluble CD4 protein (referred to as T4_exl). The deduced amino acid sequence of the

T4_exl protein is represented below the nucleotide sequence. (See U.S. Patent Application Serial No. 07/217,475, filed July 11, 1988, the teachings of which are incorporated herein by reference). Figure 2 is a schematic diagram of the CD4 protein and expanded portions thereof relating to substitution positions in 17 mutants.

Figure 3a show the binding of T51 B cells to COS-1 cells transfected with CD4, CD4 mutant Mll or CDM8 vector only.

Figure 3b shows an fluorescence-activated cell sorter (FACS) analysis of anti-CD4 monoclonal antibody OKT4 and gpl20 binding to COS-1 cells transfected with wild type CD4 , MlB and M3 mutants.

Figure 4a is a graphic representation of the inhibition of T51 B cell binding by monoclonal antibodies and gp120.

Figure 4b shows the binding of Epstein-Barr virus (EBV) - transformed MHC class II antigen loss mutant B cell lines to CD4 transfected COS-1 cells.

Detailed Description of the Invention

The invention pertains to soluble human CD4 surface glycoprotein fragments having a decreased affinity, compared to naturally occurring CD4 molecules, for MHC class II antigens, but which are capable of binding HIV gpl20. The binding region for MHC class II antigens broadly spans the extra- cellular domain (domain I and a portion of domain II) of the CD4 surface glycoprotein and includes residues involved in gp120 binding. The portion of the CD4 molecule involved in MHC binding can be modified to alter its binding affinity for class II MHC antigens without substantially altering its affinity for gp120. The affinity of the soluble CD4 fragment for gp120 can be substantially similar to natural CD4 or it can be altered (i.e., greater or lower affinity for gp120 than natural CD4), but not so altered as to abrogate gpl20 binding. Preferably, the soluble CD4 fragments will have the ability to inhibit HIV gp120 protein binding to CD4 surface receptors of T lymphocytes, but will leave the MHC class II antigens available to bind CD4, thus allowing CD4- bearing cells to interact with MHC class II

antigens.

The CD4 fragments of this invention are soluble in aqueous medium and, therefore, contain none of the hydrophobic transmembrane region of CD4. They, however, can comprise a portion (generally six amino acids or less) of the hydrophobic region which does not prevent solubilization of the CD4 protein.

Soluble CD4 having altered affinity for gp120 has been previously described by Reinherz and Clayton, U.S. Patent Application Serial No. 07/217,475, filed July 11, 1988, and U.S. Patent Application Serial No. 07/206,585, filed June 14, 1988. The teachings of each are incorporated by reference herein.

Reinherz and Clayton found that modifications in domain I and domain II of human CD4 protein alters the affinity of the CD4 molecule for gp120. However, the amino acid residues (or sites) of human CD4 protein involved in MHC class II interaction have not been identified prior to this invention. Such critical sites have now been identified by means of oligonucleotide-directed mutagenesis to create mutant human CD4 molecules which include 1-4 amino acid substitutions. The approach used in defining such sites has taken advantage of the differences known to exist in the amino acid

sequence of the extracellular segment of murine CD4 and that of its human counterpart.

The soluble CD4 fragments of this invention correspond to a portion of the extracellular domain (domain I and II) of human CD4 that is capable of binding HIV gp120. The fragments, however, are modified by altering the amino acid sequence of soluble human CD4 at a selected site or sites in such a manner that the resulting CD4 fragment is capable of binding gpl20 but has a decreased binding affinity for MHC class II antigens than that of the corresponding (unaltered) human CD4 fragment. Thus, the CD4 fragments will not interfere with the ability of CD4 T lymphocyte surface receptors to bind class II MHC antigen bearing cells.

As a result of the identification of sites critical for binding of CD4 to class II MHC antigen bearing cells, it is now possible to produce modified soluble human CD4 fragments whose ability to bind class II MHC is altered (i.e., whose ability to bind class II MHC is different from that of the corresponding naturally-occurring human CD4 fragment). Preferably, the fragment has a substantially decreased affinity for class II MHC. As described in the Examples, such sites have been identified by oligonucleotide-directed mutagenesis to create 17 individual mutant human CD4 molecules which resulted in substitution of non-conserved (polarity and/or charge change) murine amino acid residues for human CD4 residues between amino acid positions 17 and 167. Each human CD4 mutant contained from one to four murine amino acids. The amino acid sequence of human CD4 is shown in Figure 1. Preferably, the soluble fragments of the invention consist essentially of domain I and II of human CD4 and therefore will comprise a nucleotide or amino acid sequence which corresponds to those domains. For example, mutants can be made which have a nucleotide sequence of Figure 1, except that a nucleotide triplet encoding the amino acid present in the equivalent position of murine CD4 protein is substituted at at least one nucleotide triplet site in the DNA. The triplet substitution can be at the following amino acid positions:

a) the nucleotide triplet encoding the amino acid at positions 17 and 18;

b) the nucleotide triplet encoding the amino acid at positions 23, 24 and 25; c) the nucleotide triplet encoding the amino acid at positions 27, 30, 32 and 34;

d) the nucleotide triplet encoding the amino acid at positions 72 and 73;

e) the nucleotide triplet encoding the amino acid at positions 99, 104 and 107; and f) the nucleotide triplet encoding the amino acid at positions 132, 133 and 137.

CD4 mutants which have the above substitutions have a decreased affinity (compared to natural CD4) for MHC class II antigens, but which are capable of binding HIV gp120 protein. An analysis of the binding affinities of these mutants for gp120 and MHC class II proteins is summarized in Table 1 and described in detail in the Exemplification.

Other amino acid substitutions which result in CD4 mutants having a decreased affinity for MHC class II antigens but which can bind HIV gp120 are also included within the scope of the invention. Preferably, soluble CD4 fragments will not be altered in such a way as to substantially reduce its affinity for HIV gρ120. The specificity of these CD4 mutants for gp120 and MHC class II antigens can be determined by the degree of adhesion to gpl20 and class II MHC antigen bearing B cells.

The soluble human CD4 glycoproteins of this invention include analogous or homologous amino acid sequences which encode proteins having decreased affinity for MHC class II antigens. These peptides can include sequences in which functionally

equivalent amino acid residues are substituted for residues within the sequence resulting in a silent change. For example, one or more amino acid

residues within the sequence can be substituted by another amino acid of a similar polarity which acts as a functional equivalent, resulting in a silent alteration. Substitutes for an amino acid within the sequence may be selected from other members of the class to which the amino acid belongs. For example, the nonpolar (hydrophobic) amino acids include glycine, alanine, leucine, isoleucine, valine, proline, phenylalanine, tryptophan and methionine. The polar neutral amino acids include serine, threonine, cysteine, tyrosine, asparagine, and glutamine. The charged (basic) amino acids include arginine, lysine and histidine. The

negatively charged (acidic) amino acids include aspartic and glutamic acid.

In addition, the peptide structure can be modified by deletions, additions, inversion, insertions or substitutions of one or more amino acid residues in the sequence which do not substantially detract from the desired functional properties of the peptide. Naturally occurring allelic variations and modifications are included within the scope of the invention so long as the variation does not substantially reduce the ability of the peptide to bind gp120. Modified soluble CD4 fragments having decreased affinity for MHC class II antigens but capable of binding gp120 can be made using recombinant DNA techniques, such as excising it from a vector containing cDNA encoding such a fragment, or by synthesizing DNA encoding the desired fragment mechanically and/or chemically, using known

techniques. Preferably, the DNA encodes a soluble CD4 fragment which includes none of the hydrophobic transmembrane region of CD4 or a portion of that region small enough that it does not prevent solubilization of the fragment. The CD4 fragments should be long enough to effectively bind to HIV gpl20.

An alternative approach to producing modified soluble human CD4 fragments of this invention is to use peptide synthesis to make a peptide or polypeptide having the amino acid sequence of such a fragment. The peptides or modified equivalents thereof can be synthesized directly by standard solid or liquid phase chemistries for peptide synthesis. For example, the above amino acid sequence or modified equivalent thereof encoding the cytoplasmic domain of CD4 can be synthesized by the solid phase procedure of Merrifield.

Preferably, the soluble CD4 peptides will be produced by inserting DNA encoding the peptide

(i.e., CD4 DNA which encodes the desired amino acid sequence of the extracellular domain of CD4) into an appropriate vector/host system where it is expressed. The DNA sequence which encodes the extracellular domain of human CD4 is shown in Figure 1. This sequence has been described in U.S. Patent Application Serial No. 07/217,475, filed July 11, 1988, by Reinherz et al. This native DNA sequence of CD4 can be modified by deletion, insertion or substitution of nucleotides to yield peptides which exhibit substantially the same properties of the above-described soluble peptides. The DNA sequences can be chemically synthesized or they can be obtained from natural sources using recombinant DNA technology.

A variety of host-vector systems can be used to express the peptides of this invention. Primarily, the vector system must be compatible with the host cell used. Host-vector systems include, but are not limited to, the following: bacteria transformed with bacteriophase DNA, plasmid DNA or cosmid DNA; micro-organisms, such as yeast containing yeast vectors; mammalian cell systems infection with virus (e.g., vaccinia virus, adenovirus, etc.); insect cell systems infected with virus (e.g., baculo- virus).

Any of the standard recombinant methods for the insertion of DNA into an expression vector can be used. The recombinant DNA vector can be introduced into appropriate host cells (bacteria, virus, yeast, mammalian cells or the like) by transformation, transduction or transfection (depending upon the vector/host cell system) and cultured to express the peptides of this invention.

This invention also pertains to an expression vector comprising a DNA sequence encoding a soluble human CD4 fragment having decreased affinity for MHC class II antigens but which is capable of binding gp120, and cells transformed thereby.

Modified soluble human CD4 fragments of the present invention can be used for therapy, diagnosis and prevention of infection with HIV. The soluble CD4 fragments can also be used to prevent immunosuppression since inhibition of the CD4 class II interaction by gp120 may provide the basis for observed Immunosuppression in patients infected with HIV.

According to this invention, the human CD4 fragments have the ability to inhibit HIV binding to CD4-bearing T lymphocyte receptors and deplete circulating levels of gp120, but do not interfere with class II MHC recognition events. As a result, natural CD4 receptors are made available to bind with class II MHC antigens. The soluble CD4 fragments of this invention can be used therapeutically (in vivo) to treat individuals infected with HIV. Such fragments can be administered by an acceptable route (e.g., intravenously, intramuscularly, intraperi toneally, orally), alone or in combination with a physiologically acceptable vehicle (e.g., saline buffer) and optional additives and/or preservatives. The amount of soluble CD4 should be sufficient to substantially inhibit HIV binding to CD4 bearing T lymphocyte receptors and to deplete circulating levels of gp120. Preferably, the CD4 fragments will have a decreased affinity (compared to naturally occurring CD4) for MHC class II antigens. Alternatively, the soluble CD4 fragments can have an affinity for class II MHC antigens which is substantially similar to that of naturally occurring CD4.

A portion of the extracellular domain of CD4 involved in MHC class II binding can be exploited to provide screening assays for substances that block CD4 surface receptor binding with MHC class II antigens. In the screening method, a cell

expressing CD4 on its surface is contacted with a substance to be tested under conditions which would permit the substance to complex with the CD4 surface protein. The cell is then contacted with an MHC class II antigen under appropriate conditions for binding to the surface receptor. The degree of binding of the substance of interest and MHC class II antigen is indicative of blocking activity of the substance. The invention will be further illustrated by the following Exemplification.

EXEMPLIFICATION Expression of CD4 and Cellular Adhesion with MHC Class II-Expressing B Cells in Transfected COS-1 Cells.

Plasmid pSP65-T4 (gift of Dan Littman, UCSF) was digested with BamHI and Xhol to release the CD4 insert. The insert was blunted with the Klenow fragment of DNA polymerase I, ligated to Xbci linkers (New England Biolabs) and subcloned into the Xba site of the vector, CDM8 (Seed, B. and Aruffo, A., Proc. Natl. Acad. Sci. USA 84:3365-3369 (1987).

COS-1 cells were transfected with CDM8 constructs as described by Clayton, L.K. et al . (Nature 335:363- 366 (1988)), but with the following modifications: 0.3 x 10⁶ cells are plated into each well of Falcon 6-well dishes and 6μg plasmid DNA (CsCl-banded twice) in 0.35 ml RPMI plus 0.35 ml RPMI-800 μg/ml DEAE dextran was used for transfection.

Binding of B cells to transfected COS-1 cells was assayed two days after transfection, as described by Doyle, C. and J.L. Strominger (Nature 330:256-259 (1987)), but with the following modifications: transfected COS-1 cells were washed once with RPMI and 1-2 x 10⁷ B cells were added to each

35 mm well in 0.8 ml RPMI, 2% fetal calf serum

(FCS), 1% glutamine, 1% penicillin-strep tomycin, 10 μg/ml gentamicin (final medium). The mixed cells were incubated for one hour at 37°C, the B cells aspirated and the well was washed 3-5 times by dropping 2 ml final medium into the well.

Binding was scored as +, +/- or - after viewing each well by phase contrast microscopy at a magnification of 100x. A plus value represented binding equivalent to that obtained with COS-1 cells transfected with wild-type CD4 DNA. A minus value was indistinguishable from transfection results obtained with CDM8 alone. A +/- value represented substantially reduced but still detectable binding (10-20% wild-type binding). A negative control of CDM8 - transfected COS-1 cells and a positive control of wild-type CD4 transfected COS-1 cells was ineluded in every assay.

The surface expression and structural integrity of each mutant CD4 was confirmed by FACS analysis using the monoclonal antibody OKT4. COS-1 cells transfected with the mutant indicated were scraped from the well, stained with monoclonal antibodies OKT4, 19Thy5D7 or MT321 (gift from Peter Rieber, Munich, West Germany) (1:500 and 1:100 dilution of ascites for OKT4 and 19Thy5D7, respectively, and purified MT321 at 20 μg/ml) and analyzed after staining with a 1:40 dilution of FITC - conjugated goat anti-mouse immunoglobulin (Meloy), on an Epics V cell sorter (Coulter). Propidium iodide was included to gate out dead cells. The anti-CD2 monoclonal antibody 3T4-8B5 was used as a negative control. Each mutant was also tested for gp120 binding. Cells were scraped from the well, incubated with 1 μg gp120 in 0.1 ml for 30 min on ice, stained with 200 ng of monoclonal anti-gpl20 antibody (Dupont, Wilmington, Delaware) and FITC-conjugated second antibody and analyzed by FACS as above. As a negative control for anti-gpl20 antibody staining, no gp120 was added.

Specificity of Binding of CD4 Transfected COS-1 Cells to MHC Class II Expressing B Cells.

The number of B cells bound were determined by labeling 10⁸ B cells for 2 h at 37°C in 0.5 to 1.0 mCi of ⁵¹Cr. The cells were washed four times and then used for binding as described In the previous section. An aliquot of labeled cells was used to determine the specific activity which ranged from 0.01 to 0.1 c.p.m/B cell. After washing the unbound B cells from individual wells, 1 ml PBS containing

1% Triton was added to each well. Cells were incubated 15 min at 37°C, lysates spun to remove cell debris and 100 μl supernatant counted to determine the number of B cells bound to transfected

COS-1 cells. Each B cell line was assayed for binding in duplicate wells of CD4- transfected COS-1 cells in two independent experiments. Counts bound to CDM8-transfected wells were subtracted from counts bound to CD4-transfected wells. For T51, c.p.m bound to CD4-transfected COS-1 cells were 2-3 times c.p.m bound to CDM8 - transfected COS-1 cells. For antibody inhibition studies, monoclonal antibodies were added at a 1:100 dilution to B cells and transfected COS-1 cells, the cells were incubated 30-60 min at 37°C, then mixed and assayed as described in the previous section. Monoclonal antibodies used were: anti-CD4, OKT4 ; anti-CD8, 7Pt3F9; anti-class I, W6/32; and anti-class II, 9/49. Recombinant gpl20 from the H3DCG isolate of HIV-I (Genentech) and ovalbumin (Sigma, St. Louis, MO) were added to a final concentration of 20 μg/ml to the B cells and transfected COS-1 cells. The cells were incubated for 30-60 min at 37°C, and then mixed and assayed as described above. MHC-surface expression of the class II antigen-loss mutant

EBV- transformed B cell lines was confirmed by FACS using the following monoclonal antibodies: 9/49 which binds to a common determinant on DR, DP and DQ

(Pesando, J.M. and L. Graf, J. Immunol. 136:4311-

4318 (1986), L243 which binds DR (Lampson, L.A. and R. Levy, J. Immunol. 125:293-299 (1980), and B7/21 (gift of I. Trowbridge, Salk Institute) which binds DP. Mutant B cell haplotypes: T51 (parental, DR1, 3; DQ1,2; DPX.4) (Levine, F. et al., Proc. Natl.

Acad. Sci. USA 82:3741-3745 (1985); 8.1.6 (DR-,-; DQ-,-; DP-,-) (Levine, F. et al., J. Immunol.

134:637-640 (1985); 9.28.6 (DR1,-; DQ1 , - ; DPX,-) (Levine, F. et al., Proc. Natl. Acad. Sci. USA

82:3741-3745 (1985); 8.1.6 (DR3,-; DQ2,-; DP4,X) (Levine, F. et al., Proc. Natl. Acad. Sci. USA

82:3741-3745 (1985); 4.36.4 (DR-,-; DQ1,-; DPX,-) (Pious, D. et al. , J. Exp. Med. 162:1193-1207

(1985) ; 11.11.4 (DR- , - ; DQ1 , - ; DP-,-) (Pious, D. et al. , J. Exp. Med. 162:1193-1207 (1985) .

Results

Given the known difference that exists in the binding of human class II-express ing B cells to human versus murine CD4- transfected COS-1 cells, oligonucleotide-directed mutagenesis was used to create 17 individual CD4 mutants incorporating non-conservative (polarity and/or charge change) murine for human substitutions between amino-acids 17 and 167. Each human CD4 mutant contained from one to four murine amino acid substitutions. The positions of the mutations are shown in Figure 2 with specific residue changes listed in Table 1.

Each mutant was transfected into COS-1 cells, assayed for surface expression of CD4 and for gp120 binding by FACS analysis, and tested for adhesion of T51 B cells. Results given in Table 1 are described in more detail below.

Figure 2 shows a schematic diagram of the CD4 protein showing the four immunoglobulin-like

domains, three disulfide bonds, transmembrane region and cytoplasmic tail. Numbering of amino acids was according to Hussey, R.E. et al., Nature 331:78-81 (1988). The expanded portion of the diagram shows the positions of substitutions in 17 different mutants. Mutations were created using the

thionucleotide method of oligonucleotide site- directed mutagenesis as previously described

(Clayton, L.K. et al., Nature 335:363-366 (1988), except that the template for mutagenesis was the full-length CD4 cDNA in M13 phage. The structure of all mutants was confirmed by DNA sequence analysis of the CDM8-CD4 mutant construct using the double- stranded DNA as a template.

Wild-type or mutant CD4 molecules were expressed by transfection into COS-1 cells and binding to MHC class II was measured by examining adhesion of class II MHC - express ing B cells to the COS-1 cells. The studies were conducted to determine the structural basis of CD4-class II MHC interaction which may serve to facilitate cell-cell contact ((Krensky, A.M. et al., Proc. Natl. Acad. Sci. USA 79:2365-2369 (1982); Meuer S.C. et al., Proc. Natl. Acad. Sci. USA 79:4395-4399 (1982); Biddison W. et al., J. Exp. Med. 156:1065-1076 (1982); Marrack, P. et al., J. Exp. Med. 158:1077-1091 (1983); Doyle et al., Nature 330:256-259 (1987), and influence T cell signal transduction (Eichmann, K et al., Eur. J.

Immunol. 17:643-650 (1987); Rudd et al., Proc. Natl. Acad. Sci. USA 85:5190-5194 (1988); Veillette, A. et al., Cell 55:301-308 (1988).

Figure 3a shows the binding of T51 B cells to COS-1 cells transfected with CD4 (left panel), CD4 mutant Mil (middle panel) or CDM8 vector only (right panel). As shown in Figure 3a (left panel), the EBV- transformed B - lymphoblastoid cell line, T51 (DR1, 3; DQ1.2; DPX, 4) (Mellins et al., Human

Immunol. 18:211-223 (1987), binds readily to CD4- trans fec ted CO S - 1 ce lls , but there was no T 51 B - cell binding to COS-1 cells transfected with the CDM8 vector alone (Fig. 3a, right panel).

The surface expression and structural integrity of each mutant CD4 was confirmed by FACS analysis using the monoclonal antibody OKT4. Each mutant was also tested for gp120 binding. Examples are shown in Figure 3b, and results for all mutants are presented in Table 1. Wild-type CD4- and mutant

MlB-transfected COS-1 cells bind equivalent amounts of OKT4 and gp120 (Figure 3b); M3 , although reactive with OKT4, failed to bind gp120 as previously reported (Clayton, L.K. et al., Nature 335:363-366 (1988). Although M3 reactivity with monoclonal antibodies OKT4 and 19Thy5D7 was slightly lower than that of CD4 , identical binding reactivity was observed with the anti-CD4 monoclonal antibody

MT321, which binds an epitope in domain IV

(Sattentau, Q. et al., Science 234:1120-1123 (1986) (data not shown). These results show that equivalent levels of surface CD4 were obtained when COS-1 cells were transfected with M3 and CD4 complementary cDNAs, indicating that the M3 mutation may directly or indirectly perturb the epitopes seen by OKT4 and 19Thy5D7.

The analysis of CD4 mutants for expression, gp120 binding and MHC Class II binding are summarized in Table I. The monoclonal antibody reactivity and gp120 binding were determined by FACS analysis, as previously described. For those tests, the symbols: + indicate antibody reactivity equivalent to that obtained with wild-type CD4-transfected COS-1 cells; 4- indicate levels of expression approximately 2 logs lower; - indicate no detectable expression. Results are shown for monoclonal antibody OKT4 , except in the case of M3 and M8 , for which monoclonal antibody MT321 was used. Class II MHC binding was quantitated, as previously

described; + indicates binding of T51 B cells similar to that depicted in Figure 3a (left panel) and > 90% of that obtained with wild-type CD4 ; +/- indicates binding of T51 B cells similar to that depicted in Figure 3a (middle panel) and 10-20% of that obtained with wild-type CD4; - indicates no detectable binding as depicted in Figure 3a (right panel) and <10% of that obtained with wild-type CD4. Note that no mutants gave class II MHC binding between 20% and 90% of the wild-type CD4.

To further examine the specificity of the assay, the effects of monoclonal antibodies directed against the individual components of the system were investigated. Figure 4b shows the binding of

EBV- transformed MHC class II antigen loss mutant B cell lines to CD4 transfected COS-1 cells as a function of percent T51 bound versus B cell line. As shown in Figure 4a, binding of the T51 B cells was markedly inhibited by monoclonal antibodies to CD4 (OKT4) and class II MHC (9/49), but was unaffected by monoclonal antibodies directed against CD8 (7Pt3F9) or class I MHC (W6/32). Identical results were obtained for the genotypically unrelated B cell line JY (data not shown). In

addition, the MHC class II antigen loss B-cell mutant line, 6.1.6, which fails to express any class II MHC alleles from either haplotype does not bind to CD4-transfected COS-1 cells (Figure 4b). The specific binding was 8,842 c.p.m. (Figure 4b). In contrast, the mutant cell lines which express at least one complete haplotype, 9.28.6 (DR1,-; DQ1,-;- DPX,-) and 8.1.6 (DR3,-; DQ2 ; DP4 ,X) bind, but at reduced levels. This result shows that different polymorphic alleles of class II MHC can bind to CD4. It was noted that the loss variants 4.36.4 and

11.11.4 which express only DQ1,DPX and DQ1, respectively, did not bind to CD4- express ing COS-1 cells. Whether the lack of binding of DR-negative B-cell variants was due to the low level of expression of DP and/or DQ in these cells as compared to DR or indicative of a lower affinity of CD4 for DP or DQ relative to DR cannot be determined at present. It has been shown, however, that these DP and DQ alleles are functional as restr ic tional elements in T-cell responses (Mellins, E. et al., Human Immunol. 18:211-223 (1987). Furthermore, in the murine system, liposomes containing either murine I-A or

I-E molecules bind CD4 - express ing cells (Gay et al., Proc. Natl. Acad. Sci. USA 85:5629-5633 (1988).

The murine CD4 structure is 50% identical at the amino acid level to its human homologue (Maddon et al., Proc. Natl. Acad. Sci. USA 84:9155-9159

(1987) and binds poorly to T51 B cells, the extent being similar to the binding shown in Figure 3a (middle panel) for mutant Mll.

Seven mutants dramatically reduced or eliminated class II MHC binding. These include three mutations in the CDR1 homologous region of domain I (Clayton e t al . , Nature 335 : 363 - 366 ( 1988 ) ) , mutant s M1 . 1 , M1.2 and M1B; one mutation in the CDR2 equivalent region (Clayton et al., Nature 335:363-366 (1988)), mutant M3; one mutation unrelated to any

CDR, M5; one mutation immediately distal to the CDR3 homologue, M8; and one mutation in domain II, Mll. An example of a mutation that reduced B-cell binding is shown in Figure 3a (middle panel) for the Mll mutant. M1.1, M1.2 and M5 also gave a reduced B- cell binding phenotype. The mutants M1B, M3 and M8 were indistinguishable from vector-only transfectants (Figure 3a, right panel). The other mutations were without effect.

In contrast to the large number of mutations which affect class II MHC binding, only M3 abrogated gp120 binding (Figure 3b and Table 1). Note that mutants M6 , M9 and M14 could not be evaluated as their alterations grossly affect the structure of the external CD4 domains, so that reactivity with all anti-CD4 monoclonal antibodies tested was reduced or, in the case or M14, eliminated. All the other 14 mutants, when transfected into COS-1 cells, yielded a copy number of variant CD4 molecules equivalent to the copy number of wild-type CD4, as assessed by quantative FACS analysis with anti-CD4 monoclonal antibodies OKT4, 19Thy5D7 and/or MT321.

As M3 abrogated binding of both class II MHC and gp120, the effect of gp120 itself on class II MHC interaction with CD4 was examined. Preincubation of B cells with a concentration of gp120 that saturated CD4 binding sites (20 μg/ml)

inhibited CD4-class II MHC binding (Figure 4a).

This inhibition of B cell binding to CD4-expressing COS-1 cells is not a result of down-modulation of CD4 on the transfected cells as their reactivity with OKT4 is unchanged (data not shown). Unlike gp120, soluble CD4 (T4_exl) (Hussey, R.E. et al.,

Nature 331:78-81 (1988)), even at a concentration of 1 mg/ml, has no effect on binding of B cells, leaving unaffected conjugates such as those in

Figure 3a (left panel). This is consistent with the affinity of the monomeric CD4-MHC interaction being much lower (by about ≥4 orders of magnitutde) than that of CD4 for gp120 (Lasky, L. et al., Cell

50:975-985 (1987), as well as with the lack of inhibition by soluble CD4 class II-restricted T-cell responses (Hussey, R.E. et al., Nature 331:78-81 (1988)). Given the low affinity of monomeric CD4 for class II MCH , the greatly up-regulated expression of CD4 copy number on activated T- lymphocyte clones (Meuer, S.C. et al., Nature

303:808-810 (1983)) probably enhances interactions of CD4+ cells with la-expressing cells through an increase in multipoint attachment.

Microbiological Deposit

Plasmid DNA from mutant MlB has been deposited on December 15, 1989, with the American Type Culture Collection, Bethesda, Maryland, and has been

assigned Accession Number _---------.

Equivalents

Those skilled in the art will recognize, or be able to ascertain using no more than routine

experimentation, many equivalents to the specific embodiments of the invention described herein. Such equivalents are intended to be encompassed by the following claims.

Claims

1. A soluble human CD4 glycoprotein having a

decreased affinity, compared to natural CD4, for MHC class II antigens, but which is capable of binding HIV gp120 protein.

2. The soluble human CD4 glycoprotein of Claim 1, consisting essentially of domain I and domain II of the human CD4 protein.

3. The soluble human CD4 glycoprotein of Claim 1, encoded by the nucleotide sequence of Figure 1, except that a nucleotide triplet encoding the amino acid present in the equivalent position of murine CD4 protein has been substituted at at least one nucleotide triplet site in the DNA, the triplet selected from the group consisting of:

a) the nucleotide triplet encoding the amino acid at positions 17 and 18;

d) the nucleotide triplet encoding the amino acid at positions 72 and 73;

4. An isolated DNA sequence encoding a soluble human CD4 glycoprotein having a decreased affinity, compared to natural CD4 , for MHC class II antigens, but which is capable of binding HIV gp120 protein.

5. An isolated DNA sequence of Claim 4 having the nucleotide sequence of Figure 1, except that a nucleotide triplet encoding the amino acid present in the equivalent position of murine CD4 protein has been substituted at at least one nucleotide triplet site in the DNA, the triplet selected from the group consisting of: a) the nucleotide triplet encoding the amino acid at positions 17 and 18;

d) the nucleotide triplet encoding the amino acid at positions 72 and 73;

6. A DNA expression vector containing a DNA

sequence encoding a soluble human CD4 glycoprotein having a decreased affinity, compared to natural CD4 , for MHC class II antigens, but which is capable of binding HIV gp120 protein.

7. The DNA expression vector of Claim 6, having the nucleotide sequence of Figure 1, except that a nucleotide triplet encoding the amino acid present in the equivalent position of murine CD4 protein has been substituted at at least one nucleotide triplet site in the DNA, the triplet selected from the group consisting of:

a) the nucleotide triplet encoding the amino acid at positions 17 and 18;

d) the nucleotide triplet encoding the amino acid at positions 72 and 73;

8. A cell transformed with the expression vector of Claim 7.

9. A method of treating a human infected with HIV, comprising administering to a patient, a soluble CD4 protein in an amount sufficient to deplete circulating levels of HIV gp120

protein.

10. The method of Claim 9, wherein the soluble CD4 protein has a decreased affinity, compared to natural CD4 , for MHC class II antigens, but which is capable of binding HIV gp120 protein.

11. The method of Claim 9, wherein the soluble CD4 protein is adminis trered in a physiologically acceptable vehicle.